AI and automation

86 articles Vitor By Vitor

Automatic AI Categorization: Optimizing Lead Management

youtube This article explores a new feature that automatically categorizes conversations with artificial intelligence, designed to improve lead management and strategic focus in customer communication environments. How AI Categorization Works The feature uses artificial intelligence to categorize every conversation in the inbox and give each lead a score. The process is simple: 1. Click the "Score open conversations" option in the left panel. 2. Choose the inbox you want. 3. Start the scoring ("Start scoring"). The AI processes the conversations and assigns a score to each contact. The system indicates that each scored conversation uses one AI credit. Benefits of AI Scoring AI scoring offers significant advantages: - Financial Visibility: helps you see where your investment is and where the potential return lies. - Strategic Focus: lets you direct energy and resources to the most promising leads. - Detailed Score: provides an individual score for each contact, showing how relevant or "hot" it is. Contact Details and Traffic Source For more detailed information about a specific contact, just: 1. Click the contact you want. 2. In the right panel, you can rescore the contact after an interaction ("AI rescore"). 3. Set the traffic source, such as a specific campaign, a video, etc. This helps you understand where your leads come from and how different channels perform. Lead Ranking The platform also lets you analyze all contacts at once, generating a lead ranking ("Lead ranking"). This gives you a quick view of the "hot contacts" across your entire customer base, improving your sales and customer service strategies. Conclusion The new automatic AI categorization feature is a powerful tool for improving lead management. It provides valuable insights into the value of each contact, improves how resources are allocated and offers a complete view of your customer base. Takeaways - The AI automatically categorizes and scores every conversation in the inbox. - A star score is assigned to each lead, indicating its potential. - You can rescore contacts and set the traffic source for deeper analysis. - The tool lets you build a lead ranking, making it easier to identify promising contacts. - The feature aims to save time and resources by focusing on the leads with the highest conversion potential. Start free on Papo Global → First user $0, no credit card. Learn more at papo.global.

AI-Powered Business Automation Platform

youtube This is a comprehensive introduction to an AI-powered business automation platform focused on optimizing customer communication and management. The system offers an intuitive interface and a wide range of features to centralize and automate operations, from account creation to sales and project management, with a strong AI component. Account Creation and Setup Creating an account is simple and flexible: you can sign up with Google or enter your details manually, including name, email, phone and password. After the account is created, you need to verify your email and provide company details such as your role, website, language, time zone, industry and company size. Channel Integration The platform stands out for its ability to integrate multiple communication channels, such as: - WhatsApp (official or unofficial API) - Facebook - Instagram - Email - Voice (via Twilio, for example) Note: the unofficial WhatsApp API (QR Code) is not provided by Meta, violates WhatsApp's terms, and the number may be blocked; for your main operation, use the official API. Main Menu Overview The main menu offers two view modes: - Simple Mode: Only the most frequently used settings and features. - Advanced Mode: Full access to all of the platform's tools and options. The left sidebar displays all conversations and features, including Super Chat and Radar, which are detailed later on. Calls You can set up a voice channel to make and receive calls, with recording included, centralizing your communication. Contacts The contacts section lets you manage leads and customers, with real-time automatic AI categorization based on past interactions and historical data. You can add tags (e.g. follow-up, Instagram lead, urgent, VIP) and view generated lead summaries. Extractor A powerful lead generation tool that uses the Brazilian Federal Revenue Service (Receita Federal) database to extract contacts of companies registered in the country. Leads can be filtered by niche, email, mobile phone, CNAE (business activity code), city and opening date. The platform also lets you: - Create custom audiences from extracted contacts. - Set up automatic message delivery via WhatsApp or email. - Create email sequences for lead nurturing. - Set up automatic lead extraction with defined intervals and quantities, plus automatic message sending. - Define advanced extraction settings, such as ignoring contacts already in your database, varying search pages and filtering by date. The cost per captured lead is 1 cent. Synapse This feature provides a visual view of the path taken by each lead, showing the interaction history, status (new lead, in contact, negotiation, sales, closed) and the relationships between customers in the same network. Kanban A complete sales management system, with customizable columns to organize leads into different stages (New Lead, In Contact, Negotiation, Closed). You can create automations and webhooks for each stage, as well as manage multiple pipelines. Projects The projects section is a more robust Kanban, ideal for organizing work into subprojects and tasks. Each task can include: - Action checklists. - Documents editable in real time. - Attachments (videos, photos, audio). - Links to customer conversations. - Specific filters. - Team comments. - Change history. - Priority, status, category, deadlines, assignees and subtasks. The view can be a list, boards or cards. Appointments A complete internal calendar to manage appointments, with: - Bill due dates. - Customer and professional database. - Service definitions (name, description, duration, price). - Partner configuration. - Reports and webhooks for appointments and custom events. Flows (Automations) The platform offers a system of preconfigured automations and the ability to create custom flows using Super Chat. The flow library includes ready-made templates for various market niches (e.g. law firms, clinics, real estate agencies, restaurants). Prompt and AI This is the most crucial part of the platform: setting up the AI assistant. You can create prompts (instructions for the AI) in a simplified way, through a guided interview or manual writing, or by using ready-made templates for different businesses. The prompt lets you define: - The assistant's behavior. - Main objective. - Absolute rules. - Tone of voice. - Company information. - Calls to action. - Additional notes. - General rules. You can test the prompt in real time and create multiple assistants (personas) to handle specific inboxes. Timing Settings to humanize the AI by adjusting: - Time before the AI resumes replying (pause after a human interaction). - Wait before the first reply. - Message buffer wait time (pause between sequential messages). - Maximum messages per reply. - Past interactions (how many previous messages the AI will read as context). - How long messages are kept in memory. Knowledge The AI can use reference documents (PDFs, texts) as a knowledge base for products and services, and can also suggest Help Center articles. Understand The AI can read and extract knowledge from files (PDFs, images, texts) sent by customers, using them as context for the conversation. Send Lets you configure the AI to send specific files (catalogs, price lists) when the customer asks for them. Audio The AI can reply with audio, with trigger options (only when the customer sends audio, always, or alternating) and a variety of voices from different companies. Inboxes Links each AI assistant (prompt) to an inbox (WhatsApp number, Instagram, etc.), allowing different AIs to handle different channels. Stats Basic reports on how many messages were sent, failed and skipped. Calendar Connection with Google Calendar for automatic scheduling, with AI rules for times and durations. Includes a follow-up tool that sends reminders before, during and after the meeting. Dispatcher (Email Marketing) Lets you create message broadcasts to contact lists, with options to import via CSV, sanitize users, and use tags or Kanban stages. The platform offers a range of metrics and the option to call the contact if they click the email. Companies Management of registered companies, working like a contact but oriented to the company's tax ID (CNPJ). Reports An overview of the platform with open conversations, agent status, conversation traffic and resolutions in real time. Includes team auditing, inbox reports, access levels (SLA) and bot reports. Campaigns (Live Chat) Creation of chatbots with multiple assistants to manage conversations in real time. Help Center Creation of public or private articles for customers and employees, with categorization and search options. Affiliates An affiliate program that lets you generate referral links to earn commissions on new customers' subscriptions. Settings Various account settings, such as name, language, currency, email domain and account ID, plus options to add agents, teams, agent assignment rules, inboxes, MCP servers, API, WhatsApp templates, labels, custom attributes, traffic source, automations, bots and macros. Super Chat (AI Assistant) Super Chat is the central AI tool: an assistant that lets you control 90% of the platform through text or voice commands. It checks tasks, shows priorities, sends proposals, qualifies leads, schedules meetings, captures contacts and much more. It is like an interactive, intelligent CRM. Conclusion The platform presented is a robust, highly integrated tool designed to optimize business management through automation and artificial intelligence. Its main differentiators include: - Centralized Communication: All channels (WhatsApp, Facebook, Instagram, email, voice) are managed in one place. - Advanced AI: From lead categorization to creating automation flows and scheduling, AI optimizes and humanizes interactions. - Comprehensive Management: Features such as the lead extractor, sales and project Kanban, appointments and detailed reports cover all of a business's needs. - Flexibility and Customization: Ready-made templates, customizable automations and adjustable AI settings for different niches and needs. - Super Chat: An AI assistant that lets you control much of the platform by voice or text, simplifying tasks and access to information. The platform aims to be an "AI-first" solution, prioritizing artificial intelligence to solve problems and optimize business processes. Start free on Papo Global → First user $0, no credit card. Learn more at papo.global.

Super Chat: Your CRM Controlled by Text Prompts

youtube This article presents "Super Chat", an innovative tool that lets you control your entire CRM system through simple text commands, known as "prompts". It shows how to automate marketing and management tasks, such as sending emails, and also how to organize your daily priorities. What is Super Chat? Super Chat is an interface that lets you interact with your CRM through text prompts. Instead of navigating complex menus, you simply type what you want the tool to do. Automated Email Sending The first demonstration shows how to extract contacts and send emails automatically: 1. Initial command: The user types "Extract 5 contacts with an email and send them a message". 2. Choosing the channel and mode: The tool asks which channel to use (WhatsApp, email or email marketing) and which sending mode (send now or create a campaign). The user chooses "email, send now". 3. Selecting the inbox and subject: Super Chat lists the available inboxes and asks for the email subject and body. The user chooses "Gmail Sales" as the inbox, sets the subject to "Don't miss this opportunity" and the body to a "message about sales". The tool then writes the email body automatically. 4. Confirmation and sending: After the user confirms, the tool runs the operation, extracting the contacts and sending the emails. The demonstration shows that the operation completed successfully, listing the 5 emails sent in less than a minute, with the defined subject and body, and confirming delivery in the Gmail inbox. Organizing Daily Priorities Besides sending emails, Super Chat can also help you organize your priorities: - Setting priorities: The user can set their priorities based on projects and tasks. - Daily summary: The tool provides a summary of tasks, categorizing them as "Top urgent", "Urgent" and "Medium", and showing their status (in progress, pending, overdue). - Full CRM control: Super Chat lets you control the entire CRM with a single prompt, from extracting emails to daily planning. Takeaways - Task automation: Super Chat simplifies and automates routine CRM tasks, such as sending emails, using text prompts. - Intuitive control: The chat interface makes CRM management more accessible and easier to use. - Efficient organization: The tool helps organize and prioritize daily tasks, improving productivity. - Flexibility: It allows you to customize email campaigns and adapt to different inboxes and topics. Start free on Papo Global → First user $0, no credit card. Learn more at papo.global.

Feeding AI: How to Create a Knowledge Base for Automatic Responses

With Papo (https://papo.global/), you can create a centralized Knowledge Base by adding your company's documents and information to train artificial intelligence. This process, known as RAG (Retrieval-Augmented Generation), allows the AI to consult your exclusive material to provide automatic, accurate, and consistent responses to your customers' questions, directly on service channels like WhatsApp. A well-structured Knowledge Base is the brain behind an efficient AI assistant. It ensures that the bot provides not generic answers, but information aligned with your company's policies, products, and processes. This increases customer trust and optimizes your team's time. | Benefit | Description | | :--- | :--- | | Accurate Responses | The AI uses your own documents as a source, avoiding generic or incorrect information. | | 24/7 Service | Automate responses to frequently asked questions at any time of day, without overwhelming your team. | | Communication Consistency | Ensures all customers receive the same official information, regardless of who or what responds. | | Reduced Operational Load | Frees human agents from repetitive questions so they can focus on complex cases and sales opportunities. | | Scalability | Allows you to serve a larger volume of customers simultaneously without needing to expand your team proportionally. | What is an AI Knowledge Base (RAG)? An AI Knowledge Base is a central repository of information you provide to train a virtual assistant. Think of it as your company's exclusive digital library. When a customer asks a question, instead of searching for answers on the internet or a limited script, the AI consults this library first. The technology behind this is called RAG (Retrieval-Augmented Generation). The process works in two steps: 1. Retrieval: The AI searches for and locates the most relevant text snippets within your documents that match the customer's question. 2. Generation: Based on these retrieved snippets, the AI formulates a cohesive and contextualized response in natural language. This differentiates Papo's AI from traditional chatbots, which rely on pre-programmed and rigid flows. With RAG, your assistant becomes dynamic and capable of accurately answering a much wider range of questions. How the Knowledge Base Works in Papo The Papo platform offers an intuitive interface for you to manage your Knowledge Bases without needing technical expertise. The general process follows these steps: 1. Base Creation: Within the platform, you can create one or more Knowledge Bases. You can organize them by theme, such as "Products and Services," "Exchange Policies," or "Technical Manuals," to facilitate management. 2. Document Upload: You feed each base with your documents. These can be files containing FAQs, product specifications, terms of service, user guides, among others. The platform processes these files and makes them searchable for the AI. 3. AI Connection: The Knowledge Base is associated with your AI assistant, such as Super Chat. From then on, any interaction requiring an informative response will first consult this base. 4. Search and Response: When a customer sends a message, Papo's AI analyzes the question, searches for relevant information in your base, and formulates the response in seconds. This system ensures that information control remains with you. If a company policy changes, simply update the document in the Knowledge Base, and the AI will immediately start using the new information. Who is this feature for? Creating a Knowledge Base is ideal for any business that wants to scale customer service intelligently and controllably. - E-commerce and Retail: To answer questions about products, shipping, order status, return policies, and technical specifications. - SaaS and Technology Companies: To provide first-level support, answer questions about features, plans, pricing, and configuration guides. - Service Providers: To inform about service types, appointments, pricing, coverage areas, and contractual terms. - Educational Institutions: To answer questions about courses, enrollments, admission processes, and academic information. - B2B Operations: To quickly provide complex technical and commercial information to potential clients and partners. Frequently Asked Questions (FAQ) What types of documents can I use to feed the AI? You can use text-based documents containing the information you want to automate. Common examples include PDF files with product manuals, text documents with lists of frequently asked questions (FAQs), company policies, and service descriptions. For details on specific file formats, please consult the platform's documentation. Does the AI respond using only the Knowledge Base? The AI's behavior is configurable through modules like "Prompt & AI." You can instruct the assistant to strictly prioritize information from your Knowledge Base to ensure maximum accuracy. In other scenarios, it's possible to combine the knowledge from the base with other prompt instructions for a more flexible dialogue. How does the AI know which document or snippet to use? RAG technology allows the AI to analyze the meaning of the customer's question and perform a semantic search across all documents in your base. It identifies the most relevant snippets for the query, even if the words are not exactly the same, and uses them to construct the response. Can I have more than one Knowledge Base? Yes. Papo allows the creation of multiple Knowledge Bases. Organizing your information into separate bases (e.g., "Sales," "Support," "HR") is a good practice to keep content organized and facilitate management and updates. Does using the Knowledge Base consume AI credits? Artificial intelligence features, such as lead categorization and response generation, may consume AI credits. The specific consumption model for Knowledge Base queries is detailed in your Papo account's plan and usage information. Do I need to know how to code to create and use a Knowledge Base? No. The Papo platform is designed for business users to create, feed, and manage their Knowledge Bases through a visual interface, without needing to write a single line of code. Read also - Super Chat: Your CRM Controlled by Text Prompt - Automatic AI Categorization: Optimizing Lead Management Start for free and automate your responses → Your account starts with the first user free for you to set up your operation.

How to Use the Flows Library to Automate Common Tasks

To automate common tasks, Papo (https://papo.global/) offers the Flows Library, a collection of ready-to-install automation templates. With it, you can quickly implement workflows for lead qualification, satisfaction surveys, customer onboarding, and other routines, without having to create the automation from scratch. The Flows Library is a powerful resource for accelerating the implementation of automations in your operation. Instead of designing each step of a conversational flow or an internal process, you can choose a pre-built template that already solves a common problem and adapt it to your needs. This saves time, reduces complexity, and allows teams of all skill levels to harness the power of automation. | Step | Action | Description | | :--- | :--- | :--- | | 1. Access | Navigate to the Flows Library | Within the Flows module on the Papo platform, locate the option to access the template library. | | 2. Explore | Find and preview a template | Browse categories (e.g., Sales, Support) and click to preview the flow's structure. | | 3. Install | Install the flow in your account | With one click, the template is copied to your list of flows. It is installed as inactive by default. | | 4. Customize and Activate | Edit and publish the flow | Review messages, adjust rules, and when ready, activate the flow to start using it. | What is the Flows Library? The Flows Library is a collection of pre-configured automation flows provided by Papo. Think of it as an app store for your automations. Each template in the library is designed to solve a specific business need, such as: - Lead Qualification: Ask initial questions to filter and prioritize new contacts. - New Customer Welcome: Send a sequence of messages to guide new customers. - Satisfaction Survey (CSAT): Automatically send a survey after a support interaction ends. - Demo Scheduling: Collect information and direct the lead to a salesperson's calendar. The main benefit is agility. Instead of spending hours building a complex flow, you can install a template in seconds and focus only on fine-tuning, such as customizing message text or defining which team a task will be sent to. How to Use the Flows Library Step-by-Step Implementing an automation from a template is a simple and secure process, as the flow only becomes active after you manually enable it. 1. Accessing the Library Within the Papo Global dashboard, navigate to the Flows module, where you manage all your automations. Look for the option that leads to the Flows Library or a templates section. 2. Browsing and Previewing Templates In the library, you will find various templates organized by category or use case, such as Sales, Marketing, Support, or Human Resources. - Browse options: Explore the available templates to find one that fits your needs. - Preview the flow: Before installing, you can click on a template to see a preview of its structure. This shows all the "nodes" (steps) of the flow, such as messages, conditions, and actions, allowing you to understand exactly what that automation does. 3. Installing a Flow from the Library Liked a template? The next step is to install it. By clicking the install button, a complete copy of that flow will be added to your list of automations in the Flows module. An important detail is that, by default, the flow is installed as INACTIVE. This is a security measure to ensure that an automation does not accidentally start running before being configured for your business. 4. Reviewing, Customizing, and Activating With the flow installed and inactive, you have complete freedom to adjust it. 1. Open the newly installed flow in your list of automations. 2. Review each node: Read the messages and adjust the text to reflect your brand's voice and policies. 3. Configure actions: Check actions such as "create a task in the Projects module," "add a tag to the contact," or "send an email" and configure them as needed. 4. Test the flow: If possible, use testing tools to simulate the customer journey. 5. Activate the flow: When you are satisfied with the customizations, simply change the flow's status from "Inactive" to "Active." From that moment on, it will be ready to be triggered and automate your tasks. Who is the Flows Library for? This resource is especially useful for: - Companies new to automation: Offers a clear and functional starting point, reducing the learning curve. - Marketing and Sales Teams: To quickly implement qualification funnels, lead nurturing, and scheduling. - Customer Support Managers: To standardize processes such as satisfaction surveys and welcome messages. - Operations that require agility: Allows testing and launching new automations in minutes, rather than days. - Users without technical knowledge: Simplifies the creation of complex flows, making automation accessible to everyone on the team. Frequently Asked Questions What exactly happens after I install a flow from the library? After installation, a copy of the template is created in your list of automations within the Flows module. It is added with the status "Inactive," meaning it will not perform any actions until you review and manually activate it. Can I edit a flow I installed from the library? Yes. Once installed, the flow is yours. You have full control to add, remove, or modify any step (node) of the flow, customizing it completely to your company's needs. Do flows from the library have any additional cost? Access to platform features, including the Flows Library, depends on your contracted plan. For detailed information on plans and costs, please consult the official Papo Global website. Do I need technical knowledge to use the templates? Advanced technical knowledge is not required. The library was created to simplify the process. However, having a good understanding of the business process you wish to automate will help you make the correct customizations to the template. Where can I find the Flows Library on the platform? Generally, the library is located within the main Flows section in the Papo Global menu. Look for a button or tab named "Library," "Templates," or similar. Start automating your tasks for free → Your free account allows you to explore the automation modules and set up your first flows.

What is Live Coach and How It Provides Real-Time Suggestions for Salespeople

Live Coach is an Artificial Intelligence assistant from Papo (https://papo.global/) that acts as a copilot for your sales agents during real-time conversations. It analyzes the dialogue with the customer and, based on a pre-configured sales script (playbook), offers suggestions for responses, arguments, and ways to overcome objections, ensuring your team follows best practices and increases conversion chances. This tool, also known as Captain, is designed to empower salespeople, from the newest to the most experienced, by offering the right support at the exact moment of interaction. Instead of making the agent search for information in documents or rely on memory, Live Coach delivers the company's knowledge directly to the chat screen, privately and contextually. | Aspect | Service Without Live Coach | Service With Live Coach | | :--- | :--- | :--- | | Consistency of Approach | Each salesperson uses their own style, which can lead to inconsistent messaging. | The entire team follows a communication standard defined in the playbook, strengthening the brand. | | Responding to Objections | The agent might hesitate or not have the best argument for complex objections. | AI suggests the best responses for common objections, prepared by company experts. | | Training New Agents | The learning curve (onboarding) is long and depends on supervisor oversight. | New salespeople become productive faster as they receive real-time guidance. | | Efficiency and Agility | The agent wastes time searching for product or pricing information in other windows. | Necessary information and arguments appear on screen, speeding up responses and the sales flow. | How Live Coach Works Live Coach integrates the intelligence of your sales operation directly into the agent's conversation interface. The process happens in three main steps: 1. Sales Playbook Configuration: The first step is to "teach" the AI. You feed Live Coach with the information it will use to generate suggestions. This playbook includes: - Product/Service Context: Descriptions, technical specifications, prices, and unique selling propositions. - Tone of Voice: Define how communication should be (e.g., consultative, direct, friendly, or formal). - Objections and Arguments: Input the most common objections ("it's too expensive," "I need to think about it," "your competitor offers X") and the best counter-arguments for each. - Inboxes: Choose which channels Live Coach should operate in (e.g., only in Sales WhatsApp or also on Instagram). 2. Real-Time Conversation Analysis: Once activated, Live Coach monitors the conversation between the agent and the customer. The AI processes customer messages to identify intentions, questions, product inquiries, and, most importantly, objections. 3. Contextual Suggestion Generation: Based on the analysis and the configured playbook, the system displays relevant suggestions to the agent. It's important to note that these suggestions are private and appear only to your employee. They have full autonomy to use the suggestion, adapt it, or ignore it. Generating these suggestions may consume AI credits from your account. The goal is not to automate responses, but to empower the human, transforming each agent into an expert in their product and sales process. Who is Live Coach for? This functionality is especially valuable for commercial operations seeking standardization, efficiency, and continuous improvement. - Growing Sales Teams: Helps standardize the sales pitch and ensures that all salespeople, new or old, communicate the product's value consistently. - Sales Managers: Acts as a continuous training tool, ensuring that the defined sales process is followed and accelerating the learning curve for new hires. - Businesses with Consultative Sales or Complex Products: Where technical details and the correct handling of objections are crucial for gaining customer trust and closing the sale. - High-Volume Service Operations: Allows maintaining a high standard of quality in interactions, even when the team is managing multiple conversations simultaneously. By using Live Coach, companies reduce reliance on the individual "talent" of each salesperson and create a more predictable and scalable sales system. Frequently Asked Questions Does Live Coach respond to the customer automatically? No. Live Coach is a copilot for the human agent. It displays suggestions privately to the salesperson, who decides if, how, and when to use that information in the conversation with the customer. It does not send automatic messages. How does Live Coach know what to suggest? It combines two sources: the sales playbook you configure (with product information, tone of voice, and arguments for objections) and real-time conversation analysis performed by Artificial Intelligence. What is a "playbook" in the context of Live Coach? The playbook is the set of rules and information you provide to guide the AI. It's your digitized sales script, containing everything a salesperson needs to know to have a successful conversation. Can I use Live Coach across all my service channels? Yes. During configuration, you can define which inboxes (WhatsApp, Instagram, Facebook, etc.) connected to Papo Live Coach should be active in. Does using Live Coach incur additional costs? The functionality itself is part of the Papo platform modules. However, generating suggestions via Artificial Intelligence may consume AI credits, depending on your plan and usage. Details about credit consumption are available on the platform. What is the difference between Live Coach and a chatbot like Super Chat? Super Chat is designed to automate conversations and respond directly to customers, qualifying leads and resolving queries without human intervention. Live Coach, on the other hand, operates in the background, assisting a human agent who is actively conducting the conversation. Start using Papo for free → Centralize your customer service and sales with AI to never miss another opportunity.

Meet Voice Assist: The voice assistant for your website

Voice Assist is a Papo (https://papo.global/) feature that adds an AI-powered voice assistant to your website's chat widget. It allows visitors to ask questions and interact using voice commands, offering a more modern, accessible, and faster customer service experience, with instant answers based on your Knowledge Base. In a world where voice interaction with devices has become common, bringing this capability to your website is a competitive advantage. Voice Assist transforms the traditional text chat into a dynamic conversational touchpoint, where customers can simply speak instead of type, making communication more natural and efficient, especially on mobile devices. | Aspect | Traditional Chat Widget | Widget with Voice Assist | | :--- | :--- | :--- | | User Interaction | Limited to text typing. | Allows voice and text interaction. | | Accessibility | Can be challenging for people with motor or visual difficulties. | Increases accessibility, allowing more people to interact easily. | | Mobile Device Experience | Typing on small screens can be slow and error-prone. | Speaking is faster and more practical than typing on a mobile phone. | | Modernity and Innovation | Standard functionality, expected by most users. | Positions the brand as innovative and attentive to new technologies. | | Query Speed | The user needs to formulate and type the entire question. | The user simply speaks, speeding up the start of the conversation. | How does Voice Assist work? Voice Assist is integrated directly into the chat widget you install on your website through the Papo platform. When a visitor clicks the chat icon, they have the option to start the conversation by voice. The process occurs in a few simple steps: 1. Voice Activation: The visitor clicks the microphone icon in the widget and asks their question aloud. 2. AI Processing: The Papo platform transcribes the audio to text and analyzes it using the same artificial intelligence that powers Super Chat. The AI searches for the most relevant answer in your Knowledge Base (documents, FAQs, company policies). 3. Smart Response: The assistant formulates a cohesive response and presents it to the user. Depending on the configuration, the response can be in text or also in audio, simulating a real conversation. 4. Actions and Handoff: In addition to answering questions, Voice Assist can be configured to perform actions, such as directing the user to WhatsApp, scheduling a meeting (if integrated with Google Agenda), or transferring the conversation to a human agent when necessary. The entire "brain" of Voice Assist is the same AI system you already configure in Papo, including the assistant's prompt and the Knowledge Base. This ensures consistency in information, regardless of whether the customer types or speaks. How to set up Voice Assist on your website Voice Assist activation is done within the settings of your "Website" type inbox on the Papo platform. 1. Access Inboxes: In the Papo dashboard, go to Settings → Inboxes. 2. Select the Website Widget: Locate the inbox corresponding to your website or create a new one by selecting the Website option. If it's a new one, follow the steps to name it and get the installation script. 3. Enable Voice Assist: Within the website widget settings, you will find the options for Voice Assist. Activate the feature. 4. Define the Prompt: Associate Voice Assist with a specific AI prompt. This prompt will guide the assistant's tone of voice and behavior. 5. Configure Actions: Define the interaction modes, such as the options the assistant can offer at the end of a conversation (e.g., start a WhatsApp conversation, schedule a meeting, or speak with an agent). 6. Save Changes: After configuring, save the changes. The voice assistant will be active on your website's widget. Who is Voice Assist for? This feature is ideal for companies looking to innovate in customer experience and optimize communication on their main digital channel: the website. - E-commerce and Retail: For customers who want to quickly ask questions about products, shipping, or order status while browsing, without having to type. - Technology and SaaS Companies: To offer a first level of technical support or answer questions about features interactively. - Service Providers: To facilitate requesting quotes or clarifying questions about services, making the first contact more agile. - Real Estate Sector: For potential buyers to ask by voice about property details while viewing the page. - Businesses Focused on Accessibility: For companies that want to ensure their website is accessible to all audiences, including people with disabilities that make typing difficult. Frequently Asked Questions (FAQ) Does the voice assistant also respond with audio? Yes. Papo's artificial intelligence can be configured to respond to customers' voice messages with its own audio responses, making the interaction even more natural and conversational. Where does Voice Assist get its information to respond? It uses the same Knowledge Base (RAG) that you configure for your Super Chat. You feed the platform with documents, FAQs, and information about your company, and the AI uses this material as the exclusive source to generate accurate responses. Can Voice Assist transfer service to a human? Yes. Just like text chat, the voice assistant can be configured to transfer the conversation to an available human agent whenever the AI doesn't know the answer or when the visitor explicitly requests it. Is the setup complex? No. Activating and configuring Voice Assist is done directly in the Papo panel, within your website's inbox settings, without the need for programming. Does Voice Assist work well on mobile phones? Yes, it is especially useful on mobile devices, where typing long questions can be inconvenient. Speaking with the assistant makes the user experience on mobile much more fluid and faster. Can I define what the voice assistant can or cannot do? Yes. Through the AI prompt configuration, you define the rules, objective, tone of voice, and limitations of your assistant, ensuring it acts according to your company's guidelines. Read also - Feeding the AI: How to Create a Knowledge Base for Automatic Answers - What is Live Coach and how it provides real-time suggestions to salespeople Start for free and activate your voice assistant → The first user of your account is free, allowing you to configure and test automations.

Creating Your First Automation Flow from Scratch

To create your first automation Flow from scratch in Papo (https://papo.global/), access the "Flows / Chatbot" module, create a new visual flow, and add the nodes that will define the contact's journey. You will start with an initial trigger, connect actions such as sending messages or asking questions, and, after activating the Flow, you can enroll contacts to start the automation. Flows are one of Papo's most powerful tools for automating interactions, qualifying leads, conducting surveys, and much more, all without the need for human intervention at every step. They function like a visual flowchart where you design your customer's journey. Flow Components Every Flow is built from a few fundamental blocks. Understanding the role of each is the first step to creating effective automations. | Component | Description | Example Usage | | :--- | :--- | :--- | | Trigger (Start) | The starting point that initiates the Flow for a contact. | A contact is manually enrolled in the Flow or by another automation. | | Action | A task that the Flow executes automatically. | Send a welcome message, add an "Interested" tag, update a contact field in the CRM. | | Condition | A check that splits the Flow's path based on criteria. | If the contact answered "yes" to a question, follow path A; if they answered "no", follow path B. | | Completion | The point where the automated journey ends for that contact. | After the contact receives the last message in the sequence or is transferred to an agent. | How to Create Your Flow Step-by-Step Building an automation might seem complex, but Papo's visual editor simplifies the process. Follow these steps to create your first Flow. 1. Access the Flows Module In the side menu of your Papo dashboard, locate and click on the Flows / Chatbot option. This area centralizes all your flow-based automations. 2. Create a New Flow Within the module, click the button to add a new Flow. The platform will generate an empty and inactive workflow, containing only the initial ("Start") node. Your entire automation will be built from there. 3. Build the Visual Structure The Flow editor is a blank canvas where you add "nodes" (the boxes representing each step) and connect them with lines to define the logical sequence. Drag and drop components to organize your journey. 4. Configure Action Nodes Click the "+" sign or an empty area to add new nodes after "Start". Each node represents an action the system should perform. Common actions include: - Send Message: Send text, images, videos, or documents to the contact. - Ask a Question: Capture the user's response to use in condition nodes. This is the basis for creating triage, surveys, or interactive forms. - Add/Remove Tag: Organize and segment your contacts based on their interactions within the Flow. - Update Contact: Modify fields in the contact's profile, such as the funnel stage in Kanban or the customer status. - Enroll in another Flow: Route the contact to another automation, creating more complex and modular journeys. 5. Use Condition Nodes to Create Branches The true intelligence of a Flow lies in its conditions. After an "Ask a Question" node, for example, you can add a condition node that creates different paths for each type of response. If the customer replies "I want to buy," the Flow can direct them to a salesperson; if they reply "I need support," the Flow routes them to the support team. 6. Activate the Flow Once your flow is complete and tested, don't forget to save and activate it. Only Flows with "Active" status can receive and process contacts. How to Enroll Contacts in a Flow An active Flow does nothing until contacts are enrolled in it. Enrollment is the act of inserting a contact at the beginning of the journey you designed. There are several ways to do this: - Manually: You can open a contact's profile in the CRM and choose which active Flow you want to enroll them in. Ideal for testing or one-off actions. - In Bulk: Filter a list of contacts in the CRM and apply the action to enroll all of them in a specific Flow. - Via Automation: Other automations or even another Flow can have the enrollment of a contact in a new flow as their final action. - Via API: For advanced integrations, your external system (such as a website or ERP) can send a command to Papo's API to enroll a new contact in a Flow, for example, immediately after they fill out a form. Frequently Asked Questions What is the difference between a Flow and an Email Sequence? Flows are interactive journeys that can use multiple channels (like WhatsApp) and react in real-time to user responses with complex conditions and actions. Email Sequences, such as the "Lead Nurturing" sequence that comes pre-configured in your Papo account, are linear and focused on sending a series of predefined emails at timed intervals. Can I use Papo's AI within a Flow? Yes. Papo was designed with AI at the center of its operations. Within a Flow, you can use artificial intelligence to interpret open-ended user responses, categorize intentions, and direct the flow more intelligently. You can even trigger Super Chat for a more natural conversation before returning the contact to the Flow or a human agent. How do I test my Flow before activating it for everyone? The best practice is to use a test phone number (yours, for example) registered as a contact on the platform. After building the Flow, activate it, but do not publish it. Manually enroll your test contact and interact with the automation to ensure all paths, conditions, and actions work as expected. Can a contact be in more than one Flow at the same time? Details regarding the simultaneous execution of Flows may vary depending on account settings. It's important to consult the platform's documentation or check business rules to understand how Papo manages competition between automations for the same contact. What happens if a contact responds with something the Flow doesn't expect? Good Flows anticipate alternative paths. You can configure a "default response" for invalid inputs or, in more complex cases, create a "handoff" action that transfers the conversation to a human agent. This ensures no contact is left without a response and that the customer experience is always positive. Read also - How to Use the Flow Library to Automate Common Tasks - Feeding the AI: How to Create a Knowledge Base for Automatic Responses Start automating for free → The first user is free, and no credit card is required to get started.

Meet Captain: How to Configure Your Personalized AI Assistant

Captain is Papo's (https://papo.global/) customizable artificial intelligence assistant, designed to automate responses and tasks across your customer service channels. You can train it with your company's specific knowledge base, including documents, FAQs, and action scenarios, so it can answer customers, qualify leads, and assist your team with accurate and contextual information, 24 hours a day. With Captain, you transform your customer service, leaving repetitive responses to automation and freeing your agents to focus on complex and strategic cases. It functions as a central brain that can be connected to various inboxes, such as WhatsApp, Instagram, or your website's chat, ensuring consistency and agility in communication. | Knowledge Type | Description | Use Cases | | :--- | :--- | :--- | | Documents | Files (PDF, DOCX, TXT) with detailed information, such as manuals, privacy policies, or product catalogs. | Answer complex questions about a product's technical specifications by consulting its manual. | | FAQs | A list of direct questions and answers for common queries. | Provide quick answers about operating hours, payment methods, or return policy. | | Scenarios (Handoffs) | Specific instructions for what the AI should do in certain situations, such as transferring to a human. | When a customer says "I want to speak to an agent," Captain executes the handoff action to the team. | | Tools | Actions the assistant can perform, such as scheduling meetings or querying external systems via API. | A customer asks to schedule a demo, and Captain checks the calendar and books the time. | How Captain Works? Captain operates based on a knowledge base that you build and manage. Instead of using generic internet responses, it exclusively consults the sources you provide. This ensures that the information is always aligned with your company's policies and data. The process is simple: when a customer sends a message, Captain analyzes it and searches for the most relevant answer within its trained "brain." It can combine information from a PDF document with an FAQ response to formulate a complete and cohesive explanation. If the question requires an action, such as transferring the service, it activates the corresponding scenario. Training Your Captain Assistant: Step-by-Step Configuring your Captain assistant involves feeding it the necessary knowledge so it can act autonomously and efficiently. 1. Create Your Assistant: The first step is to create a new assistant on the Papo platform. You can give it a name and description, defining its identity and purpose. It's possible to have multiple assistants for different areas of the company, such as one for sales and another for support. 2. Feed with Documents: Use Papo's File Center to upload important documents: technical manuals, commercial policies, catalogs, terms of service, etc. After uploading, you link these files to your Captain's knowledge base. The AI will then use the content of these documents to answer questions. 3. Add Frequently Asked Questions (FAQs): Register a list of questions and answers for your customers' most common queries. This creates a shortcut for quick and direct responses. The platform can also suggest new FAQs based on customer questions that the assistant couldn't answer, allowing you to continuously improve the knowledge base. 4. Configure Action Scenarios (Handoffs): Define rules and instructions for specific situations. The most common scenario is "handoff," where you instruct Captain to transfer the conversation to a human agent if the customer requests it or if the topic is too complex. 5. Enable Tools: Captain can go beyond answering questions. You can enable native "tools," such as integration with Google Calendar to schedule meetings, or configure "custom tools" via HTTP so the assistant can query or send data to your own systems (ERPs, CRMs, etc.). 6. Link to an Inbox: With the assistant trained, the final step is to activate it on one or more channels. You can link your Captain to a WhatsApp inbox, Instagram, your website's chat, or any other channel connected to Papo. From that moment on, it will begin interacting with customers on that channel. Frequently Asked Questions (FAQ) What is the difference between training with Documents and with FAQs? Documents are ideal for broad and detailed knowledge, where the AI needs to read and interpret context to formulate a response, like a 50-page manual. FAQs are for direct and objective answers to specific questions, such as "What is the store's address?". Using both makes the assistant more robust. What happens if Captain doesn't know the answer? You configure the default behavior for this situation. The best practice is to create a "handoff" scenario that automatically transfers the conversation to a human agent, ensuring the customer never goes unanswered. Can I have more than one Captain assistant? Yes. The Papo platform allows you to create and manage multiple assistants. You can have a Captain specialized in sales, with knowledge about products and prices, and another focused on technical support, trained with manuals and troubleshooting guides. Does Captain work on all channels (WhatsApp, Instagram, etc.)? Yes. Captain is a central intelligence that can be activated in any inbox configured in your Papo account. Simply link the desired assistant to the inbox of the corresponding channel (WhatsApp, Facebook Messenger, website chat, etc.). What are pending FAQ suggestions? It's a continuous improvement feature. When Captain receives a question it cannot answer with the current knowledge base, the system can record that question as a "pending FAQ suggestion." An administrator can then review these suggestions, create the correct answer, and approve it, making the assistant's knowledge base increasingly comprehensive. What are Captain's "Tools"? These are tools that allow Captain to perform actions. Native tools are functionalities integrated into Papo, such as scheduling an event in Google Calendar. Custom tools allow developers to connect Captain to external systems via HTTP calls (API), enabling the AI to check an order status in an e-commerce store or register a lead in an external CRM, for example. Read also - How to Manage Documents in Papo's File Center: /hc/papo/articles/como-gerenciar-documentos-central-arquivos-papo - Meet Voice Assist: The voice assistant for your website: /hc/papo/articles/conheca-o-voice-assist-assistente-de-voz-para-seu-site Start using for free → Your account starts with the first user free for you to set up and test your automations.

What is Synapse and how to use the customer radar to find opportunities

Synapse is Papo's (https://papo.global/) intelligence module that functions as a customer radar and an advanced search tool for conversation history. It's designed to automatically analyze all interactions, identify which customers need immediate attention, and allow you to uncover business opportunities or churn risks that would otherwise be lost among thousands of messages. Over time, customer conversation history becomes a goldmine of information. However, without the right tools, this valuable data remains inaccessible. Synapse solves this problem by using artificial intelligence to scan, interpret, and organize this information, transforming your service history into an active source of business intelligence. It has two main functionalities that work together to give your team a strategic advantage: the Customer Radar and History Search. | Functionality | What it does | What it's for | | :--- | :--- | :--- | | Customer Radar | Lists customers who need attention, ordered by criticality. | Prioritize follow-ups, prevent customer loss (churn), and identify sales opportunities or proactive support. | | History Search | Allows searching for specific terms across the entire conversation history. | Find mentions of products, competitors, technical issues, or purchase intentions in any conversation. | How Synapse Helps Find Opportunities In a busy operation, it's impossible for a human to read every message exchanged to identify patterns or opportunities. Phrases like "I'm thinking of upgrading," "your competitor offered me this," or "I've had this problem for days" often get lost in the daily volume of conversations. Synapse acts as an intelligence layer over your unified inbox. It reads and analyzes the content of conversations across all channels connected to Papo (WhatsApp, Instagram, etc.) to identify important signals. Based on this analysis, it feeds its two main tools, ensuring that no crucial information goes unnoticed. Using the Customer Radar to Prioritize Service The Customer Radar is your daily, AI-generated priority list. Instead of your team having to randomly decide who to call or which conversation to resume, the radar presents an ordered list of contacts who need the most attention, from most to least critical. "Criticality" is determined by AI based on a combination of factors, such as: - Signs of dissatisfaction: Mentions of problems, complaints, or words indicating frustration. - Sales opportunities: Customers who have mentioned interest in new products, services, or upgrades. - Churn risk: Contacts who have mentioned cancellation or who have gone a long time without interacting after a negotiation. - Lack of response: Leads who asked an important question and did not receive an adequate follow-up. By starting the day by looking at the Customer Radar, your team can act proactively, contacting the right people at the right time, whether to resolve a problem before it escalates or to close a sale that might otherwise cool off. Performing Smart Searches in Conversation History Synapse's second function is a powerful search that goes beyond simply finding a contact. It allows you to search for keywords or phrases within your entire conversation history. This opens up a range of possibilities for sales, marketing, and support. Examples of use: - Sales: Search for "quote" or "proposal" to find all customers who requested a quotation in recent months and follow up. - Marketing: Search for a competitor's name to understand what your customers are saying about them and how your offering compares. - Customer Success: Look for terms like "cancel" or "dissatisfied" to identify at-risk customers and act preventively. - Product Development: Search for "I wish it had" or "suggestion" to collect valuable feedback directly from customers. This functionality transforms your chat history from a simple archive into a searchable and strategic database. Frequently Asked Questions (FAQ) What's the difference between normal contact search and Synapse search? The normal search in Papo's CRM allows you to find a specific contact by name, phone, or email. Synapse's search is deeper: it searches the content of all conversations, allowing you to find mentions of specific terms within the dialogue with customers. How does Synapse define a customer as "critical"? The platform uses artificial intelligence to analyze various signals in conversations, such as expressed sentiment (positive, negative, neutral), the use of keywords related to problems or opportunities, contact frequency, and the time since the last interaction. The combination of these factors generates a score that orders customers on the radar. Does Synapse work for all channels (WhatsApp, Instagram, etc.)? Yes. As Synapse operates on Papo's unified database, it analyzes conversations from all channels you've connected to your inbox, providing a complete and centralized view. Do I need to configure anything for Synapse to work? No. Synapse is designed to work automatically, leveraging the conversation history already present in your Papo account. It begins analyzing data as soon as interactions occur, with no manual configuration required. Does Synapse use customer tags? Synapse works independently, but its results can be even more powerful when combined with the tagging system. For example, you can use Synapse to find everyone who mentioned "upgrade" and then filter those results to see only those with the "Basic Plan Customer" tag, creating a perfect list for an upsell campaign. Start for free and uncover hidden opportunities → Papo offers your first user for free so you can organize your customer service and test the power of AI.

How to Configure Your AI Agents in Papo's Prompt Module

To configure your artificial intelligence agents, access the "Prompt and AI" module in the side menu of the Papo platform (https://papo.global/). In this area, you can create and edit the prompts that guide AI responses, define behavior rules, assign agents to specific channels, and analyze usage statistics to optimize your customer service automation. The Prompt and AI module centralizes control over how your artificial intelligence interacts with customers. You can have a main agent, which serves as the default for the entire account, and create sub-assistants with specific instructions for different products, channels, or service phases. The table below summarizes the main configuration areas available in the module: | Configuration Area | What you can do | | :--- | :--- | | Agent List | View the "General" agent (default) and all created sub-assistants. | | Prompt Editor | Define the instructions the AI will follow. This can be free text or divided into sections. | | General Rules | Establish behavior guidelines that apply to all agent interactions. | | AI Settings | Make technical adjustments to the AI model, such as response creativity. | | Inboxes | Enable or disable AI for a specific channel (e.g., WhatsApp, Instagram) and choose which agent will respond. | | Statistics | Monitor AI performance, with data on messages sent, failures, and skipped interactions. | How to Access the Prompt and AI Module To start configuring your assistants, follow these steps: 1. Log in to your Papo account. 2. In the left-hand side menu, locate and click on Prompt and AI. If the item is not visible in your menu, there might be two reasons: - Module disabled: The module might not be active for your account. - Lack of permission: Your account administrator might have restricted your access using custom roles. In this case, ask the administrator to review your permissions. Step-by-Step Guide to Configure an AI Agent Within the module, you will find the "General" agent and the option to create new sub-assistants. The configuration process is the same for both. 1. Select or Create an Agent You can edit the "General" agent, which functions as the default, or click to add a new sub-assistant. New sub-assistants are created as inactive drafts, allowing you to configure them completely before activating them. 2. Define the Instructions (Prompt) This is the most important step. The prompt is the "brain" of your agent. You can provide instructions in two ways: - Free Text: Write a detailed paragraph describing who the AI is, its objective, the tone of voice it should use, what it can and cannot do. - Structured Instructions: Fill in up to seven predefined sections to better organize the guidelines, covering aspects such as persona, objectives, restrictions, and response examples. A good prompt is clear, specific, and provides enough context for the AI to respond accurately and in line with your brand. You can also preview how the final prompt will be interpreted by the system. 3. Adjust General Rules and Settings In the agent's settings area, you can define general rules that apply to all its interactions and adjust AI model parameters. The platform comes with default settings that work well for most cases, but advanced users can make fine adjustments. 4. Activate the Agent in an Inbox A configured agent, even if saved, will only take action when associated with one or more inboxes. 1. Go to the Inboxes tab or section within the Prompt module. 2. You will see a list of your connected channels (WhatsApp, Instagram, etc.). 3. For each channel, you can turn the AI on or off. 4. When turning it on, select which agent (the General or a sub-assistant) should be responsible for that channel. This allows, for example, having a sales-specialized agent responding on WhatsApp and another focused on support responding to Instagram comments. Analyzing Usage Statistics To understand if your AI is working as expected, access the Statistics area. There, you will find graphs and numbers about AI interactions, including: - Succeeded: The number of messages the AI successfully sent. - Failed: Interactions where the AI attempted to respond but an error occurred. - Skipped: Situations where the AI decided not to respond, either due to a rule or a lack of confidence in the answer. Analyzing this data helps identify the need to adjust prompts, rules, or the knowledge base that feeds the AI. Frequently Asked Questions What is the difference between the main agent and sub-assistants? The "General" or main agent is the default for your account. If you activate AI in an inbox without specifying an agent, it will be the one to respond. Sub-assistants are specialized agents you can create for specific tasks (e.g., lead qualification, technical support) and assign to distinct channels or situations. What happens if I don't configure an AI agent? If you don't edit the prompt, the agent will use a set of default Papo instructions, which are generic. To get personalized responses aligned with your business, it is crucial that you edit the "General" agent's prompt with your own information. Can I have a different AI for each channel (WhatsApp, Instagram)? Yes. In the "Prompt and AI" module, you can access the configuration of your inboxes and assign a specific agent to each one. You can use the "General" agent for one channel and a specialized sub-assistant for another. How do I know if my AI is responding correctly? The best way is to monitor conversations in the inbox and analyze the "Statistics" section in the Prompt module. Check the success rates and review conversations where the AI interacted to ensure that the responses are aligned with your instructions. I can't see the "Prompt and AI" module. What should I do? First, check with your Papo account administrator if the module is enabled for the company and if your user profile has permission to access it. Access to advanced modules is managed through custom roles. Read Also - Feeding the AI: How to Create a Knowledge Base for Automatic Responses - Creating your first Automation Flow from scratch Start for free → The first user is free, no credit card required.

How to Automate Your Instagram: Replies to Comments, Mentions, and More

Yes, it's possible to automate replies to Instagram comments, mentions, and direct messages (DMs) to capture leads and streamline customer service using the Papo platform (https://papo.global/). The tool allows you to create rules to automatically send a DM when someone comments on a post, react to mentions in Stories, and initiate conversation flows based on keywords, all within a unified inbox with CRM and AI. Instagram automations are an efficient way to ensure no customer interaction is lost. They allow your business to respond instantly, 24 hours a day, qualifying leads and directing them to the next step, whether it's a sale, an appointment, or a material download. | Instagram Automation | How it Works with Papo | Key Benefit | | :--- | :--- | :--- | | Comment to DM | A user comments a keyword on a post; Papo replies publicly and sends a private DM with a link or message. | Capture the lead from the public feed into a private, measurable conversation, increasing campaign conversion. | | React to Mentions | When your brand is mentioned in a Story, the mention appears in Papo's inbox, which can trigger a thank-you flow or a CRM tag. | Strengthen customer relationships by showing the brand is attentive, and organize mentions for marketing analysis. | | DM Triggers | A user sends a DM with a specific word (e.g., "price", "catalog"); Papo initiates an automation flow to answer that question. | Provide immediate support for common questions, freeing up human staff for more complex cases. | | AI in Conversations | After the initial automation, Papo's AI can continue the conversation, qualify the lead, understand audio, and schedule meetings in Google Calendar. | Maintain lead engagement with intelligent and contextual responses, even outside business hours. | How Instagram Automations Work with Papo Papo integrates directly with the Instagram API, allowing all interactions (DMs, comments, mentions) to arrive in a centralized inbox. From there, the Automation and Flows modules spring into action. - Automatic reply to comments: You define one or more keywords for a specific post (or for all posts). When a user comments using that word, the automation replies to the comment (e.g., "I've messaged you privately!") and simultaneously sends a direct message to the person's profile. This is ideal for campaigns like "Comment 'I WANT IT' to receive the link." - Reacting to mentions in Stories: Mentions in Stories appear as a new conversation in your Papo inbox. You can create rules to automatically reply to these mentions with a thank-you message or to apply a tag to the contact, such as "brand-advocate," facilitating future marketing actions. - Keyword triggers in DMs: Using the visual automation builder (Flows), you can create complete journeys that are initiated when a follower sends a message with terms like "quote," "support," or "hours." The flow can ask questions to qualify the lead and, if necessary, transfer the conversation to a human agent with the full history. Step-by-Step to Set Up Your First Automation Setting up a comment-to-DM automation is a simple process done directly in the Papo dashboard. | Step | Action | Details | | :--- | :--- | :--- | | 1. Connect your account | Go to Settings → Inboxes and add your Instagram profile. | You will need to log in with a Facebook account that is an administrator of the page linked to your Instagram business profile. | | 2. Create the rule | Access the Automations module and create a new rule. | Set the "trigger" as "New Instagram post comment." | | 3. Define the conditions | Specify the conditions for the rule to be triggered. | You can filter by a specific post, by keywords in the comment (e.g., "comment text contains 'want'") or leave it open for all. | | 4. Configure the actions | Determine what the automation will do. | Add two actions: 1) "Send reply to comment" (with public text) and 2) "Send private message to contact" (with the link or main information). | | 5. Activate and test | Save and activate the rule. | Make a test comment on the configured post using another Instagram account to verify that the automation works as expected. | Who is this automation for? Instagram automation with Papo is ideal for businesses and content creators who use the platform as a customer acquisition and sales channel. It is especially useful for: - E-commerce and infoproduct creators: To automate the delivery of discount coupons, checkout links, and access to rich materials. - Local businesses: Clinics, restaurants, and service providers who receive many questions about scheduling, location, and menus. - Marketing and sales teams: To ensure that no lead generated by Instagram campaigns is lost due to lack of quick response. - Businesses with high interaction volume: Who need a scalable way to manage comments and DMs without proportionally increasing their team. Frequently Asked Questions Do I need another tool like ManyChat to automate Instagram? No. Papo has native Instagram automation functionality. You can configure replies to comments, DMs, and mentions directly on the platform, without the need to subscribe to a separate tool. Does automation only reply to comments? No. In addition to replying to comments and sending DMs, you can create automations to react to mentions in Stories and to initiate conversation flows when a user sends a direct message with a specific keyword. How much does it cost to use Instagram automation with Papo? Instagram automation, including comment replies, is already included in Papo's plans. The starter plan offers the first user free forever, allowing you to begin automating your customer service without initial cost. Can I use Papo's AI in Instagram conversations? Yes. Papo's AI works across all connected channels, including Instagram. After an automation initiates the conversation, the AI can take over to qualify the lead, answer frequently asked questions, and even schedule appointments, only passing to a human agent when necessary. Is it complicated to set up automations? No. Papo offers a visual automation builder (Flows module) where you drag and drop blocks to create the logic. For simpler rules, such as comment-to-DM, the setup is done through a step-by-step form. Can I connect other channels besides Instagram? Yes. Papo is an omnichannel platform. You can centralize customer service from Instagram, WhatsApp (Official and unofficial API), Facebook, Telegram, Email, SMS, and website chat in a single inbox. Start for free → The first user is free forever, with no credit card required.

How to Configure and Train AI Automatic Conversation Categorization

Papo's (https://papo.global/) automatic conversation categorization uses artificial intelligence to analyze, classify, and score all your leads, allowing your team to focus on contacts with the highest potential. Setup begins with evaluating conversations in your inbox, and the model is continuously trained through re-evaluations and the addition of contextual information, such as traffic source and the use of tags. Papo's AI categorization feature is designed to bring efficiency and strategic focus to your sales and customer service teams. By analyzing conversation content, the AI assigns a score to each lead, helping to quickly identify which contacts are "hot" and deserve immediate attention. | Feature | Description | Key Benefit | | :--- | :--- | :--- | | Lead Scoring | AI assigns a star rating to each contact based on conversation analysis. | Focus on leads with the highest conversion potential. | | Batch Evaluation | Allows analyzing all conversations in an inbox at once. | Time savings and an overview of contact base quality. | | Re-evaluation | Option to request a new AI analysis for a contact after new interactions. | Keeps the lead score always updated and accurate. | | Source Definition | Allows manually recording the lead's origin (e.g., campaign, video, event). | Improves analysis of the return from specific marketing channels. | How AI Categorization and Scoring Works Papo's Artificial Intelligence system reads the interaction history with a customer to identify signs of interest, urgency, and purchase potential. Based on this analysis, it performs two main actions: 1. Categorization: Groups conversations and contacts based on their potential. 2. Scoring: Assigns a rating, represented by stars, indicating how "hot" a lead is. A higher number of stars means a lead with a greater probability of closing a deal. This process transforms your inbox into a priority list, allowing agents to direct their efforts where the return is most likely, optimizing time and increasing conversion rates. Step-by-Step Guide to Evaluating Your Conversations To start using automatic categorization, you need to ask the AI to analyze conversations in one of your inboxes. Each analyzed conversation consumes an AI credit. | Step | Action | Details | | :--- | :--- | :--- | | 1 | Access the Feature | In the main dashboard, click on the "Evaluate open conversations" option, usually located in the upper left corner. | | 2 | Select the Inbox | Choose which inbox (WhatsApp, Instagram, etc.) you want the AI to analyze. | | 3 | Start Evaluation | Click the "Start evaluation" button for the AI to begin processing conversations and assigning scores. | | 4 | Analyze Results | After processing, you will see the star rating next to each contact in your conversation list. | How to Train and Refine the AI An AI model becomes more accurate over time and with more information. Although Papo's system learns automatically, you can accelerate and refine this process by providing more context. 1. Re-evaluating contacts after new interactions A lead may become more or less qualified after a new conversation. If a cold customer re-engages showing great interest, it's useful to update their score. - Select the desired contact from the conversation list. - In the contact information panel, on the right, click the "Re-evaluate" button. - The AI will perform a new analysis considering the most recent interactions and update the lead's score. 2. Defining the traffic source Knowing where your best leads come from is crucial for optimizing marketing investments. By informing a contact's origin, you enrich the data for future analysis. - In the same contact information panel, locate the source field (it may appear as "Unknown source"). - Click to edit and define where the lead came from (e.g., "Black Friday Campaign", "YouTube Video", "Instagram Ad"). 3. Using tags to create your own categories While the AI provides an automatic score, you can create your own manual categorization system using tags. This helps organize contacts granularly and can be used in conjunction with the AI's score. - Create tags relevant to your business, such as "Quote requested", "Technical issue", "Interest in Gold Plan", or "Follow-up in 7 days". - Apply tags to contacts during or after conversations. - Use these tags to filter and create segmented lists, combining your criteria with the prioritization suggested by the AI. By combining automatic scoring with your manual categorization, you create a robust lead management system tailored to your operation's specific needs. Frequently Asked Questions What do the stars in lead scoring mean? The stars are a visual representation of the lead's score. The more stars a contact has, the higher their conversion potential according to artificial intelligence analysis. This helps prioritize service. Does conversation evaluation have a cost? Yes, each conversation evaluated by artificial intelligence consumes an AI credit from your account. Details about the credits available in your plan can be found in your account settings or on Papo's pricing page. How often should I re-evaluate a contact? It is recommended to re-evaluate a contact whenever a significant interaction occurs that might change their level of interest. For example, if a client requests a proposal, schedules a demo, or returns after a long period of inactivity. Does the AI only analyze message text? The AI is trained to analyze the textual content of conversations. It looks for keywords, intent, sentiment, and other language patterns that indicate the customer's stage in the buying journey. Can I define the categories the AI uses? The AI autonomously assigns a score. However, you can and should complement this analysis by creating your own categories through Papo's tagging (labels) system. Using tags like "Qualified Lead" or "VIP Customer" in conjunction with the AI's scoring creates an even more powerful classification system. Does AI analysis work for all channels? Yes, the feature is designed to work with conversations from any inbox connected to Papo, including WhatsApp, Instagram, Facebook Messenger, Email, and other text-based channels. Start using AI to qualify your leads for free → Your free Papo account allows you to start organizing your conversations and test the power of automation.

How to Set Up and Train AI-Powered Automatic Conversation Categorization

Papo's (https://papo.global/) automatic conversation categorization uses artificial intelligence to analyze and assign a score to each conversation, allowing your team to prioritize leads with the highest conversion potential. Setup is done by requesting an initial evaluation of your inbox, and the AI model is continuously trained through re-evaluations and the addition of contextual information, such as lead origin and the use of tags. Papo's AI categorization feature is designed to bring efficiency and strategic focus to your sales and customer service teams. By analyzing conversation content, the AI assigns a score to each lead, helping to quickly identify which contacts are "hot" and deserve immediate attention. | Feature | Description | Main Benefit | | :--- | :--- | :--- | | Lead Scoring | The AI assigns a star rating (score) to each contact based on conversation analysis. | Focus on leads with the highest conversion potential. | | Batch Evaluation | Allows analyzing all conversations in an inbox at once. | Time savings and an overview of contact base quality. | | Re-evaluation | Option to request a new AI analysis for a contact after new interactions. | Keeps the lead score always updated and accurate. | | Origin Definition | Allows manually recording the lead's origin (e.g., campaign, video, event). | Improves analysis of the return from specific marketing channels. | How AI Categorization and Scoring Works Papo's Artificial Intelligence system reads the history of interactions with a customer to identify signs of interest, urgency, and purchase potential. Based on this analysis, it performs two main actions: 1. Categorization: Groups conversations and contacts based on their potential. 2. Scoring: Assigns a rating, represented by stars, indicating how "hot" a lead is. A higher number of stars means a lead with a greater likelihood of closing a deal. This process transforms your inbox into a priority list, allowing agents to direct their efforts where returns are most likely, optimizing time and increasing conversion rates. Step-by-step to activate automatic evaluation To start using automatic categorization, you need to request that the AI analyze conversations in one of your inboxes. Each analyzed conversation consumes an AI credit. | Step | Action | Details | | :--- | :--- | :--- | | 1 | Access the feature | In the main panel, click on the "Evaluate open conversations" option, usually located in the top left corner. | | 2 | Select the Inbox | Choose which inbox (WhatsApp, Instagram, etc.) you want the AI to analyze. | | 3 | Start Evaluation | Click the "Start evaluation" button for the AI to begin processing conversations and assigning scores. | | 4 | Analyze Results | After processing, you will see the star rating next to each contact in your conversation list. | How to train and refine AI accuracy An AI model becomes more accurate over time and with more information. While Papo's system automatically learns from new interactions, you can accelerate and refine this process by providing more context. 1. Re-evaluating contacts A lead may become more or less qualified after a new conversation. If a previously cold customer interacts again showing great interest, it's useful to update their score. - Select the desired contact from the conversation list. - In the contact information panel on the right, click the "Re-evaluate" button. - The AI will perform a new analysis considering the most recent interactions and update the lead's score. 2. Adding context with traffic origin Knowing where your best leads come from is crucial for optimizing marketing investments. By informing a contact's origin, you enrich the data for future analysis and train the system to recognize patterns from high-converting channels. - In the same contact information panel, locate the origin field (it may appear as "Unknown origin"). - Click to edit and define where the lead came from (e.g., "Black Friday Campaign", "YouTube Video", "Instagram Ad"). 3. Using tags for manual categorization While AI offers automatic scoring, you can create your own manual categorization system using tags. This helps organize contacts granularly and can be used in conjunction with AI scoring to refine segmentation. - Create relevant tags for your business, such as "Quote requested", "Technical issue", "Interest in Gold Plan", or "Follow-up in 7 days". - Apply tags to contacts during or after conversations. - Use these tags to filter and create segmented lists, combining your criteria with AI-driven prioritization. Frequently Asked Questions How does the AI decide what score to give a lead? The AI analyzes the content of exchanged messages, looking for keywords, questions about price, demonstrations of urgency, general sentiment of the conversation, and other indicators to determine the lead's level of interest. Is a lead's score updated automatically? No. The initial score is based on the history up to the time of evaluation. To update the analysis based on important new interactions, you must use the "Re-evaluate" function in the contact's profile. What are "AI credits"? Each conversation analyzed or re-evaluated by the categorization feature consumes an AI credit. For details on the number of credits included in your plan and additional costs, please refer to the information on our website. Can I define my own categories for the AI to use? The AI assigns a potential score, not custom textual categories. To create your own categories, such as "VIP Customer" or "Marketing Lead", you should use Papo's tagging system. What if I don't agree with the AI's score? The AI's score is a powerful suggestion to help prioritize, but human judgment remains essential. Use the score as a guide and complement the analysis with your own tags and notes in the contact's profile. Is conversation analysis instantaneous? Processing time depends on the volume of conversations in the selected inbox. Generally, it is a quick process, and scores begin to appear in your conversation list within minutes. Read also - What is Lead Score and how to use it to prioritize your best contacts - How to Configure your AI Agents in Papo's Prompt Module Start for free and activate AI for your leads → The first user on your account is free for you to validate the complete flow.

How to Use the Webhook Action in Automation Flows to Integrate with Other Systems

To integrate your automated processes with other systems, Papo (https://papo.global/) allows the use of the "Send Webhook" action within Automation Flows. With this feature, you can send real-time contact data to external platforms, such as your CRM, spreadsheets, or automation tools (like n8n and Zapier), whenever a contact reaches a specific stage in your flow. The Webhook action acts as a communication bridge, sending an automatic notification from one system (Papo) to another. This opens up a range of possibilities for creating connected work ecosystems, where information flows without the need for manual data export and import. For example, you can register a new lead in your main CRM the exact moment they are qualified by your chatbot on WhatsApp. | Configuration Parameter | Description | Usage Example | | :--- | :--- | :--- | | Webhook URL | The address (endpoint) of the external system that should receive the data. | URL provided by a system like n8n, Zapier, or your own backend. | | HTTP Method | The way data will be sent. POST is the most common for sending data. | Select POST to send information in the request body. | | Body (Payload) | The data you want to send, usually in JSON format, using Papo variables. | {"nome": "{{contact.name}}", "email": "{{contact.email}}", "telefone": "{{contact.phone}}"} | | Headers | Additional information for the request, such as content type or authentication keys. | Content-Type: application/json | How to configure the Webhook action step-by-step Integrating an external system via a Webhook in a Papo Flow is a technical but straightforward process. Follow the steps below to configure data sending. 1. Add the "Send Webhook" node to your Flow Within Papo's visual Flow editor, at the point in the journey where you want data to be sent, add a new action node. Find and select the "Send Webhook" action. This node will be responsible for triggering communication to the external system. For example, you can add this node right after a contact answers a qualification question or at the end of a welcome sequence. 2. Configure the Destination URL In the Webhook node's configuration panel, the first field to fill is the Webhook URL. This is the unique URL provided by the system that will receive the data. - Where to find this URL? You should obtain it from the documentation or dashboard of your external system (e.g., your CRM, ERP, or an automation tool like n8n). Tools like Webhook.site are also great for generating temporary URLs for testing. 3. Choose the HTTP Method You will need to choose the method for the request. The most common are: - POST: Recommended for most cases. It allows sending a larger volume of structured data in the request body (payload). It's ideal for creating or updating records in other systems. - GET: Used to request data from a server. Data is sent as parameters in the URL itself. It's less common for this type of integration, as it's less secure and limited in size. Select POST to send your contact's data. 4. Assemble the Request Body (Payload) The body (or "payload") is where you define which contact information will be sent. Papo allows you to use dynamic variables to insert the data of the contact passing through the Flow. The most commonly used format is JSON (JavaScript Object Notation). In the body configuration field, you can assemble a structure like this: { "nome_completo": "{{contact.name}}", "email_contato": "{{contact.email}}", "telefone": "{{contact.phone}}", "etiqueta": "Novo Lead Qualificado", "origem": "Flow de Boas-Vindas" } In this example: - {{contact.name}}, {{contact.email}}, and {{contact.phone}} are variables that Papo will automatically replace with the contact's real data. - "etiqueta" (tag) and "origem" (origin) are fields with fixed values that you can add to provide context for the information in the destination system. 5. Test and Activate the Flow Before activating your Flow for all contacts, it's crucial to test the integration. 1. Use a test URL (like Webhook.site) to verify that data is being sent correctly and in the expected format. 2. Enroll a test contact in your Flow and observe if the Webhook is triggered and if the external system receives the information. 3. After confirming that everything is working, replace the test URL with your system's production URL and activate the Flow. Frequently Asked Questions (FAQ) What is the difference between a Webhook and an API? Simply put, a Webhook is a "push" mechanism. Papo sends data to your system automatically when an event occurs. An API query, on the other hand, is a "pull" mechanism, where your system actively requests data from Papo. Webhooks are ideal for real-time notifications. How do I get the URL for my Webhook? You need to generate this URL in the system that will receive the data. Consult the documentation of your CRM, ERP platform, or automation tool (like n8n, Zapier, Make) to find the "Webhooks" or "API" section and create a new receiving endpoint. What contact data can I send via Webhook? You can send all standard contact data stored in Papo, such as name, phone, email, as well as custom fields and tags you have configured. Use variable syntax (e.g., {{contact.field_name}}) to dynamically insert this information into the request body. What happens if the external system is offline when the Webhook is sent? If the destination system does not respond or returns an error, the Webhook delivery will fail. Papo will attempt to send the notification, but guaranteed receipt depends on the availability and correct configuration of the receiving system. It's important to ensure your webhook endpoint is stable. Can I send data securely? Yes. Webhooks operate over the HTTPS protocol, which encrypts data in transit. For an extra layer of security, many systems allow the use of authentication keys (tokens) in the request headers, which you can also configure in Papo's "Send Webhook" action. Besides contacts, can I send other information? Yes. Although the most common use case is sending contact data, you can structure your Webhook's payload to include any static or dynamic information available in the Flow's context, such as the Flow's name, the answer to a specific question, or the event date. Start automating for free → Create your account and explore Automation Flows to connect your processes.

How to Use AI to Summarize Long Conversations and Get Up to Speed Quickly

Papo (https://papo.global/) uses artificial intelligence to analyze and summarize the context of long conversations, allowing agents to get up to speed on customer history in seconds, without having to read the entire dialogue. Through features like lead scoring, automatic categorization, and AI assistant insights (Capitão), the platform delivers the most important information for a quick and accurate response. In a dynamic service environment, an agent often takes over an ongoing conversation or needs to respond to a customer who hasn't interacted in weeks. Reading the entire history to understand the context consumes valuable time and delays the response. Papo's AI solves this problem by processing the conversation and presenting key points in a condensed form. | Benefit | Description | | :--- | :--- | | Time Savings | Agents get up to speed in seconds, eliminating the need to manually read extensive histories. | | Better Context | Provides an immediate understanding of the customer's needs, problem history, and level of interest. | | Agile and Accurate Responses | With context in hand, responses are more relevant, personalized, and faster, increasing customer satisfaction. | | Service Consistency | Ensures that all agents, even when taking over a conversation, have the same information base to work with. | How Papo's AI Summarizes Conversation Context Papo's artificial intelligence doesn't just generate a summarized text block; it provides a set of insights that, combined, offer a complete and quick overview of the customer's status. The system reads the entire interaction history to identify signs of interest, urgency, past problems, and business potential. The main tools for this are: 1. Lead Scoring (Score): The AI analyzes the conversation content and assigns a star rating. A high score (e.g., 5 stars) indicates a "hot" lead with high conversion potential, signaling to the agent that this conversation is a priority. 2. Categorization and Tags: The system can identify the main subject of the conversation and automatically apply tags, such as "Quote requested," "Technical issue," or "Follow-up." This allows the agent to know the central theme of the dialogue even before reading the first message. 3. AI Assistant (Capitão): The Capitão module acts as a co-pilot for the agent within the chat. It can provide suggestions, highlight crucial information, and offer a contextual summary so the agent can formulate the best response. Together, these features create an instant customer "dossier," allowing any agent to understand the scenario and act strategically. Step-by-Step to Get Up to Speed Quickly in a Service Interaction Imagine a conversation with dozens of messages has been transferred to you. Instead of spending precious minutes reading everything, you can follow a much more efficient workflow with Papo's AI. | Step | Action | Details | | :--- | :--- | :--- | | 1. Open the Conversation | Upon receiving or taking over a conversation, the first action is to observe the contact information panel. | This panel is usually located next to the chat window and centralizes customer data and AI insights. | | 2. Analyze the Score | Check the star rating assigned to the contact. | A high score means the customer has shown great interest or urgency. A low score might indicate a first contact or a support issue. | | 3. Check the Tags | Look at the tags applied to the conversation, whether automatic or manual. | They inform the main topic, such as "Interest in Gold Plan" or "Billing Query." | | 4. Consult AI Insights | Use the AI assistant (Capitão) features to get a summary of the most recent or important points. | The assistant can highlight the customer's last question or the main obstacle to be resolved. | | 5. Respond with Confidence | With all this context, formulate a precise and personalized response. | You can get straight to the point, demonstrating that you understand the customer's history, which builds trust and speeds up resolution. | Frequently Asked Questions How does AI know what's important in a conversation? AI is trained to recognize patterns, keywords, intent, and sentiment in text. It identifies questions, expressions of urgency ("I need it yesterday"), terms related to purchases ("price," "quote"), and problems ("doesn't work," "error"), using these signals to build the contextual summary and score. Does the AI summary completely replace reading the conversation? For most situations, AI insights are sufficient for a quick and effective response. In very complex or sensitive cases, the summary serves as a guide, directing the agent to the most important parts of the history that might warrant a detailed reading. Does the summary and analysis feature have an additional cost? Yes. Analyses performed by artificial intelligence, such as categorization, scoring, and other advanced functions, consume AI credits, which are part of your Papo plan. Can I train the AI to generate more accurate summaries and scores? Yes. The AI model becomes more accurate as you feed it more context. Actions such as clicking the "Re-evaluate" button after new interactions, defining the lead source (e.g., "Instagram Campaign"), and consistently using tags help train the system on what is most relevant to your business. Does this work for conversations transferred between agents? Yes, this is one of the main advantages. When a conversation is transferred, the new agent instantly receives all the context generated by the AI, eliminating the need for lengthy manual explanations and ensuring a smooth transition for the customer. Besides a summary, what else can AI do to help the agent? Papo's AI goes beyond summarizing. It can suggest pre-written responses, automatically qualify leads, schedule meetings in the agent's calendar from a chat request, and automate the sending of follow-up messages, freeing up the team to focus on strategic tasks. Read also - How to Configure and Train AI Automatic Conversation Categorization - Meet Capitão: How to Configure Your Personalized AI Assistant Start using for free → The first user is free, and you can start organizing your service interactions today.

How to Use Quick Replies to Streamline Customer Service

With Papo (https://papo.global/), you can create and manage Quick Replies, also known as "canned responses," to standardize and accelerate human customer service. This feature allows your agents to answer frequently asked questions in seconds, using keyboard shortcuts to insert pre-approved texts directly into the chat window, ensuring consistency and efficiency. Quick Replies are an essential tool for teams that handle a high volume of interactions and need to maintain quality and speed. Instead of typing the same information repeatedly, agents simply invoke a ready-made response, freeing up time to focus on more complex issues. | Benefit | Description | | :--- | :--- | | Speed in Service | Answer common questions in seconds with a simple shortcut, reducing the average response time (ART). | | Communication Consistency | Ensure that the entire team uses the same information, tone of voice, and correct links, strengthening the brand image. | | Error Reduction | Avoid typos and incorrect information in sensitive data such as prices, links, policies, or technical instructions. | | Simplified Training | Facilitates the onboarding of new agents, who already have access to a library of standard responses to start assisting customers. | | Increased Productivity | Frees agents from repetitive tasks, allowing them to manage more conversations simultaneously and dedicate themselves to cases requiring special attention. | What are Quick Replies? Quick Replies are pre-defined text blocks that agents can quickly insert into a conversation. They are created to answer customers' most common questions (FAQs), such as inquiries about prices, operating hours, exchange policies, or usage instructions. It's important to differentiate Quick Replies from AI automation: - Knowledge Base (AI): Feeds Papo's AI so it can automatically answer customer questions, 24/7, without human intervention. It's ideal for the first level of support. - Quick Replies (Agents): These are tools for human agents. When a conversation is transferred to an agent, they can use these ready-made responses to streamline their own work, maintaining consistency. The two functionalities complement each other. AI can resolve most queries, and when a case requires a human touch, the agent uses Quick Replies to maintain efficiency. How to create and organize your Quick Replies in Papo Response management is centralized, allowing administrators to create a cohesive knowledge base for the entire team. 1. Access the Quick Replies area: This feature is usually found in your Papo account settings. Look for a section called "Quick Replies" or "Canned Responses." 2. Create a new reply: When adding a new reply, you'll need to fill in a few fields: - Shortcut (or Shortcode): A short, easy-to-remember name that will be used to invoke the reply. For example: greeting, gold-plan-price, or exchange-policy. - Message Content: The full text that will be sent to the customer. Here you can include links, formatting, and even variables to personalize the message, such as the customer's or agent's name. 3. Organize into Folders: To keep your library organized, especially if you have many replies, create folders by topic or department. For example, you might have folders like "Sales," "Technical Support," "Finance," and "Appointments." This helps teams quickly find the replies relevant to their roles. How to use Quick Replies in a conversation Once created, using replies is a simple and quick process, designed not to interrupt the flow of conversation. Within the text box where you type the message to the customer, type a slash (/) followed by the shortcut of the reply you want to use. Example: If you type /greeting, the platform will automatically replace the shortcut with the full text: Hello! My name is [Agent Name] and I'm here to help you. How can I assist you today? Most platforms also offer a button or icon in the text box that opens the complete list of replies, allowing you to search by keyword or navigate through folders if you don't remember the shortcut. Examples of Quick Replies by department To make the concept clearer, here are some practical examples that can be applied in different areas of your company: - For Sales: - Shortcut: /gold-price - Message: "Our Gold Plan costs $XXX.XX per month and includes unlimited AI access, support for 5 agents, and advanced reports. You can see all the details here: [link to pricing page]." - For Support: - Shortcut: /reset-password - Message: "To reset your password, please access this link and follow the instructions: [link to password reset]. The process takes less than a minute!" - For Appointments (Clinics, Salons): - Shortcut: /available-times - Message: "Certainly! We have available times this week on Wednesday at 2 PM and Friday at 10 AM. Which one works best for you?" Frequently Asked Questions (FAQ) What is the difference between Quick Replies and the AI Knowledge Base? Quick Replies are text shortcuts for human agents to use during a conversation. The Knowledge Base is a set of documents you provide to train the Artificial Intelligence, allowing it to answer customer questions autonomously. Can I use variables like the customer's name in the reply? Yes, the Quick Replies feature generally allows the use of variables (placeholders) like {{contact.name}} or {{agent.name}} to automatically personalize messages, making communication more personal and humanized. Who can create or edit Quick Replies? Typically, account administrators have permission to create, edit, and organize all Quick Replies. In some configurations, this permission can be extended to other agents or team leaders. Can replies be shared with the entire team? Yes. Quick Replies are created at the account level and are available to all authorized agents. Organizing them into folders helps direct which replies are most relevant to each team (Sales, Support, etc.). Is there a limit to how many Quick Replies I can create? Details about storage limits or the number of replies may vary depending on your plan. For specific information, please consult the plans and pricing documentation on the Papo website. Can I include images or documents in Quick Replies? The primary functionality focuses on text. To send files, agents can use the attachment function within the chat screen itself. Read also - Feeding the AI: How to Create a Knowledge Base for Automatic Replies - Anatomy of the Conversation Screen: Sending messages, notes, and tags Start free on Papo → Your first user is free, and you don't need a credit card to get started.

How to Use Custom Attributes to Segment Automations and Campaigns

To segment automations and campaigns based on specific business data, you can use custom attributes in Papo (https://papo.global/). This feature allows you to create unique data fields for your contacts, such as "Contracted Plan" or "Last Purchase Date," and use this information to create dynamic segments, personalize Flows, and target WhatsApp and email campaigns with high precision. While tags are great for general categorizations, custom attributes take your organization to another level. They allow you to store structured information — such as text, numbers, dates, or list options — which can be used as powerful filters. This enables hyper-segmented communication, ensuring the right message reaches the right person at the ideal time. What are Custom Attributes? Custom attributes are information fields you create in Papo's CRM to store data that is not part of the standard fields (like name, email, and phone number). They are fully adaptable to your company's needs, allowing you to enrich each contact's profile with relevant information for your marketing, sales, and service strategies. Examples of custom attributes include: - For E-commerce: "Last Purchase Date", "Category of Interest", "Total Spent". - For SaaS: "Contracted Plan", "Subscription Start Date", "User ID". - For Education: "Enrolled Course", "Student Status" (e.g., active, inactive), "Completion Date". - For B2B: "Contact Role", "Company Size", "Market Segment". This data transforms your CRM into a source of business intelligence, ready to be activated in your automations. | Characteristic | Tags | Standard Attributes | Custom Attributes | | :--- | :--- | :--- | :--- | | Data Type | Categorical, binary (has or doesn't have) | Text, phone, email (fixed) | Text, Number, Date, Dropdown List | | Purpose | Group contacts, indicate a quick status | Store basic contact information | Store structured and business-specific data | | Practical Example | "Hot Lead", "VIP Customer", "Awaiting Payment" | Contact name, primary email | contracted_plan, last_purchase_date, product_interest | | Use in Segmentation | Simple (e.g., contacts with "Hot Lead" tag) | Simple (e.g., contacts whose name contains "Silva") | Advanced (e.g., last_purchase_date is older than 90 days) | How to Create and Populate Custom Attributes The process involves three main steps: planning, creating in the platform, and populating with data. Step 1: Plan Your Attributes Before you start creating fields, define which information is truly strategic for your operation. Think about how you would like to segment your customers. What data, if you had it, would make your campaigns more effective? Start with a few essential attributes and add others as needed. Step 2: Create Attributes in Papo Attribute creation is done in the platform's settings panel. 1. Access Settings → Custom Attributes. 2. Click to add a new attribute. 3. Give it a descriptive name (e.g., "Customer Plan") and a technical name for internal use (e.g., customer_plan). 4. Choose the field type that best suits the information: - Text: For open-ended information, such as "Job Title" or "City". - Number: For numerical values, such as "Age" or "Order ID". - Date: To record specific dates, such as "Date of Birth" or "Invoice Due Date". - List: To create a list of predefined options (dropdown), such as "Basic", "Pro", "Enterprise" plans. Step 3: Populate Attributes with Data An attribute is only useful if it's filled. There are several ways to add data to a contact's attributes: - Manually: By editing the contact's profile directly in Papo's CRM. - Via Automation Flows: Ask the customer a question in a chatbot and use the answer to automatically fill an attribute. - Via Integrations (API or Webhooks): Send data from your internal system (ERP, e-commerce platform) to Papo whenever an event occurs, such as a new purchase or a registration update. Using Attributes to Segment Automations and Campaigns With your attributes created and populated, you can start using them to refine your actions. 1. Create Dynamic Segments The main application of attributes is in creating Segments. A segment is a saved filter that updates in real-time. 1. Go to the Contacts screen. 2. Use the filter tool to add a new rule. 3. Select the custom attribute you created as a criterion. For example: "Contracted Plan" is equal to "Gold". 4. You can combine multiple filters. Example: "Contracted Plan" is equal to "Gold" AND "Subscription Start Date" is earlier than 30 days ago. 5. After applying the filters, save the view as a new segment, giving it a clear name like "Gold Customers - Older than 30 days". 2. Target WhatsApp and Email Campaigns These saved segments become your target audience for campaigns. - In the WhatsApp Sender: When setting up a new campaign, you can select the "Gold Customers - Older than 30 days" segment as the audience that will receive your message. - In Email Marketing: Similarly, when creating an email campaign, you can choose one or more segments to target your send. 3. Personalize Flows and Sequences Attributes can also be used as triggers or conditions within an automation Flow. For example, you can create a Flow that checks a contact's "Category of Interest" attribute: - If the value is "Shoes", the contact is enrolled in the email Sequence about shoes. - If the value is "Bags", they enter the Sequence about bags. This allows you to create fully personalized and automated customer journeys at scale. Frequently Asked Questions (FAQ) What is the difference between a custom attribute and a tag? A tag works like a sticky note: it's great for quick, binary statuses (the contact either has or doesn't have the "Hot Lead" tag). A custom attribute is like a field in a registration form: it stores specific, structured data (e.g., the exact date of the last purchase, the name of the contracted plan). Attributes allow for much more advanced filters, such as "customers whose last purchase was more than 60 days ago." How is data entered into custom attributes? Data can be entered in four main ways: manually, by editing the contact's profile; automatically, through a Flow that captures a customer's response; via spreadsheet import (when available); or through API or Webhook integrations that synchronize data from other systems with Papo. Can I use attributes to personalize message text? Yes. Just as you use variables for the contact's name (e.g., {{contact.name}}), you can also insert the value of a custom attribute into your automation messages, emails, and WhatsApp templates. This allows you to create messages like: "Hello, {{contact.name}}! We noticed your {{contact.custom_attributes.customer_plan}} plan is about to expire." Consult Papo's documentation for the exact variable syntax. Is there a limit to how many custom attributes I can create? The number of custom attributes you can create may vary depending on your subscription plan. For specific details about the limits applicable to your account, please consult the pricing page or the official platform documentation. Are attribute-based segments updated automatically? Yes. Segments in Papo are dynamic. Once you save a filter (e.g., "contacts with a score greater than 80"), the list of contacts in that segment is updated in real-time. If a contact gains points and exceeds 80, they will be automatically added to the segment. Similarly, if a contact's custom attribute value is changed, their presence in segments will be re-evaluated instantly. Start segmenting your contacts for free → Create your account and start using the integrated CRM to organize your customer base.

What are Capitão's Tools and how to use them to automate tasks?

Capitão's Tools are actions that Papo's (https://papo.global/) artificial intelligence assistant can execute to automate tasks and interact with other systems. They allow the AI to go beyond just answering questions, enabling it to perform practical actions such as scheduling meetings, checking the status of an order in an external system, or creating a new deal in your CRM, all directly from the customer conversation. With Tools, you transform your Capitão assistant from a passive informant into a proactive agent. It can understand the customer's intent and use a specific tool to resolve the request on the spot, without the need for human intervention. This opens up a wide range of possibilities for complex and efficient automations. | Tool Type | Description | Example Usage | Technical Knowledge | | :--- | :--- | :--- | :--- | | Native | Pre-configured and ready-to-use integrations, available on the Papo platform. | Schedule a demo in the salesperson's Google Calendar. | Low (just activate and connect the account). | | Custom (HTTP) | Custom actions that connect to external systems (ERPs, CRMs) or Papo's own API. | Check the status of an order in the company's e-commerce system. | Advanced (requires knowledge of APIs and HTTP requests). | How Capitão's Tools Work? The process of how a Tool works is an intelligent cycle that connects the customer's conversation to a concrete action: 1. Intent Interpretation: The customer sends a message like "I'd like to know where my order 12345 is" or "I want to schedule a meeting with a consultant." 2. Tool Selection: Capitão analyzes the message and, based on the instructions and descriptions of the tools you have configured, identifies which one is most suitable to resolve the request. 3. Action Execution: The AI triggers the selected tool. If it's a tool to check an order, it will send the number "12345" to your external system. If it's for scheduling, it will check for available times in the calendar. 4. Response Return: The external system or native tool returns a result (e.g., "Order in transit" or "Available times: 2 PM and 3 PM"). 5. Communication to the Customer: Capitão uses this information to formulate a natural and complete response for the customer, autonomously finalizing the request. This flow transforms customer service, allowing the AI to solve problems end-to-end, and not just provide basic information. Types of Tools: Native vs. Custom You can equip your Capitão assistant with two types of tools, each with a different purpose. Native Tools These are ready-to-use integrations, developed and maintained by Papo. The goal is to simplify connection with popular services without you having to worry about technical details. The main example is the Google Calendar integration. By activating it, you allow Capitão to: - Check real-time availability in the calendar of one or more team members. - Offer available times to the customer during the conversation. - Automatically create the event in the calendar as soon as the customer confirms the desired time. Configuration usually involves simply authorizing Papo to access your account in the external service and activating the tool for your assistant. Custom (HTTP) Tools Custom tools offer total flexibility, allowing Capitão to communicate with virtually any system that has an API (Application Programming Interface). This includes your ERP, CRM, e-commerce platform, or even other modules within Papo itself. To configure a custom tool, you need to provide the details of the HTTP request that the AI should make: - Endpoint URL: The address of the API to be called (e.g., https://api.mysystem.com/v1/orders). - HTTP Method: The action to be performed (GET to retrieve data, POST to create, PUT to update, etc.). - Headers: Additional information, such as authentication keys (Authorization: Bearer sk_your_secret_key) to ensure the request is secure and authorized. - Body (Request Body): The data to be sent, usually in JSON format. You can use conversation variables to make the action dynamic (e.g., {"order_id": "{variable_order_number}"}). With this, you can create automations such as "check loyalty points balance," "create a lead in Papo's CRM," or "open a support ticket in your internal system." Step-by-Step: How to Configure a Tool in Capitão Although the exact menu names may vary, the conceptual process for enabling a tool for your assistant follows a clear logic. 1. Access Assistant Settings: In the Papo panel, navigate to the configuration area of the Capitão you want to enhance. 2. Locate the Tools Section: Within the assistant's settings, look for the option to manage or add new tools. 3. Add a New Tool: Choose between using a "Native Tool" from the list or creating a "Custom Tool." 4. For Native Tools: - Select the desired integration (e.g., Google Calendar). - Follow the on-screen instructions to connect your account and grant the necessary permissions. - Activate the tool so Capitão can start using it. 5. For Custom (HTTP) Tools: - Give a Name and Description: This is a crucial step. The name should be clear (e.g., "CheckOrderStatus") and the description needs to explain in natural language what the tool does and when it should be used (e.g., "Use this tool to get the status of an order when the customer provides the number"). The AI will use this description to decide when to trigger the action. - Configure the Request: Fill in the technical fields: endpoint URL, HTTP method, authentication headers, and the request body, using variables when necessary. - Save and Activate: After filling in all the data, save the configuration. 6. Test the Tool: Send a message to the assistant in a test channel, simulating a customer's request, to ensure the tool is being triggered correctly and the response is as expected. Frequently Asked Questions (FAQ) What is an API and why do I need it for custom tools? An API (Application Programming Interface) is a set of rules and definitions that allows different software systems to communicate. For custom tools, your system's API (ERP, CRM, etc.) is the "bridge" that allows Capitão to send and receive information, such as checking an order or creating a customer. Do I need to know how to code to use Capitão's Tools? For native tools, no. Configuration is simple and guided, involving clicks and account authorizations. For custom (HTTP) tools, yes, basic knowledge of how APIs and HTTP requests work is required. This is a more advanced feature, aimed at developers or teams with technical support. Can Capitão use tools to perform actions within Papo itself? Yes. You can use custom tools to interact with Papo's own API. For example, you can create a tool that, upon identifying a new lead in the conversation, makes a call to Papo's API and creates a new contact in the CRM or adds a specific tag to it. How does the AI know which tool to use in each situation? The AI decides based on two main factors: the intent detected in the user's message and the description you provide when configuring the tool. A clear and precise description, such as "Use this tool to schedule a technical visit," is fundamental for Capitão to understand the context and choose the correct action. Are custom tools secure? The security of connections via custom tools depends on correct implementation. Papo stores your API keys securely, but it is your responsibility to ensure that you are connecting to secure endpoints (HTTPS) and using API keys that only have the necessary permissions to perform the action, minimizing risks. Read Also - Meet Capitão: How to set up your personalized AI assistant - What is Papo's API and what is it for? Start for free and automate your tasks → Centralize your customer service and start using AI to respond to customers in seconds.

Flows, Rules, or Sequences: Which Automation to Use in Each Situation?

To automate marketing, sales, and support processes, Papo (https://papo.global/) offers powerful tools like Flows and Sequences, which are governed by Rules. Use Flows to create interactive and complex journeys, such as lead qualification chatbots; use Sequences to send linear message cadences, like follow-ups; and use Rules as the conditional ("if-then") logic that defines the behavior of your automations. Understanding the function of each tool is essential for building an efficient automation system. While Flows are flexible and ideal for dynamic conversations, Sequences are optimized for scheduled communication over time. Rules, in turn, are the brain behind everything, enabling personalization and segmentation. | Tool | Main Purpose | Structure | Practical Use Case | | :--- | :--- | :--- | :--- | | Flows | Create interactive journeys and multi-step processes. | Visual and non-linear, with message nodes, actions, and conditions (branches). | A chatbot that asks questions to qualify a lead and, based on the answers, directs them to the sales or support team. | | Sequences | Send a series of predefined messages over time. | Linear and time-based (e.g., send message after 1h, 2 days, 1 week). | A cadence of 5 follow-up messages sent to a lead who requested a quote, with 2-day intervals between each message. | | Rules | Define the conditional logic that triggers and directs automations. | "If-then" logic applied within Flows, Segments, and triggers. | Within a Flow: IF the contact has the tag "VIP Customer", THEN offer a special discount. | What are Flows? Flows are Papo's most versatile automation tool. Through a visual editor, you build a complete journey for your contact, step by step. Each step is a "node" that can be a message, a question, an action (such as adding a tag or updating an attribute), or a condition. Think of Flows as a conversational flowchart. They are ideal for situations that require interaction and decision-making based on user responses. Use Flows for: - Lead Qualification: Ask questions to understand the contact's profile and needs before passing them to a salesperson. - Support Chatbots: Create a self-service menu to answer common questions. - Onboarding Processes: Guide new customers through the first steps of using your product or service. - Satisfaction Surveys (CSAT/NPS): Send a survey and take different actions based on the score received. - Integration with other systems: Use the Webhook action to send data to your CRM, ERP, or spreadsheets in real time. To speed up your work, you can use the Flow Library, which contains ready-made templates for the most common use cases. What are Sequences? Sequences are a specific type of automation, optimized for sending a series of messages (either via WhatsApp or email) at predetermined time intervals. They are the perfect tool for creating communication cadences. Unlike a Flow, which can have multiple paths, a Sequence is linear: Message 1 today, Message 2 in 3 days, Message 3 in 1 week, and so on. The goal is to nurture the contact or maintain engagement over time without the need for complex interaction. Use Sequences for: - Sales Follow-up: Send automatic reminders to a lead who hasn't responded to a proposal. - Nurturing Campaigns: Deliver a series of educational content to a new lead. - Customer Reactivation: Send a series of offers to customers who haven't purchased in a while. - Simple Welcome Series: Send a series of tips in the first few days after a new user signs up. What are Rules? Rules are not a separate tool, but rather the logical engine that powers Flows, Sequences, and segmentation in Papo. They are the "if-then" conditions you define for the system to make automatic decisions. You use rules in several places on the platform: - Within a Flow: To create branches. For example: "IF the contact's response contains the word 'price', THEN execute the action of sending the price list link." - To trigger an automation: "IF a contact receives the tag 'Qualified Lead', THEN start the demo scheduling Flow." - When creating Segments: To filter contacts based on their data. For example, create a dynamic segment with the rule: "Contacts whose custom attribute 'Contracted Plan' is equal to 'Pro'." Rules are what make your automation intelligent and personalized, ensuring the right action is executed for the right contact, at the right time. Frequently Asked Questions (FAQ) Can I use a Flow for follow-up? Yes, you can build a Flow to send messages with time delays. However, for a simple and linear follow-up cadence, a Sequence is the most direct and easy-to-configure tool, as it was specifically designed for this purpose. Can a Sequence have conditions or branches? A Sequence itself is linear (one message after another). Conditional logic (the "Rule") is applied either in the trigger that starts the Sequence (e.g., "enroll in the Sequence only if the contact has tag X") or in the stop condition (e.g., "remove from Sequence if the contact replies"). For complex logic during the journey, with multiple paths, a Flow is the appropriate tool. Where do I configure "Rules"? You won't find a module called "Rules." Instead, you configure them in the context where they are applied: within the Flow editor when adding a condition node, in the filters on the Contacts screen to create Segments, or when defining the triggers for an automation. What's the difference between a Library Flow and one I create from scratch? A Library Flow is a pre-built template for a common task (like a CSAT survey). It saves you time because the main structure is already ready. After installing a Flow from the library, it becomes a copy in your account, fully editable for you to customize messages and actions according to your needs. Can I integrate a Flow with my company's CRM? Yes. Papo's Flows have an action called "Send Webhook." With it, you can send contact data (such as name, phone, and responses) to any external system that accepts webhooks, such as your CRM, spreadsheets, or tools like Zapier and n8n, at the exact moment the contact reaches that stage in the flow. Read also - How to use the Flow Library to automate common tasks - How to Use Custom Attributes to Segment Automations and Campaigns Start automating for free → Create your account and start using automation features to qualify leads and optimize your support today.

What are Automation Rules and how to use them?

Automation Rules in Papo (https://papo.global/) are a set of "if-then" logical instructions that allow you to automate repetitive tasks. With them, you can configure the system to automatically perform actions — such as adding a tag, assigning a conversation to a team, or initiating a Flow — whenever a specific trigger and conditions are met, optimizing the organization and efficiency of your customer service. Rules are the engine behind a smart operation, eliminating the need for manual actions to organize, categorize, and direct conversations. Instead of an agent having to read each new message to decide what to do, rules perform this initial triage in seconds, ensuring nothing is missed and every customer receives appropriate attention. | Component | Description | Practical Example | | :--- | :--- | :--- | | Trigger (If) | The event that initiates the rule's check, acting as the "start". | A new message is received in an inbox. | | Condition (And if) | The specific criterion that the conversation, contact, or message must meet. | The message contains the word "budget" OR "price". | | Action (Then) | The task that the platform automatically performs when the conditions are met. | Assign the conversation to the "Sales" team and add the "Hot Lead" tag. | How Do Automation Rules Work? Automation Rules operate based on a simple yet powerful conditional logic, composed of three main elements: Triggers, Conditions, and Actions. 1. Trigger (The initial event): This is the event that fires the automation. The platform constantly monitors these events. Common trigger examples include: - A new conversation is created. - A conversation is closed. - A message is received. - A tag is added to a contact. 2. Condition (The logical filter): Once the trigger occurs, the system checks if one or more conditions are true. Conditions are the filters that ensure the action will only be executed in the correct scenario. You can create conditions based on: - Message content: If the message contains, does not contain, or is equal to certain keywords (e.g., "support", "cancellation", "invoice"). - Origin channel: If the conversation came from WhatsApp, Instagram, Email, etc. - Contact attributes: If the contact has a specific tag (e.g., "VIP Client") or if a custom attribute has a certain value. 3. Action (The executed task): If the conditions are met, the platform performs the action you defined. Actions are the tasks you want to automate. Examples of actions include: - Add or remove a tag: To categorize the conversation (e.g., "Financial Query"). - Assign the conversation: To direct customer service to a specific agent or team (e.g., "Technical Support Team"). - Send a message: To provide a quick response or initiate a conversation Flow. - Initiate a Flow: To trigger a more complex automation, such as a qualification chatbot. Step-by-Step: How to Create an Automation Rule While the exact location may vary, Automation Rules are generally configured within sections like "Flows" or in the general automation settings of your Papo account. The conceptual process is as follows: 1. Define the objective of your automation: Start with a clear need. For example: "I want all technical support requests received via WhatsApp to be automatically directed to the Support team." 2. Choose the appropriate trigger: Identify the event that should start the process. For the example above, the trigger would be "New conversation created." 3. Establish the conditions: Determine the criteria that must be met. In this case, there would be two combined conditions: - Condition 1: The conversation channel is "WhatsApp." - Condition 2: The first message of the conversation contains the words "support", "problem", or "help." 4. Configure the action to be executed: Define what should happen if the conditions are true. For our example, the action would be "Assign the conversation to the 'Support' team." You could add a second action, such as "Add the 'Technical Support' tag." 5. Save and activate the rule: After configuring all steps, save your automation. It's good practice to send a test message to ensure the rule is working as expected before letting it run in your operation. Frequently Asked Questions (FAQ) What is the difference between a Rule and a Flow? A Rule is a simple automation based on "if-then" logic to perform one or more immediate actions (e.g., if the message contains "price", add the "Sales" tag). A Flow is an interactive and complex journey, with multiple steps, branches, and user interactions (e.g., a complete qualification chatbot). Often, a Rule is used to trigger the start of a Flow. Can I create a Rule that sends an automatic message? Yes, one of the available actions in an Automation Rule is "Send message." You can use it to send automatic responses based on keywords. Remember that for channels like WhatsApp, sending proactive messages outside the 24-hour window may require the use of approved message templates. How many conditions and actions can I have in a single Rule? The Papo platform allows you to combine multiple conditions (using "AND" / "OR" logic) and execute multiple actions in a single rule. This offers great flexibility to create precise automations. For specific limits, please consult the rule configuration interface on the platform. Do Rules work for all channels (WhatsApp, Instagram, etc.)? Yes. Since Papo unifies all your customer service channels into a single inbox, Automation Rules can be configured to work based on triggers and conditions from any connected channel, such as WhatsApp, Instagram, Facebook Messenger, Email, and others. Where can I find Rules on the Papo platform? Automation Rules are not an isolated module but rather a fundamental component integrated with other functionalities. You will generally find them within the Flows builder (to create branching logic) or in a general account automation settings section, where you can create global rules for conversation triage. Read Also - Flows, Rules, or Sequences: Which automation to use in each situation? - Creating your first Automation Flow from scratch Start for free and automate your tasks → Create your account to centralize channels and use automations to respond to leads in seconds.

What are Macros and How to Use Them to Streamline Repetitive Actions

With Papo (https://papo.global/), you can create and use Macros to execute a set of repetitive actions with just one click, streamlining your agents' manual work. A Macro acts as a shortcut that can, for example, add a tag, assign the conversation to another team, and close it, all at once, ensuring consistency and efficiency in your customer service processes. Macros are a powerful tool for standardizing workflows and saving time on tasks that repeat throughout the day. Instead of performing three or four clicks to organize a conversation, your agent triggers a single Macro, and the platform executes the predefined sequence of actions. To better understand the role of Macros, it's useful to compare them with other automation tools available in Papo. | Tool | What it does | When to use | | :--- | :--- | :--- | | Macros | Executes a set of manual actions with one click (tag, assign, close). | To apply a standard workflow to a specific conversation during human interaction. | | Quick Replies | Inserts a predefined text block into the conversation. | To quickly answer frequently asked questions without typing the same text multiple times. | | Automation Rules | Executes actions automatically based on triggers (e.g., new message). | To automate processes that do not require manual intervention, such as assigning a new lead to a team. | What are Macros? Macros are "recipes" of actions that can be applied to one or more conversations with a single command. They are designed to automate manual and repetitive tasks that agents perform within the inbox. While Quick Replies focus on accelerating communication (by inserting text), Macros focus on accelerating management (by executing actions within the conversation). The most common actions that can be included in a Macro are: - Add or remove tags (labels): To categorize the conversation (e.g., "Hot Lead", "L2 Support", "Feedback"). - Assign the conversation: To forward the interaction to a specific agent or team (e.g., "Sales", "Finance"). - Change the conversation status: To mark a conversation as "Pending" or "Resolved" (closed). - Add an internal note: To leave a comment for other team members. By combining these actions, you create fast and error-proof workflows for common day-to-day situations. How to create a Macro in Papo Macro creation is centralized, allowing administrators to configure shortcuts for the entire team. 1. Access Settings: Navigate to your Papo account's settings panel. 2. Find the Automations section: Look for the "Macros" option or a similar area within "Automations". 3. Create a new Macro: Click the button to add a new Macro. You will need to define: - Macro Name: A clear and descriptive name that helps agents understand its function. For example: "Qualify and Send to Sales" or "Close and Mark as Resolved". - Actions: Select from the list of available actions what the Macro should do. You can add multiple actions that will be executed in sequence. For example, for the "Qualify and Send to Sales" Macro, the actions could be: 1. Add tag: "Qualified Lead" 2. Assign to team: "Sales" 4. Save the Macro: After configuring the actions, save the Macro so it becomes available to agents. How to apply a Macro to a conversation Once Macros are created, agents can easily apply them during customer service, without leaving the conversation screen. 1. Open the conversation: Select the conversation to which you want to apply the actions. 2. Locate the Macros menu: Within the conversation interface, usually near the tagging or assignment options, there will be a button or menu for "Macros". 3. Select the desired Macro: Click the button, and a list of all available Macros will appear. Choose the one that corresponds to the action you need to execute. 4. Confirm application: The platform will immediately execute all actions configured in the selected Macro. For example, by clicking the "Close and Mark as Resolved" Macro, the conversation will be tagged, and its status will automatically change to "Resolved". This functionality can also be used in bulk. In some platform views, it's possible to select multiple conversations and apply the same Macro to all of them at once, further optimizing the management of a large volume of interactions. Frequently Asked Questions (FAQ) What is the difference between Macros and Quick Replies? The main difference lies in what each does. Quick Replies insert text blocks into the message box to speed up typing. Macros execute a set of management actions within the conversation, such as tagging, assigning to a team, or closing the interaction, all with a single click. Can I create Macros that send messages to the customer? No. Macros are designed to execute internal organization and workflow actions (tagging, assigning, etc.). To send predefined messages, you should use the Quick Replies feature. Who can create and use Macros? Typically, Macro creation and editing are restricted to users with administrator permissions to ensure process standardization. However, applying Macros to conversations is a tool available to all customer service agents. Permissions can be adjusted in the account settings. Are Macros automatic? No, Macros are semi-automatic. They need to be manually triggered by an agent with a click. Although they execute actions automatically after the click, they are not triggered by events like the arrival of a new message. For fully automatic automations, you should use Automation Rules. Can a Macro be edited after it's created? Yes. Administrators can access the Macros section at any time to edit, add, or remove actions from an existing Macro, as well as delete it or create new ones. Can I apply a Macro to multiple conversations at once? Yes, the Papo platform allows for bulk Macro execution. You can select multiple conversations in your inbox and apply a single Macro to all of them simultaneously, which is ideal for organizing interactions in bulk. See also - How to use Quick Replies to streamline customer service - What are Automation Rules and how to use them? Start using for free → The first user is free and no credit card is required.

What are Scenarios and how to use them to train Captain's AI

Scenarios are specific instructions you provide to Captain, Papo's (https://papo.global/) artificial intelligence assistant, so it knows exactly what action to take in predefined situations. Unlike FAQs, which serve to answer questions, Scenarios are triggers for actions, such as transferring the conversation to a human agent (handoff) when the customer expresses this need, ensuring the AI performs the correct task at the right time. Scenarios are an essential component for refining your assistant's behavior, allowing you to define exceptions and action paths for requests that should not be answered with information from the knowledge base, but rather with a concrete action. | Knowledge Type | Main Objective | Usage Example | | :--- | :--- | :--- | | Scenarios | Execute a specific action. | If the customer types "I want to speak to a person," the Scenario triggers a transfer to an agent. | | FAQs | Provide a direct answer to a question. | The customer asks "What are your operating hours?" and the AI responds "From 9 AM to 6 PM." | | Documents | Provide detailed and contextual information. | The customer asks "How does the extended warranty policy work?" and the AI extracts the answer from a PDF manual. | How do Scenarios work? Scenarios function as priority rules for your Captain assistant's brain. When a new customer message arrives, the AI analyzes it to understand the intent. If the message matches the instruction of a Scenario you've configured, the assistant will immediately execute the action associated with it, instead of trying to formulate a response from the knowledge base (Documents or FAQs). This mechanism is fundamental for controlling the AI's behavior at critical moments in the conversation. It ensures that certain requests are handled in a specific way, without room for interpretation. The flow of a Scenario is simple and effective: 1. Intent Identification: The customer sends a message like "I need to speak to an expert" or "I want to cancel my order." 2. Scenario Match: The AI identifies that the customer's intent matches an active Scenario instruction (e.g., a Scenario for "Handoff Request" or "Cancellation Request"). 3. Action Execution: Instead of searching for answers in files, Captain triggers the action defined in the Scenario. The most common action is handoff, which transfers the conversation directly to a human agent or team. 4. Continued Service: The human agent receives the conversation with its full history and can continue the service, resolving the customer's complex or sensitive request. Step-by-step: How to create and configure a Scenario Configuring a Scenario is a straightforward process that allows you to refine your assistant's behavior. Although the exact menu names may vary, the conceptual logic is as follows: 1. Access Captain's Settings: In the Papo panel, navigate to the AI assistant configuration area you wish to train. 2. Find the Scenarios Section: Within the assistant's knowledge base options, locate the area dedicated to Scenarios (sometimes called "Handoffs"). 3. Create a New Scenario: Start creating a new Scenario. The platform will request two main pieces of information: a Title and an Instruction. 4. Define a Title: Give your Scenario a clear and descriptive name, making it easier to identify later. For example: "Handoff to Support Team" or "Escalation for Complaints." 5. Write the Instruction: This is the heart of the Scenario. Describe the situation or keywords that should trigger the action. Be clear and objective. Good examples of instructions are: - When the customer asks to speak to an agent, a human, an expert, or a person. - If the customer mentions the words "complaint," "consumer protection," or "lawsuit." - If the user says they want to cancel their account or service. 6. Define the Action to be Executed: Select which action Captain should perform when the instruction is met. The primary action available for Scenarios is handoff to a human agent. 7. Activate the Scenario: By default, new Scenarios may be created as disabled to allow for testing. After reviewing the configuration, activate it so it begins to function in real-time in conversations. Scenarios vs. Automation Rules: What's the difference? While both automate tasks, Scenarios and Automation Rules operate at different levels within the Papo platform. - Scenarios: These are direct instructions for Captain's AI behavior during a conversation. They tell the AI how to act when interpreting a specific user intent. A Scenario is part of the assistant's "reasoning." - Automation Rules: These are broader automations, based on "if-then" logic, that run at the platform level. They are triggered by events (triggers) such as "new conversation created" or "tag added" and perform actions like assigning the conversation to a team or sending an automatic response, regardless of whether the AI is active or not. In summary, a Scenario guides the AI's decision, while a Rule automates a platform process. Frequently Asked Questions (FAQ) What is the main difference between a Scenario and an FAQ? The fundamental difference lies in the objective. An FAQ (Frequently Asked Question) is used to provide information; its goal is to answer a customer's question. A Scenario is used to execute an action; its goal is to make the AI perform a specific task, such as transferring the conversation, instead of just responding. What do I write in the "Instruction" field of a Scenario? Be as specific as possible about the trigger. Include the exact phrases or keywords that customers typically use in the situation you want to automate. For example, for a handoff, include variations like "speak to an agent," "speak to a person," "I want a human," "expert help." The clearer the instruction, the more precise the Scenario's activation will be. Can I create Scenarios for actions other than transferring the conversation? The primary and most common use of Scenarios is to perform a handoff to a human agent. Depending on the platform's configuration, it may be possible to associate Scenarios with other actions, such as initiating an automation Flow. Consult the platform's documentation to see the complete list of available actions. How many Scenarios can I create? You can create multiple Scenarios to cover various situations that require a specific AI action. There is no rigid predefined limit, allowing you to refine Captain's behavior according to your operation's needs. For details on your plan's limits, consult the information on the platform. Do Scenarios take priority over Document responses? Yes. Scenarios have maximum priority in the AI's decision-making. If a customer's message matches an active Scenario instruction, Captain will immediately execute the defined action, bypassing the search for answers in Documents or FAQs. This ensures that critical actions always happen as planned. Read also - What are Captain's Tools and how to use them to automate tasks? - What are Automation Rules and how to use them? Start using for free → The first user on your account is free, no credit card required.

Understanding the Automation Report: How to measure the efficiency of your Flows

Papo's Automation Report (https://papo.global/) is the analytical tool that measures the performance of your Flows, consolidating key metrics such as the number of contacts who started each flow, the quantity of messages sent, and the completion rate. It allows you to clearly visualize which automations are most efficient, identify bottlenecks in the customer journey, and calculate the return on investment (ROI) of your automation initiatives. Automating processes like lead qualification, scheduling, or answering frequently asked questions saves time and resources, but how do you know if your automations are truly working? The Automation Report transforms your flows' activity into concrete data, answering questions like: "How many people did our welcome chatbot assist this month?" or "What is the success rate of our sales qualification flow?". This article details the report's metrics and how to use them to continuously refine your automation strategy. | Key Metric | What it measures | Why it's important for analysis | | :--- | :--- | :--- | | Contacts | Total number of unique contacts who entered a specific Flow. | Shows the reach and volume of interactions each automation is managing, indicating its relevance. | | Messages Sent | Total number of messages the Flow sent to contacts. | Helps quantify the work performed by automation, which would otherwise be done manually by an agent. | | Completed | Number of contacts who reached a defined final step in the Flow. | Indicates the flow's effectiveness in guiding the user to the final objective (e.g., qualification, scheduling, resolution). | | Completion Rate | Percentage of contacts who completed the Flow (Completed ÷ Contacts). | It is the main indicator of a Flow's efficiency. Low rates can signal problems in the journey. | How to access and interpret the Automation Report To analyze the performance of your flows, navigate to the Reports module within the Papo platform. Like the Conversations and Tags reports, the Automation report offers a focused view on a specific pillar of your operation. Within the report, you will find: 1. Period Filters: Tools to select the time interval you want to analyze (e.g., last 7 days, last month). Use them to compare performance and identify seasonal trends or the impact of changes you've implemented. 2. List of Flows: A detailed table with all your active Flows, each on a separate line. 3. Performance Metrics: For each listed Flow, you will see columns with the main metrics: Contacts, Messages Sent, Completed, and Completion Rate. Interpretation begins by comparing the numbers. A Flow with many contacts but a low completion rate deserves immediate attention. Conversely, a Flow with a high completion rate can serve as a model for future automations. Using insights to optimize your operation and measure ROI The true value of the report lies in transforming numbers into strategic actions. Use the data to make decisions that improve the efficiency of your processes and your customer's experience. 1. Measure the volume of automated work and ROI The "Contacts" metric shows how many customers were served without the need for initial human intervention. The "Messages Sent" column quantifies the volume of communication performed by the automation. Together, these metrics demonstrate how much time and effort your team has saved. To calculate a simplified ROI, estimate the average time an agent would take to perform a Flow's tasks and multiply it by the number of contacts who passed through it. Compare this value (in saved work hours) with the cost of your platform subscription. 2. Identify bottlenecks and abandonment points The Completion Rate is your main thermometer for a Flow's health. A low rate is a strong indication of a bottleneck. This can happen for several reasons: - A question in the flow is confusing or difficult to answer. - The customer cannot find the option they are looking for. - The value offered by the flow is not interesting enough for them to continue. - A technical step is broken (e.g., a button that doesn't work). When you identify a Flow with low performance, review each step of its journey. Simplify questions, offer clearer options, or add an "emergency exit" so the customer can easily speak with a human. 3. Compare the effectiveness of different approaches Do you have two different Flows to qualify leads from distinct campaigns? The report allows you to directly compare which one has the best completion rate. Flow A, with a more direct approach, might convert more than Flow B, which asks more questions. Use this data to conduct A/B tests. Create variations of the same flow, direct traffic to both, and use the report to decide which version is the winner, applying the learnings across all your automations. 4. Optimize resource allocation If a "Level 1 Technical Support" Flow can resolve 80% of cases (high completion rate), it means your support team can focus on the 20% of more complex cases. On the other hand, if a "Demo Scheduling" Flow has a low completion rate, it might be necessary to allocate a human agent to intervene earlier in the process and not lose the sales opportunity. Frequently Asked Questions (FAQ) What is considered a "completed" contact in a Flow? A contact is considered "completed" when they reach a final step in your automation flow. This final step is defined by you when creating the Flow and can be an action such as sending a last message, applying a "Qualified" tag, or transferring the conversation to a human agent. How often are the Automation Report data updated? Like other Papo reports, automation data is updated almost in real-time. This allows you to closely monitor the performance of your Flows and make agile decisions to optimize journeys. Is a low completion rate always a bad sign? Not necessarily. Context is crucial. For example, a rigorous pre-qualification Flow, designed to filter out leads that don't have the ideal profile, will naturally have a low completion rate for the "qualified" path. In this case, success is measured by the quality of the few who complete, not by the quantity. How can I improve the completion rate of my Flows? To improve performance, review the Flow for friction points. Simplify questions, reduce the number of steps, ensure that the options offered are clear and relevant, and always provide an option for the customer to speak with an agent if the automation cannot help. Does the report show the exact path each contact took in the Flow? The Automation Report provides aggregated data on the overall performance of each Flow (how many entered, how many completed, etc.). To analyze the specific journey of a single contact, you can consult their conversation history on the Conversations screen, where all interactions, including Flow messages, are recorded. Where can I find Flows to create or edit? Flows are created and managed in the Flows module of the Papo platform. There you can use a visual builder to design customer journeys, and you can also install ready-made templates from the Flows Library. Start automating for free → With Papo, you can start creating and measuring your automation flows on the same day.

What is the Playbook and how to use it to train the Live Coach?

The Playbook is the sales script you configure in Papo (https://papo.global/) to "teach" the Live Coach, the Artificial Intelligence assistant that acts as a co-pilot for your salespeople. By feeding the Playbook with information about your products, tone of voice, and the best responses to objections, you train the AI to provide contextual and real-time suggestions, ensuring your team follows best practices and increases conversion rates. Think of the Playbook as the "brain" behind the Live Coach's suggestions. It centralizes the strategic knowledge of your commercial operation in a format that artificial intelligence can understand and use to empower your agents during customer conversations. A well-structured Playbook is key to transforming every salesperson into an expert, standardizing communication, and scaling your sales with consistency. Components of a Sales Playbook To effectively train the Live Coach, your Playbook needs to be fed with four main types of information. Each plays a crucial role in ensuring that the AI's suggestions are accurate and useful. | Playbook Component | Main Objective | Configuration Example | | :--- | :--- | :--- | | Product/Service Context | Describe in detail what you sell, including differentiators and prices. | "Our Gold Plan includes unlimited AI access, priority support, and 10 users. It costs R$ 499/month." | | Tone of Voice | Define the communication style the AI should suggest. | Select the "Consultative" option for a more didactic approach, or "Direct" for a more objective communication. | | Objections and Arguments | Prepare the agent to overcome the main sales obstacles. | Objection: "It's expensive." Argument: "I understand your concern. Our plan includes X and Y, which the competitor doesn't offer, generating a higher ROI in 3 months." | | Inboxes | Choose which service channels the Live Coach should operate in. | Activate the Live Coach only for the "WhatsApp Sales" inbox, leaving "General Support" without sales suggestions. | How the Playbook Feeds the Live Coach The relationship between the Playbook and the Live Coach is direct: one is the knowledge base, and the other is the tool that applies this knowledge in real-time. The process works in a continuous cycle: 1. Configuration: You access the Live Coach configuration area in Papo and fill in the Playbook fields with product information, tone of voice, objections, and arguments. 2. Conversation Analysis: When a salesperson is in a chat with a customer, the Live Coach analyzes the customer's messages in real-time to identify intentions, questions, and, most importantly, objections that match what was configured in the Playbook. 3. Suggestion Generation: Upon identifying a match (for example, the customer says "your competitor is cheaper"), the Live Coach searches the Playbook for the best counter-argument you defined for that specific objection. 4. Display to the Agent: The system displays the suggested response privately, only to your salesperson. It does not automatically send anything to the customer. 5. Agent Action: The salesperson has full autonomy to use the suggestion as is, adapt it to the conversation's context, or simply ignore it and proceed with their own approach. This flow ensures that the knowledge of your best salespeople and managers is distributed to the entire team at the exact moment it is needed. Step-by-step: How to create an effective Playbook Creating your first Playbook is a strategic process. Although the exact menu names may vary, the configuration logic on the Papo platform follows these steps: 1. Access AI Settings: In the Papo dashboard, navigate to the automations or AI area and locate the Live Coach settings (also known as Captain). 2. Define Product and Context: In the appropriate field, describe your product or service. Be clear and include important details such as functionalities, differentiators, technical specifications, and pricing information. This is the text the AI will use to answer questions like "What does Plan X include?". 3. Choose the Tone of Voice: Select the communication style that best represents your brand: consultive (consultative), direct (direct), friendly (friendly), or formal. This will guide the AI in formulating suggestions. 4. Map Objections and Arguments: This is the most critical part. List the most common objections your sales team faces (e.g., "I need to think about it," "I don't have time now," "I'll talk to my partner"). For each, write the best counter-argument, one that has proven to work in keeping the conversation active and moving the negotiation forward. 5. Select Inboxes: Choose which channels (inboxes) the Live Coach should operate in. It's good practice to activate it only in sales-focused channels to avoid overwhelming support agents with irrelevant suggestions. 6. Activate and Monitor: Save the settings and activate the Live Coach. Monitor the first conversations to see how suggestions are being generated and collect feedback from your team to continuously refine the Playbook. Frequently Asked Questions What is the difference between the Live Coach Playbook and the Captain's knowledge base? The Playbook is a knowledge base exclusive to the Live Coach, aiming to generate suggestions for the human agent in real-time. The Captain's knowledge base (composed of FAQs, Documents, and Scenarios), on the other hand, serves to train the AI assistant to automatically respond to customers when no agent is available. They have different purposes: one empowers the human, the other automates the response. Does the Live Coach automatically respond to the customer based on the Playbook? No. The Live Coach is a co-pilot. All suggestions generated from the Playbook are displayed privately only to your salesperson. The agent always has the final decision to use, edit, or ignore the suggestion. Do I need to be a programmer to create a Playbook? Absolutely not. Playbook configuration is done through simple text fields in the Papo dashboard. What you need is knowledge of your business, your products, and your sales process, not technical programming knowledge. How often should I update my sales Playbook? It is recommended to review and update your Playbook quarterly or whenever there are significant changes, such as new product launches, price changes, or when the sales team identifies new recurring customer objections. An updated Playbook is a more powerful tool. What happens if an objection is not in the Playbook? If a customer presents an objection that has not been mapped in your Playbook, the Live Coach will not have a specific suggestion to offer. In this case, the salesperson will rely on their own training and experience. These situations are great opportunities to learn and add the new objection and a good argument to the Playbook. Read also - What are Scenarios and how to use them to train the Captain's AI - How to Manage Your Deals in the Kanban Sales Funnel Start using for free → The first user is free and no credit card is required.

How Synapse Smart Search Works in Conversation History

Papo (https://papo.global/) offers smart search through the Synapse module, an AI tool that allows you to search for keywords, phrases, or meanings across your entire customer conversation history. This transforms your chat archive into a strategic database for finding sales opportunities, product feedback, and at-risk customers that would otherwise be lost among thousands of messages. Over time, the volume of customer conversations becomes a rich source of information, but accessing it manually is impossible. Synapse's smart search solves this problem by scanning and interpreting the content of all interactions across channels like WhatsApp, Instagram, and email. Instead of searching for a contact by name, you can search for what was said within conversations, uncovering valuable patterns and insights for your business. Standard Search vs. Smart Search (Synapse) To understand the power of Synapse's search, it's useful to compare it with a traditional CRM search. While one finds the contact, the other finds the information within the conversation. | Search Type | What it finds | Example Use | | :--- | :--- | :--- | | Standard Search (CRM) | Specific contacts by name, phone, or email. | Find "Maria Silva's" registration to view her history. | | Smart Search (Synapse) | Mentions of a term or phrase in any conversation. | Find all customers who mentioned "budget" in the last 6 months. | How Synapse's History Search Works Synapse's smart search is one of the two main functionalities of Papo's intelligence module, alongside the Customer Radar. While the Radar proactively identifies customers who need attention, the search allows you to reactively investigate history based on your own questions. The operation is simple and powerful: 1. Continuous Indexing: Papo's AI analyzes and catalogs the content of all new messages that arrive in your unified inbox. 2. Comprehensive Search: When you type a term into Synapse's search, the system doesn't just look for contacts. It scans the content of all conversations, past and present, to find mentions of that word or phrase. 3. Contextualized Results: Results are presented as a list of conversations where the term was found. Clicking on a result takes you directly to the relevant snippet of the dialogue, allowing you to understand the full context of the mention. This functionality transforms a previously static conversation archive into a dynamic and searchable knowledge base. Practical Applications: Turning Data into Action The ability to search the content of conversations opens up a range of strategic possibilities for different areas of the company. - Sales and Follow-up: Search for terms like "budget," "proposal," or "how much does it cost" to find leads who showed purchase intent in the past and re-engage them with a follow-up. - Competitive Intelligence: Search for a competitor's name to understand what your customers and leads are saying about them. Discover their strengths, weaknesses, and how your offering compares in market perception. - Customer Retention and Success: Look for words indicating dissatisfaction, such as "cancel," "problem," "doesn't work," or "dissatisfied." This allows you to identify customers at risk of churn and proactively address their issues before it's too late. - Product Development and Marketing: Find valuable feedback by searching for phrases like "I wish it had," "suggestion," or "it would be good if." Collect ideas and suggestions directly from the customer's voice to guide improvements in your products and services. Step-by-Step Guide to Performing a Search with Synapse Using smart search is an intuitive process, designed to provide quick answers without the need for technical knowledge. | Step | Action | Details | | :--- | :--- | :--- | | 1. Access Synapse | In the Papo dashboard, navigate to the Synapse intelligence module. | This area centralizes AI tools, including Customer Radar and history search. | | 2. Enter the search term | Use the search field to type the keyword or phrase you want to find. | Be specific to refine results. For example: "delivery problem," "interest in Gold plan." | | 3. Analyze the results | Synapse will display a list of all conversations containing the searched term. | Each result shows a snippet of the conversation and allows you to click to view the full dialogue. | | 4. Act based on insights | Use the information found to take immediate action. | Initiate a follow-up, add a tag to the contact, transfer to a specialist, or compile a feedback report. | Frequently Asked Questions (FAQ) What is the difference between normal search and Synapse search? Normal search, present in most CRMs, is used to find a specific contact by name, phone, or email. Synapse search is deeper: it searches the content of all conversations, allowing you to find mentions of specific terms within the dialogue with any customer. Does Synapse search work across all channels? Yes. Smart search analyzes the conversation history from all channels connected to your Papo inbox, such as WhatsApp, Instagram Direct, Facebook Messenger, email, among others. This ensures a complete and unified view of interactions. Can I search for exact phrases or just keywords? You can search for both isolated keywords ("cancel") and more specific phrases ("I would like to cancel my plan"). Using exact phrases helps refine results and find more precise contexts. Is Synapse the same as Customer Radar? No. Both are features of the Synapse intelligence module, but with different functions. Customer Radar is a proactive tool that uses AI to automatically list customers who need attention. History search is a reactive tool that you use to actively search for specific information. Does the search find information in very old conversations? Yes. Synapse search is designed to scan the entire conversation history stored on the Papo platform, from the most recent to the oldest. This allows for retroactive analysis and finding information from months or years ago. Do I need any technical knowledge to use smart search? No. The interface was designed to be as intuitive as using a search engine like Google. Simply type what you're looking for and analyze the results, without needing to configure complex filters or write code. Read also - How to use AI to summarize long conversations and get up to speed quickly - How to Use the CNPJ Extractor to Prospect Companies on WhatsApp Start for free and turn your conversation history into intelligence → The first user is free and allows you to explore the platform's features.

What are AI Agents and how do they work in Papo?

AI Agents in Papo (https://papo.global/) are advanced virtual assistants that use artificial intelligence to understand, interpret, and respond to conversations autonomously and contextually. Unlike simple automations, they are capable of maintaining dialogues, qualifying leads, scheduling meetings, and providing first-level support, acting as a 24/7 member of your team. While traditional automations follow fixed rules, AI Agents are designed to handle the complexity and unpredictability of human language. They are configured through natural language instructions (prompts) in the Prompt & AI module, allowing you to define their personality, goals, and limits without needing code. AI Agents vs. Automation Rules It's important to distinguish AI Agents from Automation Rules. Both automate tasks, but they operate at different levels of complexity. | Characteristic | AI Agent | Automation Rule | | :--- | :--- | :--- | | Operational Logic | Conversational, based on dialogue intent and context. | Conditional ("if-then"), based on triggers and keywords. | | Complexity | Manages complex dialogues with multiple steps and questions. | Executes simple, immediate actions (e.g., add tag). | | Interaction | Proactive: can ask questions, seek clarification, and guide the customer. | Reactive: responds to a specific event or condition. | | Usage Example | Qualifying a lead by asking about their budget and needs. | If the message contains "price", assign the conversation to the Sales team. | How do AI Agents work in Papo? The operation of AI Agents in Papo is centered on the Prompt & AI module, which serves as the brain of your artificial intelligence operation. The process involves three main components: 1. The Prompt: This is the set of instructions that defines the AI Agent's behavior. In it, you describe who the agent is, its tone of voice, its goals (sell, support, schedule), what it can and cannot do, and provide examples of interactions. A well-written prompt is the key to an effective agent aligned with your brand. 2. The Agent: This is the entity that executes the prompt's instructions. In Papo, there is a "General" agent that serves as the default for the account, but you can create multiple "sub-assistants". Each sub-assistant can have a specific prompt for a different task, product, or channel. For example, one agent for sales qualification on WhatsApp and another for technical support in Instagram comments. 3. Association with Inboxes: An agent, even when configured, only takes action when activated for a specific channel. Within the Prompt & AI module, you associate an agent (General or a sub-assistant) with one or more inboxes (WhatsApp, Instagram, Email, etc.). This gives you total control over where and when the AI should respond. Step-by-step: How to configure an AI Agent Configuring an AI Agent in Papo is an intuitive process, focused on instructions rather than code. 1. Access the AI Module: In the Papo platform's side menu, click on Prompt & AI. If it's not visible, ask your account administrator to check your access permissions. 2. Select or Create an Agent: You can edit the default "General" agent or click to create a new sub-assistant. New agents are created as drafts, allowing you to configure them completely before activating them. 3. Write the Prompt: This is the most crucial step. In the agent's editing area, define the instructions. You can use a free text field to describe the expected behavior in detail or fill in the structured sections that guide the creation of the persona, goals, and restrictions. 4. Activate the Agent on a Channel: Go to the Inboxes tab within the module. You will see a list of all your connected channels. For each one, activate the AI toggle and select from the menu which agent (the General or a sub-assistant) will be responsible for that channel. 5. Monitor and Optimize: Use the Statistics area to track your agent's performance. Analyze successfully sent messages, failures, and ignored interactions to identify opportunities for improving your prompt. Frequently Asked Questions (FAQ) What is the difference between an AI Agent and a Flow? An AI Agent is designed for dynamic, open-ended conversations, interpreting user intent in real-time. A Flow is a structured, visual journey with a more defined path, ideal for step-by-step processes like registration or a quiz. They can work together: an AI Agent can initiate a Flow when it identifies a specific intent. Can an AI Agent transfer the conversation to a human? Yes. One of the most important functions of an AI Agent is knowing when to pass the baton. You can instruct it in the prompt to transfer the conversation to a specific team (like "Sales" or "Support") when a lead is qualified, when it doesn't know the answer, or when the customer explicitly requests to speak with a person. Do I need to know how to code to create an AI Agent? No. The configuration of AI Agents in Papo is done through the Prompt & AI module, where you write instructions in natural language, as if you were giving orders to a team colleague. No programming knowledge is required. Where do I define the AI Agent's personality and tone of voice? The agent's personality, tone of voice (formal, friendly, technical), name, and role are all defined within the prompt. For example, you can instruct: "You are 'Zap', a friendly virtual assistant for Store X. Always respond in a fun way and use emojis." Can I have different AI Agents for sales and support? Yes. Papo allows you to create "sub-assistants". You can configure a sub-assistant with a sales-focused prompt and associate it with your WhatsApp inbox, and another sub-assistant with a technical support prompt to respond to Instagram comments and support emails. Does the AI Agent have access to the customer's history? Yes, the AI Agent operates with the full context of the conversation and the contact's history in Papo. This allows it to provide more personalized answers and avoid asking questions the customer has already answered, providing a smoother experience. Read also - What are Automation Rules and how to use them? - How to transfer a conversation between agents and teams in Papo Start for free and set up your AI → Papo's AI responds in seconds, qualifies leads, and schedules meetings automatically.

How AI Call Summary Works

Papo's call summary feature (https://papo.global/) uses artificial intelligence to automatically transcribe the audio of voice calls, such as those made via WhatsApp or telephony integrations (Wavoip), and then generates a concise summary. This summary highlights the most important points, decisions made, and actions to be taken, allowing your team to optimize analysis time and improve the quality of follow-up. After a voice call is completed and recorded on the platform, Papo's AI springs into action to process the audio. Instead of having to listen to a multi-minute call again, agents and managers can read a structured summary that gets straight to the point. This ensures that no crucial information is lost and speeds up the transition between services and the follow-up of negotiations. The AI Summary Process The transformation of a voice conversation into an intelligent summary occurs in the background, without the need for any manual action from the agent. The process is divided into clear steps, from recording to presenting the summary. | Step | Platform Action | Result for the User | | :--- | :--- | :--- | | 1. Call Recording | The voice call (via WhatsApp, Wavoip, etc.) is recorded, depending on your provider's settings and plan. | The audio file is available in the contact's history, under "Call Report". | | 2. Automatic Transcription | The AI processes the audio recording and converts it into text, creating a complete transcription of the conversation. | A written record of everything said in the call, facilitating keyword searches. | | 3. Analysis and Summarization| The AI model analyzes the transcription to identify main topics, questions, answers, and action items. | The essence of the conversation is extracted, separating relevant content from the rest. | | 4. Summary Generation | The AI generates a short, structured text, presenting key points, next steps, and agreed-upon decisions. | A summary ready for quick reading, attached to the call record in the customer's history. | How AI Call Summary Works in Practice For the user, the experience is fluid and integrated into the daily workflow on Papo. 1. Make or Receive a Call: The agent uses Papo to make or receive a voice call through a configured channel, such as WhatsApp or Wavoip. 2. Call Completion: At the end of the call, it is automatically recorded in the contact's history, visible on the conversation timeline. 3. Background Processing: In the following minutes, Papo's AI processes the call recording. 4. Access the Summary: When accessing the call record in the history, the agent will find not only the details (duration, date, time), but also the complete transcription and, just below or highlighted, the summary generated by the AI. 5. Quick Action: Based on the summary, the agent can quickly create a follow-up task, update the lead status in the CRM Kanban, or send an email with the discussed points, all without leaving the platform. This automation is especially useful when an agent needs to take over a conversation started by a colleague, as it allows them to understand the full context of the last call in seconds. Who Can Benefit from AI Call Summary? This feature adds value to various roles within a sales or support operation. - Sales and Support Agents: Save time by not having to listen to long calls again to remember details or prepare for follow-up. - Team Managers: Can review call quality and negotiation progress much faster, focusing on key points identified by the AI instead of listening to hours of recordings. - Quality Analysts (QA): Use summaries and transcriptions to audit script compliance, service quality, and identify training needs more efficiently. - Collaborative Teams: Facilitate handovers between agents or shifts, ensuring that the next agent has the complete context of the previous interaction without noise. Frequently Asked Questions Where can I find the call summary? The summary and transcription are available directly in the contact's conversation history, attached to the corresponding call event. They appear shortly after the recording is processed by the AI. Does the AI summary work for all calls? The functionality depends on the activation of call recording in your voice provider (such as PapoCall or Wavoip) and your plan on Papo. Calls need to be recorded for the AI to transcribe and summarize them. Is the call transcription 100% accurate? AI transcription technology is highly advanced, but, like any automated system, it may occasionally present inaccuracies, especially with low-quality audio, background noise, or very specific accents. However, the accuracy is generally high enough to provide clear context and generate a useful summary. Can I disable the call summary feature? AI feature settings are usually managed at the account or plan level. For details on the availability and configuration of this feature, consult your account settings or the documentation on the Papo website. In which languages do transcription and summary work? Papo is constantly expanding its AI capabilities. For the most up-to-date list of supported languages for call transcription and summary, check the detailed information about AI features on the official website. Is there an additional cost to use AI call summary? The availability of advanced AI features, such as call summary, may depend on your subscription plan. We recommend consulting the pricing page on the Papo website or contacting support for specific information about your plan. Read Also - How to use AI to summarize long conversations and get up to speed quickly - Reports Overview: Where to find each metric in Papo Start using for free → The first user is free forever, allowing you to experience the platform's features.

How to use AI to suggest the best response during customer service?

With Papo (https://papo.global/), you can use artificial intelligence as a co-pilot for your customer service team, suggesting the best responses in real-time during a conversation. Through modules like Capitão, the platform analyzes the customer's question and recommends relevant answers, either from a list of predefined Quick Replies or by generating text based on your Knowledge Base documents, increasing the speed and consistency of human support. Efficiency in customer service depends on quick and accurate responses. However, even the most experienced agents can take time to find the correct information or type out answers to repetitive questions. Papo's agent-assist AI solves this problem by acting as an assistant that reads the conversation and offers response suggestions with a single click. This allows human agents to focus on customer relationships, while technology handles the repetitive task of finding and formulating information. Types of AI-Generated Suggestions Papo's AI uses two main sources to suggest responses to your agents during customer service. Each has a specific purpose and, together, they create a robust support system. | Suggestion Type | Information Source | Ideal For | | :--- | :--- | :--- | | Quick Replies | Pre-approved texts manually created by your team. | Very frequent and standardized questions, such as greetings, prices, operating hours, useful links, and exchange policies. | | Knowledge Base | Documents (PDFs, product manuals, internal policies) that you upload to the platform. | More complex or specific questions that require consulting detailed information not found in a ready-made response. | Quick Replies ensure standardization and speed for common queries, while the Knowledge Base offers depth and flexibility for unexpected questions. How AI response suggestion works The process is designed to be intuitive and not interrupt the agent's workflow. Papo's Capitão module functions as a co-pilot integrated into the conversation screen. 1. Customer Message Analysis: When a customer sends a question, the AI analyzes the content in real-time to understand the intent behind the message. 2. Search for the Best Response: Based on this analysis, the AI searches on two fronts: - Checks if the question matches any of the configured Quick Replies. - Consults the Knowledge Base to find document excerpts that answer the customer's query. 3. Suggestion Presentation: The platform displays the most relevant suggestions directly in the agent interface, next to the typing box. Suggestions can include both Quick Replies and texts generated from documents. 4. Agent Action: The agent can then: - Click on a suggestion to insert it directly into the conversation. - Edit the suggestion before sending it to personalize the message. - Ignore the suggestions and type a manual response, if none are suitable. This cycle happens in seconds, transforming customer service into a more dynamic and efficient process. How to configure AI to suggest responses For AI to assist your team, you first need to provide it with the necessary knowledge. Configuration is divided into simple steps and does not require advanced technical knowledge. | Step | Action | Description | | :--- | :--- | :--- | | 1. Feed the Knowledge Base | Upload relevant documents (PDFs, text files) to the "Knowledge Base" section of Papo. | Include product manuals, company policies, troubleshooting guides, and frequently asked questions (FAQs) lists. | | 2. Create Quick Replies | Access the "Quick Replies" area and create shortcuts and texts for the most common questions. | Organize them into folders by department (Sales, Support) to facilitate management and use by agents. | | 3. Activate the AI Assistant | In your AI assistant (Capitão) settings, link the Knowledge Bases and Quick Replies it should use. | You can have different assistants for different teams, each with its own set of knowledge. | | 4. Train the Team | Show agents how suggestions will appear on the conversation screen and how they can use them. | Usage is intuitive, but a brief alignment ensures everyone gets the most out of the functionality. | Frequently Asked Questions (FAQ) What is the difference between AI that responds autonomously and AI that suggests responses? At Papo, there are two main applications of AI. Super Chat is an autonomous AI agent that responds to customers 24/7, qualifies leads, and schedules appointments without human intervention. Capitão, on the other hand, is an AI co-pilot that assists the human agent, suggesting responses during the conversation so the agent can work faster. Where does AI get the information to suggest responses? AI retrieves information from two locations that you configure: the Quick Replies library, which contains pre-approved texts for common questions, and the Knowledge Base, composed of documents (such as PDFs and manuals) that you upload to the platform. Can the agent edit an AI-suggested response before sending it? Yes. AI suggestions are a starting point. The agent has full control and can click on the suggestion to insert it into the text box, and then edit, add, or remove information to personalize the message before sending it to the customer. Do I need technical knowledge to configure the Knowledge Base? No. The process involves uploading files your company already has, such as product manuals, PDF return policies, or text documents with FAQs. The platform automatically processes these files to make them searchable by AI. Are Quick Replies the same as what chatbots use? Not exactly. Quick Replies are primarily a tool for human agents, used to streamline manual customer service. The chatbot's AI (Super Chat) primarily uses the Knowledge Base to formulate its own responses autonomously, although both functionalities can be fed by similar information sources. Does AI learn from the responses agents choose? AI is trained to identify the best response based on the content you provide in the Knowledge Base and Quick Replies. The effectiveness of suggestions is directly linked to the quality and organization of this information. The platform does not use agent choices to autonomously retrain the model, ensuring that control over the information remains with you. Read also - How to use Quick Replies to speed up customer service - What are AI Agents and how do they work at Papo? Start using for free → The first user is free, and you can start configuring your customer service operation right away.

How to use the Flows Library to install ready-made automations?

To install ready-made automations and streamline business processes, Papo (https://papo.global/) offers the Flows Library, a collection of templates that can be installed in your account with a few clicks. With this feature, you can quickly implement workflows for lead qualification, satisfaction surveys, and customer onboarding, without having to create everything from scratch. The Flows Library was created so that teams of all skill levels can harness the power of automation, saving time and reducing the complexity of workflow creation. | Step | Action | Description | | :--- | :--- | :--- | | 1. Access | Navigate to the Flows Library | Within the Flows module on the Papo platform, locate the option to access the template library. | | 2. Explore | Find and preview a template | Browse categories (e.g., Sales, Customer Service) and click to preview the complete flow structure. | | 3. Install | Install the flow in your account | With one click, a copy of the template is added to your list of flows, with the status Inactive by default. | | 4. Customize and Activate | Edit and publish the flow | Review messages, adjust actions, and when ready, activate the flow to start using it. | What is the Flows Library? The Flows Library is a collection of pre-configured automation flows provided by Papo. Think of it as an app store for your automations. Each template in the library is designed to solve a specific business need, allowing you to start automating tasks in minutes, rather than hours. Some examples of templates you might find include: - Lead Qualification: A chatbot that asks initial questions to filter and prioritize new contacts before sending them to the sales team. - New Customer Welcome: A sequence of messages to guide new customers through the first steps with your product or service. - Satisfaction Survey (CSAT/NPS): A flow that automatically sends a survey after a service interaction ends and takes actions based on the score received. - Demo Scheduling: A flow that collects lead information and directs them to a salesperson's calendar. The main benefit is agility. Instead of designing a complex flow from scratch, you can install a tested model and focus only on fine-tuning, such as customizing message text or defining which team a task will be sent to. How to install an automation from the Library step-by-step Implementing an automation from a template is a simple and secure process. The flow only becomes active after you review, customize, and manually activate it. 1. Accessing the Library Within the Papo dashboard, navigate to the Flows module, where you manage all your automations. On the Flows screen, look for the option that leads to the Flows Library or a templates section. 2. Exploring and choosing a template In the library, you will find various templates organized by categories or use cases, such as Sales, Marketing, Customer Service, or Human Resources. - Browse options: Explore the available templates to find one that fits your needs. - Preview the flow: Before installing, you can click on a template to see a preview of its visual structure. This shows all the "nodes" (steps) of the flow, such as messages, conditions, and actions, allowing you to understand exactly what that automation does even before adding it to your account. 3. Installing the template Like a template? The next step is to install it. By clicking the install button, a complete copy of that flow will be added to your list of automations in the Flows module. A crucial detail is that, by default, the flow is installed as INACTIVE. This is a security measure to ensure that an automation does not accidentally start running before being specifically configured for your business. 4. Customizing and activating the flow With the flow installed and inactive, you have complete freedom to adjust it to your needs. 1. Open the newly installed flow in your list of automations. 2. Review each node: Read the text messages and adjust the language to reflect your brand's voice and policies. 3. Configure actions: Check actions such as "add a tag to the contact," "create a task in the Projects module," or "send a webhook" and configure them according to your operation's rules. 4. Test the flow: Use the platform's testing tools to simulate the customer journey and ensure everything is working as expected. 5. Activate the flow: Once you are satisfied with the customizations, simply change the flow's status from "Inactive" to "Active." From that moment on, it will be ready to be triggered by the configured triggers and automate your tasks. Who is the Flows Library for? This feature is especially useful for: - Companies new to automation: Offers a clear and functional starting point, reducing the learning curve. - Marketing and Sales Teams: Allows for quick implementation of qualification, nurturing, and scheduling funnels. - Customer Service Teams: Facilitates the creation of flows for satisfaction surveys and frequently asked questions. - Managers and Analysts: Accelerates process standardization and the implementation of new strategies without relying on technical development. Frequently Asked Questions (FAQ) What happens after I install a flow? The flow is copied to your list of automations in the Flows module with the status "Inactive." It will not perform any actions until you review, customize, and manually activate it. Can I edit a flow I installed from the library? Yes. The installed template is entirely yours. You can edit, add, or remove any step (node), change texts, reconfigure actions, and adapt the flow completely to your needs. Do library templates cost money? Access to the Flows Library and the installation of templates are features included in the Papo platform. Costs related to the execution of automations, such as sending messages, follow the rules and limits of your contracted plan. For details on plans and pricing, please consult our official website. What types of automation can I find in the library? The library includes a variety of templates for common use cases in sales, marketing, and customer service, such as lead qualification, meeting scheduling, satisfaction surveys (CSAT and NPS), new customer onboarding, welcome sequences, and much more. Do I need technical knowledge to use the templates? No. The main advantage of templates is that they are designed to be used by anyone, without the need for programming knowledge. Basic customization primarily involves editing text. More advanced adjustments, such as configuring webhooks, may require a bit more familiarity with the platform, but the core of the flow will already be ready. Read also - Flows, Rules, or Sequences: Which automation to use in each situation? - What are Automation Rules and how to use them? Start for free → The first user is free for you to test the complete flow.

What are Capitão's Tools and how to use them?

Capitão's Tools are actions that Papo's (https://papo.global/) artificial intelligence assistant can perform to automate tasks and interact with other systems. They allow the AI to go beyond just answering questions, enabling it to carry out practical actions such as scheduling meetings, checking the status of an order in an external system, or creating a new deal in your CRM, all directly from the customer conversation. With Tools, you transform your Capitão assistant from a passive informant into a proactive agent. It can understand the customer's intent and use a specific tool to resolve the request on the spot, without the need for human intervention. This opens up a range of possibilities for complex and efficient automations. | Tool Type | Description | Example Usage | Technical Knowledge | | :--- | :--- | :--- | :--- | | Native | Pre-configured and ready-to-use integrations, available on the Papo platform. | Schedule a demo in a salesperson's Google Calendar. | Low (just activate and connect the account). | | Custom (HTTP) | Custom actions that connect to external systems (ERPs, CRMs) or Papo's own API. | Check the status of an order in the company's e-commerce system. | Advanced (requires knowledge of APIs and HTTP requests). | How Do Capitão's Tools Work? The process of how a Tool works is an intelligent cycle that connects the customer's conversation to a concrete action: 1. Intent Interpretation: The customer sends a message like "I'd like to know where my order 12345 is" or "I want to schedule a meeting with a consultant." 2. Tool Selection: Capitão analyzes the message and, based on the instructions and descriptions of the tools you've configured, identifies which one is most suitable to resolve the request. 3. Action Execution: The AI triggers the selected tool. If it's a tool to check an order, it will send the number "12345" to your external system. If it's for scheduling, it will check for available times in the calendar. 4. Response Return: The external system or native tool returns a result (e.g., "Order in transit" or "Available times: 2 PM and 3 PM"). 5. Communication to the Customer: Capitão uses this information to formulate a natural and complete response for the customer, autonomously finalizing the request. This flow transforms customer service, allowing the AI to solve problems end-to-end, rather than just providing basic information. Types of Tools: Native vs. Custom You can equip your Capitão assistant with two types of tools, each with a different purpose. Native Tools These are ready-to-use integrations, developed and maintained by Papo. The goal is to simplify connection with popular services without you having to worry about technical details. The main example is the Google Calendar integration. By activating it, you allow Capitão to: - Check real-time availability in the calendar of one or more team members. - Offer available times to the customer during the conversation. - Automatically create the event in the calendar as soon as the customer confirms the desired time. Configuration usually only involves authorizing Papo to access your account in the external service and activating the tool for your assistant. Custom (HTTP) Tools Custom tools offer total flexibility, allowing Capitão to communicate with virtually any system that has an API (Application Programming Interface). This includes your ERP, CRM, e-commerce platform, or even other modules of Papo itself. To configure a custom tool, you need to provide the details of the HTTP request that the AI should make: - Endpoint URL: The address of the API to be called (e.g., https://api.mysystem.com/v1/orders). - HTTP Method: The action to be performed (GET to retrieve data, POST to create, PUT to update, etc.). - Headers: Additional information, such as authentication keys (Authorization: Bearer sk_your_secret_key) to ensure the request is secure and authorized. - Body (Request Body): The data to be sent, usually in JSON format. You can use conversation variables to make the action dynamic (e.g., {"order_id": "{variable_order_number}"}). With this, you can create automations such as "check loyalty points balance", "create a lead in Papo's CRM", or "open a support ticket in your internal system". Step-by-Step: How to Configure a Tool in Capitão Although the exact menu names may vary, the conceptual process for enabling a tool for your assistant follows a clear logic: 1. Access Assistant Settings: In the Papo dashboard, navigate to the configuration area of the Capitão you wish to enhance. 2. Locate the Tools Section: Within the assistant's configuration options, find the area dedicated to "Tools." 3. Add a New Tool: Choose between adding a "Native" or "Custom (HTTP)" tool. 4. For Native Tools: Select the desired service (such as Google Calendar) from the list of available integrations. Follow the on-screen instructions to connect and authorize your account. 5. For Custom Tools: - Give a Name and Description: Create a clear name (e.g., "Check Order Status") and a detailed description for the AI. The description is crucial, as it teaches Capitão when and how to use the tool (e.g., "Use this tool to check a customer's order status when they provide the order number"). - Configure the HTTP Request: Fill in the fields for the endpoint URL, HTTP method, headers (for authentication), and the request body, if necessary. 6. Activate and Test: After configuration, activate the tool so Capitão can start using it. Perform tests by sending messages that should trigger the new tool to ensure everything is working as expected. Frequently Asked Questions (FAQ) What is the difference between Tools and Scenarios? Tools are designed to perform actions in external systems or other parts of Papo's API, such as querying a database or scheduling an event. Scenarios, on the other hand, are instructions for Capitão to perform an internal action within the platform, such as transferring the conversation to a human agent (handoff) in specific situations. Do I need to know how to code to use the Tools? For Native Tools, such as the Google Calendar integration, no programming knowledge is required. Configuration is done through a simple interface. For Custom (HTTP) Tools, technical knowledge of how APIs work, including HTTP methods, authentication, and JSON format, is necessary. How does the AI know which Tool to use? The AI decides which tool to use based on the description you provide when configuring it. A clear and detailed description, explaining the tool's purpose and what information it needs, is crucial. Capitão analyzes the customer's intent and compares it with the descriptions of all available tools to choose the most appropriate one. Can I use the Tools to interact with Papo's own API? Yes. You can configure a Custom (HTTP) Tool to make calls to Papo's API. This allows you to create advanced automations, such as using AI to create a new contact in the CRM, send a message through another channel, or update a customer attribute, all from an instruction in the conversation. Is it safe to use Custom Tools with my API keys? Yes. Authentication information, such as API keys (tokens), is securely stored on the Papo platform and used in server-to-server communications, which are more secure. It is good security practice to create API keys with limited permissions specifically for this integration, rather than using keys with full access. Read also - What are Scenarios and how to use them to train Capitão's AI - How to generate and use API keys to integrate with other systems Start for free and automate your tasks → The first user is free, and you can start configuring your AI to perform real actions right away.

How to Create an Automation that Sends a Message After a Specific Time?

To create an automation that sends a message after a specific time, you should use the "Wait" (Delay) step within a Flow on the Papo platform (https://papo.global/). This feature allows you to define a strategic pause of minutes, hours, or days before the automation proceeds to the next action, such as sending a follow-up message, a reminder, or a satisfaction survey. The ability to pause an automation is crucial for creating communication journeys that feel natural and are delivered at the most opportune moment for the customer, without overwhelming them with instant messages. | Component | Main Function | Time Units | Practical Use Case | | :--- | :--- | :--- | :--- | | "Wait" Step | Pause the execution of a Flow for a determined period. | Minutes, Hours, Days. | After sending a proposal, wait 2 days and then send a follow-up message asking if the client has any questions. | | Automation Flow | Orchestrate the complete contact journey with actions and conditions. | N/A | A chatbot that qualifies a lead, sends material, waits 24 hours, and then asks what they thought of the material. | | "Send Message" Action | Send a message (text or HSM template) to the contact. | N/A | The message that is triggered immediately after the "Wait" period ends. | What is the "Wait" step and what is it for? The "Wait" step is one of the action "nodes" available in Papo's visual Flow builder. Its sole function is to introduce a temporal delay in the automation sequence. Instead of one action happening immediately after another, the Flow remains in a waiting state for the period you configure. This feature is essential for building effective communication cadences, such as: - Sales follow-up: Maintaining contact with a lead who has shown interest but hasn't yet made a decision. - Lead nurturing: Sending a series of educational content or tips about your product/service over several days. - Reminders: Notifying customers about appointments, invoice due dates, or important events in advance. - Post-interaction surveys: Collecting feedback some time after a service interaction is completed or a product is delivered, allowing the customer time to have a complete experience. By combining the "Wait" step with other action and condition nodes, you transform a simple chatbot into a powerful relationship and sales tool. How to configure the "Wait" step in a Flow Creating a pause in your automation is a simple and visual process. Follow the steps below to add a waiting time to your Flow. 1. Access the Flows Module: In the main menu of the Papo platform, navigate to Flows / Chatbot. 2. Create or Edit a Flow: You can start a new Flow from scratch or add the waiting step to an existing automation. 3. Add a New Node: At the point in the journey where you want to insert the pause (for example, after sending an initial message), click the + icon to add a new step. 4. Select the "Wait" Action: In the list of available actions, find and select the "Wait" (or "Delay") option. 5. Define the Waiting Time: A configuration window will open. In it, you must define the duration of the pause: - Value: Enter the corresponding number for the time (e.g., 24). - Unit: Choose the unit of time from the dropdown menu (Minutes, Hours, or Days). 6. Connect the Next Action: After the "Wait" node, add the action that should be executed at the end of the waiting period. Generally, this will be a "Send Message" node. 7. Configure the Follow-up Message: In the "Send Message" node, write the text the contact will receive. Remember that if the pause exceeds WhatsApp's 24-hour customer service window, the message sent must be a Message Template (HSM) previously approved by Meta. 8. Save and Activate the Flow: After building and reviewing your logic, save the changes and ensure that the Flow is set to "Active" so that it can process contacts. With these steps, any contact passing through this part of your Flow will be "paused" at the "Wait" node and will only proceed with the automation once the configured time has elapsed. Frequently Asked Questions (FAQ) What is the difference between using "Wait" in a Flow and a Sequence? The "Wait" step within a Flow offers maximum flexibility, as it allows you to create branches and conditional logic after the wait. For example, you can wait 2 days and then check if the contact has a specific tag before deciding which message to send. A Sequence, on the other hand, is a simpler, more linear tool designed specifically to send a series of messages at fixed intervals, without the complexity of interactions or conditions. What happens if the contact replies during the waiting period? By default, the "Wait" step only pauses the execution of that specific Flow path. If a contact sends a message during this period, the conversation will be opened in the inbox for a human agent or may trigger other rule-based automations, but the original Flow will continue its countdown. The automation is not interrupted unless you configure a specific rule to remove the contact from the Flow in case of a reply. Can I use the "Wait" step to send messages outside of WhatsApp's 24-hour window? Yes, it is one of the main uses for this feature. However, it is crucial to understand that to initiate a conversation with a customer on WhatsApp after the 24-hour window has closed, you must use a Message Template (HSM) approved by Meta. You cannot send a free-form text message after the wait if the window is already closed. Does the waiting time only consider business days? No. The "Wait" step counts time continuously. For example, if you configure a "2-day" wait on a Friday, the next action will be executed on Sunday, at the same time. The system does not differentiate between business days, weekends, or holidays in the default configuration. Is it possible to define a different waiting time for each contact? In the default configuration of the "Wait" node, you define a fixed value (e.g., 24 hours) that applies to all contacts passing through that step. For more advanced scenarios requiring dynamic waits, it may be necessary to build more complex logic using rules and custom attributes. What is the maximum time I can configure in the "Wait" step? Details about specific time limits can be consulted directly in the Flow configuration interface on the Papo platform. Generally, the limits are quite broad to cover most lead nurturing and follow-up use cases. Start automating your conversations for free → Create your Papo account and start using Flows to build smart journeys for your customers.

How to Create an Automation that Sends a Message After a Specific Time?

To create an automation that sends a message after a specific time, you must use the "Wait" (Delay) step within a Flow on the Papo platform (https://papo.global/). This feature allows you to define a strategic pause of minutes, hours, or days before the automation proceeds to the next action, such as sending a follow-up message, a reminder, or a satisfaction survey. The ability to pause an automation is crucial for creating communication journeys that feel natural and are delivered at the most opportune moment for the customer, without overwhelming them with instant messages. | Component | Main Function | Time Units | Practical Use Case | | :--- | :--- | :--- | :--- | | "Wait" Step | Pause the execution of a Flow for a determined period. | Minutes, Hours, Days. | After sending a proposal, wait 2 days and then send a follow-up message asking if the client has any questions. | | Automation Flow | Orchestrate the complete contact journey with actions and conditions. | N/A | A chatbot that qualifies a lead, sends material, waits 24 hours, and then asks what they thought of the material. | | "Send Message" Action | Send a message (text or HSM template) to the contact. | N/A | The message that is triggered immediately after the "Wait" period ends. | What is the "Wait" step and what is it for? The "Wait" step is one of the action "nodes" available in Papo's visual Flow builder. Its sole function is to introduce a temporal delay in the automation sequence. Instead of one action happening immediately after another, the Flow remains in a waiting state for the period you configure. This feature is essential for building effective communication cadences, such as: - Sales follow-up: Maintaining contact with a lead who showed interest but hasn't made a decision yet. - Lead nurturing: Sending a series of educational content or tips about your product/service over several days. - Reminders: Notifying customers about appointments, invoice due dates, or important events in advance. - Post-interaction surveys: Collecting feedback some time after a service interaction was completed or a product was delivered, giving the customer time to have a complete experience. By combining the "Wait" step with other action and condition nodes, you transform a simple chatbot into a powerful relationship and sales tool. How to configure the "Wait" step in a Flow Creating a pause in your automation is a simple and visual process. Follow the steps below to add a waiting time to your Flow. 1. Access the Flows Module: In the main menu of the Papo platform, navigate to Flows / Chatbot. 2. Create or Edit a Flow: You can start a new Flow from scratch or add the wait step to an existing automation. 3. Add a New Node: At the point in the journey where you want to insert the pause (for example, after sending an initial message), click the + icon to add a new step. 4. Select the "Wait" Action: In the list of available actions, find and select the "Wait" (Delay) option. 5. Define the Waiting Time: A configuration window will open. In it, you must define the duration of the pause: - Value: Enter the number corresponding to the time (e.g., 2). - Unit: Choose the unit of time from the dropdown menu (Minutes, Hours, or Days). 6. Connect the Next Action: After the "Wait" node, add the action that should be executed at the end of the wait. Generally, this will be a "Send Message" node. 7. Configure the Follow-up Message: In the "Send Message" node, write the text that the contact will receive. Remember that if the pause exceeds the 24-hour WhatsApp customer service window, the message sent must be a Message Template (HSM) previously approved by Meta. 8. Save and Activate the Flow: After building and reviewing your logic, save the changes and ensure that the Flow is set to "Active" status so that it can process contacts. With these steps, any contact passing through this part of your Flow will be "paused" at the "Wait" node and will only proceed with the automation when the configured time is completed. Frequently Asked Questions (FAQ) ### What is the difference between using "Wait" in a Flow and a Sequence? The "Wait" step within a Flow offers maximum flexibility, as it allows you to create branches and conditional logic after the wait. For example, you can wait 2 days and then check if the contact has a specific tag before deciding which message to send. A Sequence, on the other hand, is a simpler, more linear tool, designed specifically to send a series of messages at fixed intervals, without the complexity of interactions or conditions. ### What happens if the contact replies during the waiting period? By default, the Flow will continue its normal execution after the waiting period ends, even if the contact sends a message during that time. The automation will follow the path that was designed. For more advanced scenarios, you can use Rules to create logic that interrupts the wait or alters the flow if the contact interacts. ### What happens if I need to send a message after 24 hours on WhatsApp? If the waiting time configured in the "Wait" step causes the next message to be sent after the 24-hour WhatsApp customer service window, that message must be a Message Template (HSM) approved by Meta. Common text messages will be blocked. Make sure to create and approve your follow-up templates in advance. ### Is there a maximum limit for the waiting time? The platform allows you to configure the waiting time in minutes, hours, or days, covering the vast majority of use cases for marketing and sales automations. For information on specific limits, consult the detailed documentation on the platform. ### Can I use variables in the message sent after the wait? Yes. The message sent after the "Wait" step can be fully customized with contact attributes, such as name, company, or any other information stored in Papo's CRM. This makes your follow-up much more personal and effective. ### Is it possible to cancel the wait for a specific contact in a Flow? Managing individual contacts within a running Flow is an advanced feature. Typically, a contact can be manually removed from a Flow, which would interrupt any future waiting steps. Start for free and automate your follow-ups → Start with the first user for free and build automations with no upfront cost.

How to Use AI to Identify Customer Sentiment in a Conversation?

Papo (https://papo.global/) uses artificial intelligence to analyze conversation messages in real-time and classify customer sentiment as positive, negative, or neutral. This analysis allows support and sales teams to proactively identify dissatisfied customers, business opportunities, and prioritize the most critical interactions through tools like Synapse's Customer Radar and automation rules. Understanding the emotion behind a customer's message is crucial for providing quality service. A request for "help" might have a neutral tone, while "I need help NOW, nothing works" carries a clearly negative sentiment. Papo's AI has been trained to capture these nuances, transforming the conversation's tone into actionable data for your operation. | Sentiment | What it means | Strategic Use Example | | :--- | :--- | :--- | | Positive | Satisfied customer, praising the product, agreeing to a solution, or showing enthusiasm. | Identify brand advocates, request a testimonial, or direct to an upgrade (upsell) offer. | | Negative | Frustrated customer, complaining about a problem, mentioning competitors, or expressing disappointment. | Prioritize service, apply an "Urgent" tag, notify a supervisor, or trigger a workflow to prevent cancellation (churn). | | Neutral | Customer asking an objective question, requesting information, confirming data, or in a purely transactional interaction. | Follow the standard service flow, collect information, or direct to a quick response or Help Center article. | How Papo's AI Analyzes Sentiment Papo's artificial intelligence is not based solely on isolated keywords. The system has been trained with millions of conversation examples to understand the context, intent, and linguistic patterns that indicate the customer's emotional state. The analysis considers multiple factors: - Word Choice: The presence of strong adjectives like "incredible," "perfect," and "excellent" signals positivity, while terms like "problem," "disappointed," "error," and "doesn't work" indicate negative sentiment. - Sentence Context: The AI evaluates the sentence as a whole. "I didn't have any problems" is a positive phrase, despite containing the word "problem." - Intensity: The use of capital letters, word repetition ("never, never again"), and excessive punctuation (!!!, ???) are interpreted as intensifiers of the expressed sentiment. This analysis directly feeds into intelligence modules like Synapse. In it, the Customer Radar uses sentiment as one of the main factors to calculate a contact's "criticality." A conversation with persistent negative sentiment will have higher criticality, signaling to the team that the customer needs immediate attention to prevent churn. How to Use Sentiment Analysis in Automations The true power of sentiment analysis lies in using it to automate actions and optimize processes. At Papo, this is primarily done through Automation Rules. You can configure rules that are triggered based on the sentiment detected in a message. Practical examples of sentiment-based automation: 1. Prioritizing Dissatisfied Customers: - Trigger: When a message is received. - Condition: If the message sentiment is Negative. - Action: Add the tag "At-Risk Customer" and assign the conversation to a senior agent or a retention team. 2. Identifying Feedback Opportunities: - Trigger: When a conversation is resolved. - Condition: If the customer's last message had Positive sentiment. - Action: Schedule an automatic message to be sent after 24 hours asking the customer to leave a review or testimonial. 3. Triage and Organization: - Trigger: When a new conversation is created. - Condition: If the first message's sentiment is Negative. - Action: Mark the conversation as Pending and with high priority so it's the first in line to be handled. By automating the application of tags based on sentiment, you also enrich your reports. The Tags Report will then show not only the most common topics but also the proportion of satisfied and dissatisfied customers, providing a clear view of your service's health. Step-by-step: Creating a Sentiment-Based Automation Below is an example of how to create a rule to automatically escalate service for a dissatisfied customer. | Step | Action | Configuration Details | | :--- | :--- | :--- | | 1. Access Automations | In the Papo dashboard, navigate to the Automations module and select the option to create a new Rule. | Give the rule a clear name, such as "Escalate Negative Service". | | 2. Define the Trigger | Choose the event that will initiate the rule check. | Select the "Message Created" trigger and configure it to be activated only by customer messages. | | 3. Add the Condition | Specify the main condition that must be met for the rule to execute. | Add the condition: Message Sentiment > is > Negative. | | 4. (Optional) Refine the Rule | Add other conditions to make the rule more specific. | You can add a condition like Inbox > is > Technical Support so that the rule only applies to that channel. | | 5. Define the Actions | Determine what the platform should do when the conditions are true. | Add the action "Add Tag" and select the tag "Urgent - Dissatisfied". Then, add the action "Assign to Team" and choose the "Level 2 Support" team. | Frequently Asked Questions Is AI sentiment analysis 100% accurate? Sentiment analysis is based on machine learning models and has high accuracy, but it is not infallible. Sarcasm, irony, and complex cultural contexts can occasionally be misinterpreted. It should be used as a powerful aid and triage tool, but human judgment remains valuable for sensitive cases. In which languages does sentiment analysis work? The Papo platform is developed to support multiple languages. For a complete and updated list of languages covered by sentiment analysis, please consult the official documentation or the features page on the Papo website. What is the difference between sentiment analysis and AI tagging? These are two complementary functionalities. Sentiment analysis identifies the emotion or tone of the message (positive, negative, neutral). AI tagging identifies the subject or topic of the conversation (e.g., "Billing Inquiry," "Access Problem"). Together, they offer complete context: a conversation with Negative sentiment and an Access Problem tag is a clear priority. How does sentiment analysis help prevent churn (customer loss)? By detecting frustration and dissatisfaction in real-time, sentiment analysis enables proactive action. Instead of waiting for the customer to formalize a complaint or simply stop using your service, automations can escalate the service to a specialist or notify a manager immediately, increasing the chances of reversing the situation and retaining the customer. Can I create reports based on customer sentiment? Yes. By using automation rules to apply tags like "Satisfied" or "Dissatisfied" based on detected sentiment, you can use the Tags Report to visualize the distribution and trends of customer satisfaction over time, by channel, or by team. Start for free and understand your customers → The first user is free for you to connect your channels and test the power of AI.

What is the Prompt and AI Module and how does it empower your AI Agents?

The Prompt and AI Module is the control center on the Papo platform (https://papo.global/) where you define the personality, knowledge, and behavioral rules of your AI Agents. It is in this module that you create and edit "prompts" — the instructions that serve as the brain of your virtual assistant — ensuring it interacts with customers in a way that aligns with your brand, qualifies leads, and performs tasks autonomously and intelligently. The Prompt and AI Module transforms artificial intelligence from a generic resource into an expert trained for your business. Instead of relying on basic configurations, you have total control over how the AI thinks and acts, allowing you to create multiple agents for different channels or objectives. The table below summarizes the main components of the Prompt and AI Module and their functions. | Component | Main Function | | :--- | :--- | | Prompt Editor | Where you write the instructions that define the persona, objectives, and limits of the AI Agent. It is the "brain" of the operation. | | Agents | Allows you to manage the "General" (default) agent and create "sub-assistants" with specific prompts for tasks such as sales or support. | | General Rules | Defines behavioral guidelines that the agent must follow in all interactions, regardless of the specific prompt. | | Inboxes | Associates an AI Agent with one or more channels (WhatsApp, Instagram, etc.), activating it to respond to customers. | | Statistics | Shows AI performance data, such as successfully sent messages, failures, and skipped interactions, for continuous optimization. | How the Prompt Module functions as the AI's "brain" Think of the Prompt and AI Module as the training center for your AI Agent. This is where you teach it everything it needs to know to act as a member of your team. This is primarily done through the "prompt." The prompt is a set of detailed instructions in natural language. In it, you describe: - Persona: Who is the agent? Should it be formal, friendly, technical? - Objectives: What is its main function? Qualify leads, schedule meetings, offer first-level support? - Knowledge: What information can it provide about your products, services, and policies? - Restrictions: What should it not do? What topics should it avoid or transfer to a human? A well-crafted prompt is the difference between a generic chatbot and a virtual assistant that truly solves problems and generates business value. Instructions vs. Scenarios: What's the difference? Within the Papo ecosystem, it's important to distinguish how AI is trained for different purposes. The Prompt and AI Module primarily focuses on Instructions for AI Agents that interact with customers. - Instructions (Prompt): These are broad, contextual guidelines that guide the behavior of the AI Agent in open and dynamic conversations. They define the personality and reasoning logic so that the agent can handle a variety of unexpected customer questions. You configure this in the Prompt and AI Module. - Scenarios: These are more specific question-and-answer pairs or structured dialogue flows. They are used to train internal AI tools, such as Capitão (your team's assistant), to answer frequently asked questions or perform very specific tasks with a predefined script. In summary, instructions give your agent the ability to "think," while scenarios train it to "respond" to specific situations. How to empower your AI Agents with the Prompt Module Configuring an effective agent goes beyond simply turning on the AI. The module offers tools for fine-tuning and continuous adjustment. 1. Access the Module: In the side menu of the Papo platform, click on Prompt and AI. If the item is not visible, ask your account administrator to check your access permissions. 2. Create Specialized Agents: In addition to the "General" agent, which serves as the default, create sub-assistants. For example, you can have a "Sales Agent" with a prompt focused on qualification and scheduling, and a "Support Agent" with a prompt focused on resolving common queries and opening tickets. 3. Structure Your Instructions: In the Prompt Editor, you can use the free-text field or structured sections. The sections help organize guidelines into categories such as persona, objectives, and restrictions, ensuring you don't miss any important details. 4. Associate Agents with Strategic Channels: In the Inboxes area, you decide where each agent will operate. Activate your "Sales Agent" on WhatsApp and Instagram Direct, where prospecting is stronger, and your "Support Agent" on email or the website chat. 5. Monitor and Optimize: The Statistics area is crucial. If you notice many "Skipped" interactions, it may be a sign that the AI lacks confidence to respond, indicating that your prompt needs more detail or clarity. Analyze this data to continuously refine your instructions. Frequently Asked Questions What is the difference between the "General" agent and a sub-assistant? The "General" agent is the default assistant for your account. If you enable AI in an inbox without specifying an agent, it will be the one to respond. Sub-assistants are additional agents you can create with their own prompts and rules for specialized tasks, such as handling a specific channel or a marketing campaign. Can I have one AI Agent for sales and another for support? Yes. This is one of the main advantages of the Prompt Module. You can create a sub-assistant called "Sales Agent" with a prompt focused on qualifying leads and scheduling demos. Then, you can create another called "Support Agent" with instructions to answer frequently asked questions. Finally, in the "Inboxes" area, you associate each agent with the corresponding channel. What happens if the AI doesn't know how to respond? When the AI lacks confidence in the answer or a rule prevents it from interacting, the interaction is marked as "Skipped" in the statistics. The conversation remains in the inbox for a human agent to take over. This ensures that no customer is left without a response and serves as an indicator for you to improve your agent's prompt. How do I know if my prompt is good? The best way to evaluate your prompt is by analyzing the quality of the AI's responses and the data in the Statistics section. A good prompt results in a high number of successful interactions ("Succeeded") and few failures ("Failed") or skipped interactions ("Skipped"). If the AI is responding out of tone or providing incorrect information, it's time to review and further detail the instructions in the Prompt Editor. Start for free → Set up your first AI Agent at no cost and see your customer service work 24/7.

How to use Synapse to identify upsell and cross-sell opportunities?

To identify upsell and cross-sell opportunities, Papo (https://papo.global/) offers Synapse smart search, an AI module that scans your entire conversation history for keywords and phrases. This allows your sales team to proactively find customers who have shown past interest in complementary products, higher-tier plans, or additional solutions, transforming a message archive into an active source of new revenue. Your customer interaction history is a goldmine of information. Comments like "does your plan include X?" or "do you also have a solution for Y?" are clear signs of new sales potential. However, with a large volume of conversations, these opportunities are easily lost. Synapse solves this problem by allowing you to actively search for these signals, finding the exact moment a customer expressed a need that your company can meet with an upsell (a better product or plan) or cross-sell (a complementary product) offer. Search Strategies for New Sales Synapse search is a versatile tool. The secret to using it effectively is to think like a sales detective, looking for clues left by customers in previous conversations. | Opportunity Type | Suggested Search Terms | Action Goal | | :--- | :--- | :--- | | Upsell | "Upgrade", "more comprehensive plan", "limit", "I need more", "advanced feature" | Find customers who already use your product and are ready for a superior version. | | Cross-sell | Names of other products/services, "do you also have", "additionally, I need" | Offer complementary products or services to existing customers. | | Reactivation | "Quote", "proposal", "how much does it cost", "I'll think about it" | Reactivate leads who requested a quote in the past and didn't close the deal. | | Competitive Intelligence | Competitor names ("Competitor X") | Identify customers who compared your solution and present your differentiators. | How Synapse Search Reveals Opportunities Unlike a standard CRM search that finds a contact by name or phone, Synapse search investigates the content of conversations. It acts as an intelligence layer over your unified inbox, reading and indexing messages from all channels connected to Papo, such as WhatsApp, Instagram, and email. When you search for a term like "plan upgrade," Synapse doesn't just show you who said it, but presents the exact conversation, allowing you to understand the full context: - Who asked: The contact who showed interest. - When they asked: The date of the message, which helps assess if the opportunity is still relevant. - What was the context: What prompted the question and what response was given at the time. This depth of analysis allows your sales team to make much more personalized and effective approaches, based on a real need expressed by the customer themselves. Step-by-Step Guide to Finding Sales Opportunities with Synapse Using Synapse to generate new sales is a strategic process that can be divided into clear steps. | Step | Action | Details | | :--- | :--- | :--- | | 1. Access Synapse | In the Papo dashboard, navigate to the Synapse intelligence module. | Within Synapse, locate the History Search feature. | | 2. Define Strategy | Decide what type of opportunity you are looking for: upsell, cross-sell, or reactivation. | This definition will guide the terms you search for. | | 3. Execute Search | Enter the relevant search terms in the search field. | Use word variations. For example, for cross-sell, search for the name of a complementary product and also the problem it solves. | | 4. Analyze Results | Synapse will list all conversations containing the searched terms. | Click on each result to read the conversation snippet and understand the full context of the customer's request. | | 5. Act on Opportunity | Based on the analysis, start a new conversation with the customer. | You can create a follow-up task, assign the conversation to a salesperson, or initiate the approach directly with a personalized offer. | Practical Examples to Boost Your Sales Let's see how to apply these searches in your daily routine. Finding Upsell Opportunities Imagine you sell software with Basic, Pro, and Enterprise plans. A Basic plan customer might have asked a few months ago: "How do I get access to advanced reports?". At the time, they might not have been ready for the upgrade. Action: Use Synapse search to look for "advanced reports" or "upgrade". Find that conversation and re-engage the customer: "Hi, [Name]! I saw that a while ago you were interested in our advanced reports. We have a new condition for the Pro plan that might interest you. Would you like to know more?". Discovering Cross-sell Chances Your company sells digital marketing services (Product A) and also offers SEO consulting (Product B). A digital marketing customer might have commented in a conversation: "My organic traffic is very low." Action: Search Synapse for terms like "organic traffic", "SEO", or "Google". When you find customers who mentioned this pain point, you can offer your consulting service as the perfect solution, making a targeted cross-sell. Reactivating Old Proposals Many leads ask for a quote and then disappear. They might have gotten busy, or the timing wasn't ideal. Action: Search for "quote" or "proposal" and mentally filter for results from 3 to 6 months ago. Re-engage the contact: "Hi, [Name]! A few months ago, we talked about a proposal for [service]. Have things changed on your end? We have a new version of the service that might be even more useful for you." Frequently Asked Questions (FAQ) What is the difference between upsell and cross-sell? Upsell is the practice of encouraging a customer to buy a more complete, expensive, or premium version of a product they already own or are purchasing. Cross-sell is offering a product or service complementary to what the customer is acquiring. Does Synapse find opportunities automatically or do I need to search? Synapse has two functionalities. The Customer Radar proactively and automatically identifies contacts that need attention, including sales opportunities. The History Search, covered in this article, is a reactive tool where you actively search for specific terms to find opportunities based on your own hypotheses. Can I search for conversations from a specific period? Synapse search scans all conversation history available on the platform. Although the search itself does not have date filters, each displayed result shows the message date, allowing you to quickly assess the temporal relevance of the opportunity found. How do I know which customer to approach first after finding multiple opportunities? Prioritize based on context. A more recent conversation usually indicates a "hotter" opportunity. Additionally, analyze the sentiment expressed in the message (a customer who seemed frustrated with a current plan limitation is a strong candidate for an upsell) and cross-reference with other information from your CRM, such as "hot lead" tags or purchase history. Besides sales, what else can I use Synapse search for? Smart search is useful for several areas. You can use it to find product feedback (searching for "suggestion" or "wish it had"), identify customers at risk of churn (searching for "cancel" or "dissatisfied"), and conduct market research (searching for competitor names). Read also - How does smart search (Synapse) work in conversation history? - How to create and customize your sales pipeline in Kanban CRM Start using it for free and discover hidden opportunities → The first user is free for you to connect your channels and test the platform's intelligence.

How Papo's AI Generates Call Summaries and Transcriptions?

Papo's (https://papo.global/) call summary and transcription feature uses artificial intelligence to automatically convert voice call audio into text and then generate an intelligent summary. This process highlights the most important points, decisions made, and next actions, allowing teams to save time, optimize post-call follow-up, and improve service quality. In sales and support operations, many important details are defined during a call. Remembering everything or having to re-listen to a long recording to prepare for a follow-up consumes precious time. Papo's AI solves this challenge by automating transcription and summary creation, ensuring no crucial information is lost and speeding up service continuity. The Summary and Transcription Generation Process The transformation of a voice conversation into a summary and transcription happens automatically in the background, requiring no manual action from the agent. The process is divided into clear steps, from recording to making the results available. | Step | Papo AI Action | User Benefit | | :--- | :--- | :--- | | 1. Call Recording | The voice call, made through an integrated channel like WhatsApp or a telephony provider (e.g., Wavoip), is recorded on the platform. | The audio file is stored and accessible in the contact's history, within the "Call Report." | | 2. Automatic Transcription | The AI model processes the audio file of the recording and converts it entirely into text. | A complete transcription of the conversation is generated, allowing for keyword searches and detailed review of the dialogue. | | 3. Analysis and Summarization | Artificial intelligence analyzes the transcription to identify central themes, questions, answers, values, dates, and action items. | The essence of the conversation is extracted, separating relevant content from informal chats or filler words. | | 4. Summary Generation | Based on the analysis, the AI drafts a concise and structured text, presenting the key points, agreed decisions, and next steps. | A ready-to-read summary is attached to the call record, facilitating understanding of the context in seconds. | How to Access and Use Summaries in Practice For the user, the experience is fluid and fully integrated into the daily workflow in Papo. 1. Make or Receive a Call: The agent uses a voice channel configured in Papo, such as PapoCall for WhatsApp calls or a VOIP integration, to communicate with the client. 2. Completion and Registration: At the end of the call, it is automatically registered in the contact's timeline, visible along with text messages. 3. Background Processing: In the following minutes, Papo's AI begins processing the recording. This process is fully automatic and requires no intervention. 4. Access Generated Content: By clicking on the call event in the contact's history, the agent will find the call details (duration, date) and, immediately below, the complete transcription and the summary generated by the AI. 5. Execute Actions with Agility: With the summary in hand, the agent can quickly create a follow-up task, update the client's card in the CRM Kanban, or send a follow-up email with the discussed points, all without leaving the platform. This automation is especially useful for handing off tasks between agents, as it allows a new attendant to understand the entire context of the last call in seconds, without having to read a long history or listen to a recording. Who Is This Feature For? The AI summary and transcription feature adds value to various roles within a commercial or support operation. - Salespeople and SDRs: Save time by reviewing the main points of a qualification or negotiation call, preparing a faster and more accurate follow-up. - Support Agents: Quickly understand the history of a problem reported by voice, even if the call was handled by another colleague, ensuring service continuity. - Team Managers: Monitor the quality of interactions and the progress of negotiations more efficiently, focusing on summaries instead of listening to hours of recordings. - Quality Analysts (QA): Use transcriptions to audit script compliance and speech quality, and summaries to quickly identify the results of each call. Frequently Asked Questions Where can I find the call transcription and summary? Both the complete transcription and the AI-generated summary are available directly in the contact's conversation history. They are attached to the corresponding call event and appear shortly after the recording is processed. Does the AI summary work for all my calls? The functionality depends on the activation of call recording in your voice provider (such as PapoCall or Wavoip) and your plan in Papo. For the AI to transcribe and summarize a call, it must first be recorded by the platform. Is the call transcription 100% accurate? AI transcription technology is highly advanced, but its accuracy can be affected by factors such as audio quality, background noise, strong accents, or overlapping speech. Generally, clean and clear audio results in high-fidelity transcriptions. What is the difference between transcription and summary? Transcription is the literal conversion of the call audio into text, recording word-for-word what was said. The summary, on the other hand, is a short, intelligent text generated by AI that highlights only the most important points, such as decisions, agreed actions, and the central theme of the conversation. Are video calls also transcribed and summarized? Currently, Papo's AI transcription and summary feature is optimized for voice calls. For details on handling video call recordings, please consult the specifications of your plan and the communication channels configured in your account. How long does it take for the summary and transcription to appear? Processing usually occurs within a few minutes after the call ends. The exact time may vary depending on the call duration and the system's processing load at the moment. Read also - Why integrate a VOIP telephony system like Wavoip with Papo? - How to use AI to suggest the best response during service? Start using for free → Connect your channels and automate your service with the power of AI.

How Papo's AI Generates Call Summaries and Transcripts?

Papo's (https://papo.global/) call summary and transcription feature uses artificial intelligence to automatically convert voice call audio into text and then generate an intelligent summary. This process highlights the most important points, decisions made, and next actions, allowing teams to save time, optimize post-call follow-up, and improve service quality. In sales and support operations, many crucial details are defined during a call. Having to re-listen to a long recording to prepare a follow-up or remember all agreed-upon points consumes valuable time from your team. Papo's AI solves this challenge by automating transcription and summary creation, ensuring no important information is lost and speeding up service continuity. The Summary and Transcription Generation Process The transformation of a voice conversation into a summary and transcription happens automatically in the background, requiring no manual action from the agent. The process is divided into clear steps, from call recording to the availability of results on the platform. | Step | Papo AI Action | Benefit for the User | | :--- | :--- | :--- | | 1. Call Recording | The voice call, made through an integrated channel like WhatsApp or a telephony provider (VOIP), is recorded on the platform. | The audio file is stored and accessible in the contact's history, within the call record. | | 2. Automatic Transcription | The AI model processes the audio file of the recording and converts it entirely into text. | A complete transcript of the conversation is generated, allowing keyword searches and detailed review of the dialogue. | | 3. Analysis and Summarization | Artificial intelligence analyzes the transcript to identify central themes, questions, answers, values, dates, and action items. | The essence of the conversation is extracted, separating relevant content from informal chats or fillers. | | 4. Summary Generation | Based on the analysis, the AI drafts a concise and structured text, presenting key points, agreed decisions, and next steps. | A ready-to-read summary is attached to the call record, making it easy to understand the context in seconds. | How to Access and Use Summaries in Practice For the user, the experience is fluid and fully integrated into the daily workflow in Papo. 1. Make or Receive a Call: The agent uses a voice channel configured in Papo, such as PapoCall for WhatsApp calls or a VOIP integration, to communicate with the client. 2. Completion and Recording: At the end of the call, it is automatically recorded on the contact's timeline, visible along with text messages. 3. Background Processing: In the following minutes, Papo's AI begins processing the recording. This process is fully automatic and requires no intervention. 4. Access Generated Content: By clicking on the call event in the contact's history, the agent will find the call details (duration, date) and, immediately below, the complete transcript and the summary generated by the AI. 5. Execute Actions Quickly: With the summary in hand, the agent can quickly create a follow-up task, update the client card in the CRM Kanban, or send a follow-up email with the discussed points, all without leaving the platform. This automation is especially useful for handing off between agents. It allows a new attendant to understand the entire context of the last call in seconds, without having to read a long history or listen to a recording, ensuring more consistent and professional service. Who Is This Feature For? The AI summary and transcription feature adds value to various functions within a commercial or support operation, optimizing processes and freeing up time for strategic activities. - Salespeople and SDRs: Save time by reviewing the main points of a qualification or negotiation call, preparing a faster and more accurate follow-up. - Support Agents: Quickly understand the history of a problem reported by voice, even if the call was handled by another colleague, ensuring service continuity. - Team Managers: Monitor the quality of interactions and the progress of negotiations more efficiently, focusing on summaries instead of listening to hours of recordings. - Quality Analysts (QA): Use transcripts to audit script compliance and speech quality, and summaries to quickly identify the results of each call. Frequently Asked Questions Where can I find the call transcript and summary? Both the complete transcript and the AI-generated summary are available directly in the contact's conversation history. They are attached to the corresponding call event and appear shortly after the recording is processed. Does the AI summary work for all my calls? The functionality depends on the activation of call recording with your voice provider (such as PapoCall or a VOIP integration like Wavoip) and your plan with Papo. Calls need to be recorded for the AI to process the audio, transcribe it, and summarize it. Is the call transcript 100% accurate? AI transcription technology is highly advanced, but, like any automated system, it may occasionally present inaccuracies. Factors such as low audio quality, background noise, strong accents, or very specific jargon can affect accuracy. The transcript should be used as a powerful guide, but not as an infallible legal record. How long does it take for the summary and transcript to appear? Processing occurs in the background and usually takes only a few minutes after the call ends. The exact time may vary depending on the call duration and the system's processing load at the moment. Which languages are supported for transcription and summary? Papo is constantly expanding its AI capabilities. For the most up-to-date list of supported languages for call transcription and summary, please consult the official documentation on our website or contact our support team. Can I edit the AI-generated transcript or summary? The feature is designed to provide quick, automated insights, saving manual time. For details on the possibility of editing or annotating the generated texts, check the available options in the platform interface when viewing the call record. Read also - Why integrate a VOIP telephony system like Wavoip with Papo? - What are AI Agents and how do they work at Papo? Start for free and turn your calls into actions → The first user is free, allowing you to experience the power of AI at no initial cost.

How to create an automation to send a CSAT survey at the end of each service interaction

To send a CSAT survey automatically at the end of each service interaction, use a Papo Flow (https://papo.global/) with the "Wait" (Delay) step. After the conversation ends, the Flow pauses for a configurable period and then sends a satisfaction survey message, allowing you to collect feedback and measure service quality. CSAT survey automation eliminates the manual work of sending questions after every interaction. With the visual Flow builder, you create a journey that starts when a conversation is finished, applies a strategic waiting time, and sends the rating question — all integrated with the customer history and the platform's reports. | Component | Main Function | Configuration Example | | :--- | :--- | :--- | | Automation Flow | Orchestrate the post-service journey. | Flow triggered when closing a conversation or applying the "Service Finished" tag. | | "Wait" Step | Introduce a pause before sending the survey. | Wait 30 minutes, 2 hours, or 1 day. | | "Send Message" Action | Trigger the satisfaction question. | Text message (within 24h) or Message Template (HSM) with a 1 to 5 scale. | | CSAT Reports | View scores, averages, and distribution by agent. | Access the Reports module to track team performance. | How CSAT survey automation works The logic is simple: the Flow is triggered when a service interaction ends, waits an interval so the customer can reflect on the experience, and then sends a message asking for a score. The customer's response is recorded in the conversation and is available in Papo's CSAT reports, where you can analyze overall and per-agent satisfaction. Step by step to create the CSAT survey Flow 1. Access the Flows module: In Papo's main menu, go to Flows / Chatbot. 2. Create a new Flow: Give it a descriptive name, such as "Post-Service CSAT Survey". 3. Define the trigger: Configure the Flow to start when a conversation is closed. You can use a condition such as applying a "Service Finished" tag or a conversation closing event. 4. Add the "Wait" step: After the start node, click the + icon and select Wait. Enter the value and time unit (minutes, hours, or days). A delay of 30 minutes to 2 hours is usually enough for the customer to process the service. 5. Add the survey sending: After the "Wait" node, click + and choose Send Message. Write the question, for example: "How do you rate the service you received? Reply with a score from 1 to 5, where 5 is excellent." If the delay exceeds 24 hours, use a pre-approved Message Template (HSM). 6. Capture the response (optional): Papo automatically records the customer's response as part of the conversation. To route negative ratings to a supervisor, you can add a condition node that checks whether the score is less than 3 and, in that case, assign the conversation to a specific agent. 7. Save and activate the Flow: Review the logic, save your changes, and make sure the Flow status is "Active". From that moment on, every conversation that meets the trigger will receive the CSAT survey automatically. Who this automation is for - Support and service teams that want to measure service quality consistently. - Companies that use WhatsApp as their main channel and want immediate feedback without relying on external links. - Operations that need CSAT metrics to identify improvement points and recognize high-performing agents. Frequently Asked Questions (FAQ) ### How does Papo record the CSAT survey score? The platform associates the customer's response with the conversation and makes it available in CSAT reports. You can view the individual score, the overall average, and the distribution by agent, making performance analysis easier. ### Do I need to use a message template (HSM) for the survey? If the sending occurs within the 24-hour window after the customer's last interaction, you can use a regular text message. If the configured delay exceeds this period, you must use a WhatsApp-approved Message Template (HSM). ### Can I send the survey immediately after the conversation ends, without delay? Yes. Just don't include the "Wait" step in the Flow. The survey will be sent as soon as the conversation is closed. ### How do I see CSAT survey results? Go to the Reports module and find the CSAT section. There you find the satisfaction average, the list of recent responses, and performance by agent. ### Can I customize the survey question? Yes, you can write any text in the Flow message. It's recommended to use a clear scale (1 to 5) and a direct question to make it easier for the customer to respond. ### What happens if the customer replies something other than a score? The response is recorded in the conversation. You can configure a condition node in the Flow to handle non-numeric responses, for example redirecting the conversation to a human agent. Read also - How to measure customer satisfaction (CSAT) in Papo? - How to create an automation that sends a message after a specific time? Start for free → Create your free Papo account and automate CSAT surveys to measure customer satisfaction without manual effort.

How to set up Voice Assist to schedule meetings by voice command

To set up Voice Assist to schedule meetings by voice command in Papo (https://papo.global/), go to the settings of the Website-type inbox, enable Voice Assist, associate an AI Agent, and activate the scheduling action integrated with Google Calendar. This way, the website visitor can tap the microphone, ask for a time out loud, and Papo checks the calendar and creates the event automatically. | Requirement | Role in voice scheduling | | :--- | :--- | | Website chat widget | Where the microphone icon appears for the visitor. | | Voice Assist enabled | Allows receiving voice commands in the widget. | | Associated AI Agent | Interprets the request and conducts the conversation. | | Google Calendar connection | Checks free times and creates the event in Google Calendar. | | "Schedule a meeting" action | Activates the scheduling flow within the assistant. | How voice scheduling works Voice Assist turns the visitor's spoken request into a calendar event. When the user taps the microphone and says something like "I want to schedule a meeting tomorrow at 10 a.m.", Papo transcribes the audio and uses the same AI as Super Chat to interpret the request. From there, the assistant checks the connected Google Calendar, identifies free times, and creates the event with the date, time, and guest's email. The process follows three main steps: 1. Voice input: the visitor speaks the request in the website widget. 2. Interpretation and check: the AI understands the intent and checks the busy times in Google Calendar. 3. Event creation: Papo creates the meeting in the calendar without the visitor having to type. Step-by-step setup Before you start, make sure your Papo account is connected to Google Calendar. The platform shows the status of this connection and lets you list busy times and create events. Then follow the process in the dashboard: 1. Go to Settings → Inboxes. 2. Select an existing Website-type inbox or create a new one. 3. In the Voice Assist section, enable the feature. 4. Associate Voice Assist with an AI prompt already configured in your account. 5. Under interaction modes, check the Schedule a meeting option. 6. Save the changes so the widget is updated on the website. After saving, the widget will display the microphone icon. You don't need to change the website's installation code. Who this makes sense for This setup is useful for any business that relies on meetings booked through the website: - Consultants and service providers who offer an initial conversation. - Sales teams that turn visitors into demo meetings. - Clinics, salons, and studios that organize appointments without relying on typing. - Support teams that want to reduce friction for people browsing on their phones. FAQ Does Voice Assist schedule on its own or just show times? With the scheduling action enabled and Google Calendar connected, the assistant can check free times and create the event automatically in Google Calendar. Do I need to connect Google Calendar before enabling Voice Assist? Yes. For the scheduling flow to work, the Papo account must be connected to Google Calendar. Without this connection, there's no way to check availability or create the event. Does Voice Assist use the same AI as the text chat? Yes. The voice assistant is associated with an AI prompt in your account. This keeps the same knowledge base and the same behavior as text support. Does the visitor need to type any information? The request starts by voice. When necessary, the assistant can confirm the scheduling details in the conversation itself, but the main interaction is spoken. Can I use voice scheduling in any inbox? No. Voice Assist is a feature exclusive to the Website-type chat widget. Inboxes for other channels, such as WhatsApp or Telegram, follow the rules specific to each channel. Read also - How to enable and configure Voice Assist in your website chat? - How do availability hours for scheduling work? Start for free and enable Voice Assist on your website → With a free Papo account, you connect your Google Calendar and start scheduling meetings by voice command without switching tools.

How to create a birthday automation to send a discount coupon via WhatsApp

To create a birthday automation that sends a discount coupon via WhatsApp, use Papo's Custom Attributes and Flow builder (https://papo.global/). You will store the contact's birth date in a custom field and configure a Flow that is triggered daily, checking whether the current day and month match any customer's birthday and then sending the message with the coupon. This automation strategy turns a special date into an engagement and sales opportunity, without requiring manual work from your team. The combination of personalized data with visual automation logic allows each customer to receive a unique and relevant message at the right time. | Component | Main Function | Configuration Example | | :--- | :--- | :--- | | Custom Attribute | Store the birth date in the contact profile. | Create an attribute of type "Date" named birth_date. | | Automation Flow | Orchestrate the daily check and message sending. | Flow with recurring trigger, date condition, and send action. | | Condition Node | Check whether the current day and month match the contact's birthday. | Condition that compares the current date with the birth_date field. | | "Send Message" Action | Trigger the birthday message with the discount coupon. | Message template (HSM) with the coupon code and customer name. | How the birthday automation works The core logic is simple and powerful. First, you need to have your contacts' birth dates recorded in a Custom Attribute in Papo. With this data filled in, you build a Flow that runs at a defined frequency, usually once a day. On each run, the Flow checks, for each contact in your base, whether the day and month of the birth date match the current date. If so, the automation sends a birthday message containing an exclusive discount coupon directly to the customer's WhatsApp. Step by step to create the automation 1. Create the Custom Attribute: In the Contacts module, access the Custom Attributes settings. Create a new attribute of type "Date" and name it "Birth Date". This field will be available on each contact's profile to be filled in manually or via list import. 2. Fill in contact data: Make sure the birth date of the customers you want to reach is filled in the new attribute. You can do this by editing a contact individually or importing a spreadsheet with the data in bulk. 3. Access the Flows module: In Papo's main menu, go to Flows / Chatbot. 4. Create a new Flow: Give it a descriptive name, such as "Birthday Automation with Coupon". 5. Set the trigger: Configure the Flow to be started by a recurring trigger. The ideal configuration is a daily run, so the birthday check happens every day. 6. Add the Condition node: After the start node, add a Condition. The logic here is to check whether the day and month of the "Birth Date" Custom Attribute are equal to the day and month of today's date. Papo lets you dynamically compare date fields with the system's current date. 7. Configure message sending: On the "Yes" branch of the condition (when the date matches), add a Send Message action. Write a warm birthday message and include the discount coupon code. Example: "Happy birthday, {{contact.first_name}}! To celebrate, here is a 20% discount coupon for your next purchase: BIRTHDAY20. Valid until the end of the month!" 8. Save and activate the Flow: Review the logic, save your changes, and make sure the Flow status is "Active". From that moment on, the automation will run daily, ensuring that no birthday person is left without their special message and discount coupon. Who this automation is recommended for - E-commerce and retail stores that want to increase purchase recurrence with personalized offers. - Beauty salons, barbershops, and aesthetic clinics that can offer a discount on services during the customer's birthday month. - Restaurants and delivery services that seek to build customer loyalty with a complimentary dessert or a discount on the birthday order. - Any B2C business that collects customers' birth dates and wants to strengthen the relationship automatically and at scale. Frequently Asked Questions (FAQ) ### Do I need to use a message template (HSM) for sending? Yes, since the birthday message is proactive and sent outside a 24-hour customer service window initiated by the customer, you must use a Message Template (HSM) previously approved by Meta. You can create a template that includes variables such as the customer's name and the coupon code. ### How does Papo know today's date to compare with the birthday? The Flow's Condition node lets you compare a Custom Attribute of type "Date" with the system's current date. You configure the rule to check whether the day and month are the same, ignoring the birth year. ### Can I use a different discount coupon for each customer? Yes. If you have unique coupon codes stored in another Custom Attribute (such as a text field), you can insert it dynamically into the message using the corresponding variable, just as you insert the first name. ### What happens if a contact's birth date is not filled in? Simply nothing. The Condition node will not find a date match for that contact, and they will follow the "No" branch of the Flow (if one is configured) or simply will not receive the message. That's why it's important to keep your contact base up to date. ### Can I schedule sending for a specific time of day? Yes. The Flow's recurring trigger can be configured to run at a specific time, such as 9 a.m., ensuring the message arrives at the beginning of the birthday person's day. ### Is it possible to add a "Wait" step after sending for a follow-up? Yes. You can extend the Flow by adding a "Wait" node after sending the first message and then a new "Send Message" node for a follow-up, such as a reminder to use the coupon a few days later. Read more - How to create an automation that sends a message after a specific time? - How to create an automation to send a CSAT survey at the end of each service Get started for free → Create your Papo account and turn special dates into sales with smart WhatsApp automations.

How to use Automatic Categorization to route conversations to the right team

To automatically route conversations to sales, support, or finance teams in Papo (https://papo.global/), you combine the AI's intelligent classification with an Automation Rule: the AI identifies the conversation topic and applies a tag, and the automation redirects the conversation to the right team as soon as that tag is recorded. Instead of manually transferring conversations or relying on a single team to triage, you teach the AI to recognize the profile of each request and create triggers that place the conversation in the correct team's queue within seconds. This reduces response time and improves the customer experience, since they are helped by someone who can actually solve the problem. | Method | How it works | Practical result | | :--- | :--- | :--- | | Manual classification + automation | The agent reviews the conversation, applies the corresponding tag (Sales, Support, Finance), and the automation rule triggers assignment to the team. | Human control over categorization; ideal for processes that require review. | | Automatic AI classification + automation | The AI, configured in the Prompt & AI module, analyzes the content and automatically applies the appropriate tag; the automation rule then redirects the conversation. | Instant routing, with no manual intervention; scales with conversation volume. | How Automatic Categorization distributes conversations among teams The complete automatic routing flow follows three steps: 1. AI analysis – The AI agent reads the first message (and history, if any) to understand the customer's intent. Based on the training you defined, it classifies the conversation as "Sales", "Support", or "Finance". 2. Tag application – The classification materializes as a tag automatically applied to the contact and the conversation. Tags are the signal that automations will read. 3. Automation Rule – A rule configured for the trigger “Tag applied” identifies the tag that was just added and executes the action “Assign a Team”. This way, a sales lead goes to the Sales team, a billing question goes to Finance, and so on. Step by step: configuring the AI and routing rules 1. Prepare department tags Before any automation, create the tags that represent the destination teams: - Go to Settings and find the Tags section. - Create one tag for each department, such as “Sales”, “Support”, “Finance”. Use clear names and colors that help with visualization. These tags will be applied by the AI or manually and will serve as conditions for the rules. 2. Configure the AI to classify conversations automatically In the Prompt & AI module, you define how the AI agent should behave. To have it apply tags automatically: - Go to the Prompt & AI module and open the configuration for the agent handling the desired inbox. - In the instructions field (prompt), include clear guidance so the agent identifies the type of request and applies the corresponding tag. Example excerpt: “If the customer asks about pricing, a demo, or plans, classify as Sales and apply the ‘Sales’ tag. If it is a billing question or tax deadline, classify as Finance and apply the ‘Finance’ tag. For other questions and technical issues, classify as Support.” - Save the configuration. From that point on, new conversations handled by the agent will receive the tag automatically. 3. Create automation rules for each team Now turn the tag into action: - Go to Settings → Automation and click to add a new rule. - Set the trigger as “Tag applied to a conversation”. - In the condition, select the desired tag (e.g., “Sales”). - Choose the action “Assign a Team”. Then select the team that should receive the conversation (e.g., “Sales Team”). - Repeat the process for the other tags (Support, Finance), always linking each one to the corresponding team. 4. Test and adjust the flow Send a test message that simulates a sales, support, and finance question. Check whether the tag is applied automatically and whether the conversation appears in the correct team's queue. If necessary, refine the AI prompt to improve classification accuracy. Who this automatic routing is recommended for - Companies with multiple specialized teams – Sales, support, and finance operate on different channels or the same number, and manual triage causes delays. - High-volume operations – When the number of conversations doesn't allow an agent to classify before transferring, AI eliminates this bottleneck. - Those who want to use AI beyond automatic replies – The same agent that talks to the customer also organizes the back office, distributing tasks to the right people. FAQ Does the AI already classify automatically, or do I need to train it? The classification and tag application behavior is configured by you in the Prompt & AI module. The AI learns from the instructions you provide and improves with interactions. The more directions you give in the prompt, the more accurate the categorization will be. Can I use the lead score to route conversations? Yes, as a complement. You can create automation rules that trigger based on a tag applied manually after analyzing the AI score. For example, a 5-star lead can receive the “VIP Customer” tag, and a rule sends it directly to the best salesperson. However, department routing (sales, support, finance) works best with intent classification, as described above. Do I need to have teams created before configuring the rules? Yes. The “Assign a Team” action requires the teams to already be registered in your Papo account. You can create and manage teams in the account settings area. Do the rules apply to conversations that are already in progress? No. Automation rules run only on new conversations entering the inbox. To redirect an existing conversation, use manual assignment in the conversation panel. What if a conversation doesn't fit any of the tags I configured? The conversation will remain in the inbox without automatic team assignment. To keep these cases from stalling, maintain a rule or manual association that sends untagged conversations to a default team (such as a triage center) or the general support team. Read also - How to set up rules to automatically assign conversations from a specific channel to a team - How to use Automatic Categorization to prioritize VIP customer conversations Start free now → The first user is free and no credit card is required – set up your AI and routing rules in minutes.

How to Use Lead Score in a Flow to Send Personalized Offers

To send personalized offers based on each lead's interest level, you combine Papo's Lead Score (https://papo.global/) with an automation Flow, creating branches that trigger the right message according to the contact's score — hot leads receive a direct offer, warm leads receive educational content, and cold leads are nurtured with information, all without manual intervention. Instead of treating every contact the same way, you adjust communication automatically as soon as the lead enters the flow, increasing the relevance of your messages and your conversion rate. This strategy connects the CRM's intelligent qualification directly to Papo's automation journeys. | Stage | What happens | Owner | | :--- | :--- | :--- | | Qualification | Assignment of the Lead Score (AI or manual) and continuous updating of the contact's score | Papo CRM and AI | | Flow creation | Configuration of a flow with conditions based on the score (e.g., stars or tags) | Marketing / Sales | | Personalized trigger | Automatic sending of the correct message to each flow branch, such as offers, content, or follow‑ups | Papo Flow (automation engine) | How Lead Score feeds the automation flow Papo's Lead Score assigns a rating from 1 to 5 stars to each contact, indicating their purchase potential. The score can be generated in two complementary ways: - Automatic with AI: the artificial intelligence analyzes the conversation history, looking for signs of intent, urgency, and questions about prices or terms. Each evaluated conversation consumes one AI credit, and the result updates the lead ranking automatically. - Manual assignment: a team member can directly set the star rating or apply a tag, such as "Hot Lead" or "Requested a Quote", reflecting external interactions or the salesperson's perception. When you insert a condition node in a Flow, you can check the contact's Lead Score value (the stars) or a specific tag. Based on that value, the flow decides which path to follow and which message to send — all in real time, without any manual action. Step by step: creating the flow with personalized offers Build your automatic sales flow in just a few minutes using Papo's visual builder. The example below creates three communication branches, but you can adjust the number of levels and the messages according to your strategy. | Step | Action in the Flow builder | Important details | | :--- | :--- | :--- | | 1. Define the trigger | Choose how the contact enters the flow (e.g., when receiving a first message, when added manually, or when reaching a specific tag) | The trigger can be configured to run automatically whenever a new lead is captured | | 2. Insert condition (Lead Score) | Add a condition node and configure it to check the "Lead Score" field (stars) or a tag such as "Hot Lead" | You can combine more than one criterion (e.g., score > 3 AND tag "Quote") to narrow it down even further | | 3. Create the branches | Define the rules: for example, leads with 4 or 5 stars follow the "Hot" path, leads with 3 stars the "Warm" path, and the rest the "Cold" path | Each branch can contain multiple steps, such as an immediate message, a pause, and a follow‑up | | 4. Personalize the messages | In each branch, insert a "Send Message" node with text suited to the lead's profile | For hot leads, a direct and personal proposal; for warm leads, educational material or an invitation to a webinar; for cold leads, a welcome message and scheduling a new contact | | 5. Save and activate | Review the logic, save the flow, and switch the status to "Active" | The flow will start processing new contacts immediately; those already in the CRM can be added manually, if needed | If, in any branch, you include the "Wait" step and the waiting time exceeds WhatsApp's 24-hour service window, the next message must be a Message Template (HSM) approved by Meta. The platform lets you select these templates directly in the send node. Practical example of communication by interest level Imagine an equipment store that wants to turn initial interest into a sale. The flow works like this: - Hot Lead (4 or 5 stars): the contact immediately receives a message with a special offer: "Hi [Name], I noticed you were looking for information about machine X. For you, we're offering 15% off until Friday. Would you like to close the deal now?" - Warm Lead (3 stars): the system sends a short video with the product's highlights and offers to answer questions, without applying pressure. - Cold Lead (1 or 2 stars): communication is gentler and educational, with a link to the catalog and a suggestion to schedule a conversation when they are closer to a decision. This automatic personalization increases conversion because it delivers the right argument at the right time, respecting each lead's stage. Who this strategy is for - B2B and B2C sales teams that capture leads via WhatsApp, website, or social media and want to automate qualification and the first approach. - Businesses that already have lead scoring set up and want to go beyond reports, turning the score into immediate action. - Operations that need to scale relationships without growing the team — the flow does the work of classifying and responding automatically, while salespeople focus only on hot leads. Frequently asked questions (FAQ) Can I use tags instead of stars for the condition? Yes. Instead of checking the numerical score, you can configure the condition to evaluate whether the contact has a specific tag, such as "Hot Lead" or "Quote Requested". This is useful when qualification is done mostly by the team. Is the Lead Score updated on its own? AI assignment can run periodically to re-evaluate new or old conversations, consuming one credit per evaluation. In addition, manual ratings and tags can be changed at any time. The flow always uses the contact's most recent value at the moment it runs. What's the difference between using a Flow and a campaign via the Broadcaster? The Flow is ideal for individual, branched journeys, where each contact can receive a different message according to their profile and actions. The Broadcaster is aimed at mass sends to a static segment, with the same message for everyone in the group. Here, you combine the Flow's personalization with score data for a much more precise approach. Do I need AI credits to use Lead Score in a Flow? Only automatic AI analysis consumes credits. Creating the flow and reading the score (whether automatic or manual) does not consume additional credits. If you already have a scored base, you can activate the flow immediately. Can I combine Lead Score with other conditions, such as the date of the last purchase? Yes, the Flow builder lets you combine multiple criteria in the same condition node. You can, for example, create a branch for leads with a high score and who haven't interacted in more than 30 days, sending a highly targeted re-engagement offer. How does the AI decide a lead's rating? The AI processes the contact's conversation history, looking for words and expressions that indicate real purchase intent, urgency, or specific questions about prices and deadlines. Based on these signals, it assigns the stars. Read also - How to use Lead Score to segment hot contacts and send a WhatsApp campaign - How to create an automation that sends a message after a specific time? Start for free → The first user is free so you can test personalized offer automation with Lead Score and all of the platform's modules.

How to configure Voice Assist to answer in different languages

To make Papo Voice Assist (https://papo.global/) answer in different languages, you associate an AI Agent (prompt) that understands and responds in the desired language with the website widget. The voice assistant inherits the same behavior as Super Chat: if the prompt is configured to respond in Portuguese, English, or Spanish, voice recognition and responses will follow that guidance, with no need for extra per-language settings. The Voice Assist turns your website chat widget into a voice contact point, and its "brain" is exactly the same AI Agent you use for text support. This means multilingual capability depends directly on how you structure the prompt (assistant instructions). With a single comprehensive prompt or with multiple widgets for each language, you can offer voice support in Portuguese, English, and Spanish simply and consistently. | Strategy | How it works | When to use | | :--- | :--- | :--- | | Single multilingual prompt | You create an AI Agent that understands voice commands in Portuguese, English, and Spanish and responds in the same detected language. The Voice Assist associated with that prompt will serve the three languages in the same widget. | Ideal for websites with low international traffic or when the user experience does not require separating languages. | | Widgets per language | You create a Site-type inbox for each language (e.g., Portuguese page, English page). Each widget receives a Voice Assist configured with a prompt exclusive to that language. | Ideal for websites with translated sections, where each audience should speak with an assistant that already speaks their language natively. | How Voice Assist handles different languages Voice Assist does not have a separate language selector. Instead, it follows the same AI flow as text chat: 1. The user taps the microphone and asks a spoken question. 2. Papo transcribes the audio to text. 3. The text is sent to the same AI Agent that answers text conversations (Super Chat). The prompt defines the rules, including the language(s) the assistant should understand and use to respond. 4. The generated response is displayed as text in the chat window — and, if the audio response option is enabled, Papo synthesizes the audio in that language using the available voices. Therefore, the key to multilingual voice support is the prompt: - If the prompt instructs the agent to respond in Portuguese, Voice Assist will only understand and respond in Portuguese. - If the prompt includes the instruction "respond in the same language as the question," and the agent has the knowledge for it, it can switch between Portuguese, English, and Spanish automatically. - If you create a specific prompt "AI Agent – English," the Voice Assist associated with it will serve only in English. Step-by-step setup Before you start, make sure the website widget is already installed. Configuration is done per inbox. 1. Go to Settings → Inboxes in the Papo dashboard. 2. Select the Site-type inbox you want to adjust or create a new one. 3. In the Voice Assist section, enable the feature. 4. In the Prompt field, choose the AI Agent that contains the language instructions you prepared (e.g., "Multilingual Agent" or "English Support Agent"). 5. If you want the assistant to also respond by audio, check that this option is enabled in the Prompt & AI module (the audio inherits the language of the text response). 6. Save your changes. If you chose separate widgets per language, repeat the process for each Site-type inbox, associating the corresponding prompt with each one. Who this makes sense for - Online stores that sell abroad and need visitors from different countries to receive voice assistance in their language. - Tourism, hospitality, and event companies with a multinational audience browsing the website in more than one language. - Corporate websites or support portals with translated versions (e.g., /en, /es). - Businesses that already use Super Chat in multiple languages and want to extend that capability to Voice Assist without duplicating effort. Frequently asked questions Do I need to create a special Voice Assist for each language? No. You keep the same feature, but you can associate a different AI Agent with each website widget. Voice Assist inherits the language of the prompt, so you just need prompts (Agents) configured for the languages you want to serve. Does Voice Assist automatically understand whether the customer speaks English or Spanish? The language detection capability depends on the AI Agent prompt. If you configure the assistant to "respond in the same language as the question," and the AI has sufficient knowledge, it will be able to identify the language and respond appropriately. For greater predictability, the safest method is to have separate prompts per language and use them in different widgets. Can I use the same Knowledge Base for all languages? Yes, as long as the documents are in the necessary languages. The AI consults the Knowledge Base and can extract information in Portuguese, English, or Spanish if the content is available in those languages. Do I need to change the widget code for each language? No. You just create different inboxes for each version of the site (or reuse the same one if you use a multilingual prompt). The installation script is the same; the prompt association is done in the dashboard. Will the audio response come out in the correct language? Yes. Voice synthesis uses the text generated by the AI Agent. If the agent responded in Spanish, the audio will be in Spanish, within the voices available on the platform. Read also - How to enable and configure Voice Assist in your website chat? - What is the Prompt Module and how does it empower your AI Agents? Start for free and enable Voice Assist on your website → With Papo's free account, you can configure Voice Assist and start serving in Portuguese, English, and Spanish without depending on other tools.

How to create a flow that sends a daily summary of conversations by email

Yes, you can automate sending a daily summary of conversations by email in Papo (https://papo.global/) using the Flows module. The platform combines the artificial intelligence that already analyzes and summarizes interactions with a scheduled sending action, generating a report that arrives in the manager's inbox every day, with no manual work. With this automation, the person responsible for the operation receives a consolidated view of what happened in support: conversation volume, main topics, hot leads, and even customer sentiment. All extracted automatically by Papo's AI and delivered in a clear format by email. | Step | Action | Where to configure | | :--- | :--- | :--- | | 1 | Have an email credential configured (SMTP) | Settings → Inboxes | | 2 | Create a Flow with a daily trigger (schedule) | Flows module | | 3 | Add an AI action to generate the day's summary | Action node in Flow (AI/Processing) | | 4 | Include an email sending action with the summary | Email action node in Flow | | 5 | Activate and test the flow | Flows module | How the automated daily summary works Papo already uses artificial intelligence to analyze conversations in real time — scoring leads, categorizing topics, and offering insights through the Capitão assistant. The daily flow simply schedules the collection of this information and sends it by email. 1. Scheduled trigger: The Flow starts automatically once a day, at the time you set (e.g., 8 p.m., after business hours). 2. Collection and summarization: An AI action queries the conversations from the period and generates a summary with the most relevant data: total interactions, pending conversations, highest-scoring leads, most applied tags, and even an analysis of the predominant sentiment. 3. Email sending: The generated text is inserted into the body of an email and sent to the manager (or recipient list) using the SMTP credential already configured in the platform. The result is a daily report that arrives without anyone having to compile spreadsheets or read dozens of chats. Step by step to create the flow Prerequisite: email sending credential Before building the flow, you need to have an email inbox configured in Papo for sending. Go to Settings → Inboxes, add an Email inbox, and enter your SMTP server details (server, port, user, and password). It is recommended to use services such as Amazon SES or SendGrid to ensure deliverability. 1. Create a new Flow In the side menu, access the Flows module and click to create a new flow. Give it a descriptive name, such as “Daily conversation summary”. 2. Configure the scheduled trigger Add a Trigger node and choose the scheduling option (daily). Set the time when the summary should be generated and sent — for example, at 7 p.m., right after the day's support ends. 3. Add the AI action for summarization Insert an Action node that uses Papo's artificial intelligence. Configure the prompt so that the AI: - Considers conversations from the last 24 hours. - Extracts metrics such as number of open, resolved, and pending conversations. - Lists contacts with the highest lead score (hot customers). - Groups the most frequent tags (e.g., “Quote”, “Technical support”). - Optionally, includes an overall sentiment analysis. The AI will return structured text that will serve as the email body. 4. Add the email sending action After the AI node, include another Action node of type Send email. Select the credential configured in the prerequisite, enter the recipient's address (manager), and use the variable with the summary generated in the previous step as the message content. You can also set a fixed subject, such as “Interaction summary – {{date}}”. 5. Activate and test Save the flow and activate it. To test, you can force a manual execution (if available) or wait for the next scheduled time. Check that the email arrives correctly and that the content reflects the day's conversations. Adjust the AI prompt as needed to include exactly the information the manager needs. Who this automation is useful for - Support managers who need to monitor the operation without entering the platform several times a day. - Sales supervisors who want to know which leads are hottest and which deals have progressed. - Business owners who outsource support but want a daily summary of the quality and volume of interactions. - Operations with multiple channels (WhatsApp, Instagram, email) that want to consolidate everything into a single morning report. Frequently asked questions Do I need to have an email credential configured before creating the flow? Yes. Sending the summary depends on an email inbox with SMTP configured in Papo. Without it, the flow will not be able to send the message. Does the AI generate the summary automatically, or do I need to program every detail? Papo's AI already analyzes conversations in real time (score, tags, assistant). In the flow, you only guide what you want to include in the summary through a prompt. The platform takes care of collecting and formatting the information. Can I customize the email content? Yes. The email body is built from the text that the AI generates. You can adjust the prompt to include specific metrics, greetings, or even a table with the hottest leads. What happens if there are no conversations on the day? The flow will run normally, but the summary will indicate that there was no activity. You can configure an optional condition not to send the email if there are no conversations, avoiding empty messages. Can I send the summary to more than one recipient? Yes. In the email action, you can enter multiple addresses (separated by commas) or even use a predefined contact list, as long as the sending credential supports the volume. Does the flow consume AI credits? Generating the summary uses Papo's AI processing, which is included in the platform's plans. Details about usage limits can be consulted directly on the website. Read also - How to send notifications of new conversations to Slack using Webhooks - How to create and use email templates for your campaigns? Start for free and automate your reports → With the first user free, you can test the daily summary flow and scale your operation at no initial cost.

How to create an automatic welcome flow for new WhatsApp contacts

To create an automatic welcome flow for new WhatsApp contacts, use the Flows module in Papo. Build a flow with an action that sends an introduction and qualification message, and configure Synapse to automatically enroll every contact who starts a conversation. With this structure, your team doesn’t need to send manual welcome messages. The flow kicks in as soon as the lead sends their first "hi" on WhatsApp, engaging them with an immediate response and guiding them to the next step. Automatic welcome flow summary | Component | Role in the flow | Example use | | :--- | :--- | :--- | | Flow (Flows/Chatbot) | Contains the visual sequence of actions that will be executed for each enrolled contact. | Flow called “WhatsApp Welcome”. | | “Send Message” action node | Triggers the introduction and qualification message on the contact’s WhatsApp. | Text with greeting, contact name, and interest question. | | Synapse | Automation that connects the Conversations module to the Flow, automatically enrolling the contact as soon as a new conversation starts. | Rule: “When new WhatsApp conversation → enroll contact in Welcome Flow”. | | Condition (optional) | Flow branch based on the contact’s response. | If the lead replies “I want to buy”, forward to a salesperson; if “I need support”, send to the support team. | How it works: step by step 1. Create the welcome Flow In the Papo side menu, go to Flows / Chatbot. Click the button to add a new Flow and give it a descriptive name, such as “WhatsApp Welcome”. The editor will open with the initial node (“Start”). 2. Build the introduction and qualification message Click the “+” sign after Start and choose the Send Message action. In the text field, write a clear, customizable message. Use the available variables to include the contact’s name, such as {{ contact.first_name }}. Example: Hello, {{ contact.first_name }}! We’re from [Your Company]. How can we help you today? To point you in the right direction, are you looking for information about our plans, technical support, or a demo? You can also add quick reply buttons (when available) to make interaction easier. 3. Add qualification steps (optional) After the first message, insert an Ask a Question node to capture the contact’s response. Then use a Condition node to split the path: - If the contact replies with a keyword such as “plans” → send to a salesperson (“Send Message” action with a scheduling link). - If they reply “support” → forward to the support inbox (“Transfer to Agent” or “Add Tag” action). These branches make the flow interactive and reduce manual work. 4. Activate the Flow After building the sequence, click save and then set the Flow status to Active. The flow will only be processed when it is active. 5. Configure automatic enrollment with Synapse For the Flow to trigger as soon as a new contact starts a WhatsApp conversation, the contact needs to be automatically enrolled in it. Papo offers the Synapse module to create connections between modules. - Go to Synapse and create a new automation. - Define the trigger: When a new conversation is created (selecting the desired WhatsApp inbox). - Choose the action: Enroll contact in Flow and select the “WhatsApp Welcome” Flow. This way, every time a new number sends a message, the contact will be instantly enrolled in the welcome flow. From that moment on, initial support will be standardized, even outside business hours, without manual intervention. Who is this flow for? - Businesses that receive a high volume of first contacts on WhatsApp and can’t respond to everyone manually. - E-commerce businesses, service providers, and SaaS companies that want to qualify the lead before involving a salesperson. - Sales teams that need to save time: the flow already delivers basic information and filters interest, leaving agents focused on hotter leads. - Companies that provide support outside business hours: the welcome message is sent immediately, keeping the lead engaged until a human takes over. Frequently asked questions Do I need an approved message template (HSM) for the welcome flow? Not necessarily. Because the flow is triggered as soon as the contact starts a conversation (within the 24-hour customer service window), the message qualifies as a response and does not require an approved HSM. However, if you want to include buttons or media, it is recommended to use a template to ensure delivery. Can I use the flow to send an image or presentation video? Yes. In the Send Message node, you can attach an image, video, or document in addition to text. This is useful for sending a catalog, a virtual business card, or a short welcome video. How do I personalize the message with the name of the contact who started the conversation? The Flow editor lets you insert dynamic variables. When you type {{ contact., you will see options such as first_name, last_name, or custom fields. Choose the desired one, and the system will replace it with the contact’s actual value at send time. What if the contact already has a previous conversation? Will the flow be triggered again? No. The automation in Synapse can be configured to enroll only contacts who are starting their first conversation. This way, leads who have already spoken with the company will not receive the welcome message repeatedly. Can I add a delay before sending the welcome message? Yes. Between the Start node and the send action, insert a Wait (delay) node and set the desired time (for example, 10 seconds). This gives a “natural” response feel before the bot introduces itself. What happens if the contact does not respond to the qualification message? You can end the flow at that point or configure a new automation to, after a period without a response, send a follow-up or enroll them in another nurturing flow. Read also - How to create a birthday automation to send a discount coupon via WhatsApp - How to use the Flows Library to install ready-made automations ** Start for free and create your welcome flow right now →** Your team can start serving customers the same day with Papo.

How to set up automatic appointment reminders via WhatsApp

To set up automatic appointment reminders via WhatsApp and reduce no-shows, use the Appointments module from Papo. Connect your Google Calendar account, define reminder rules, and the platform sends automatic WhatsApp messages before each scheduled time, keeping your clients informed and your calendar full. With this automation, your team doesn’t need to send manual reminders. The system triggers the message at the right time, directly in the client’s WhatsApp, using your company’s official number. Setup is simple and integrates with the scheduling flow you already use, whether via AI or manually. Automatic reminder setup summary | Component | Main function | Example use | | :--- | :--- | :--- | | Appointments module | Manages appointments and sends automatic WhatsApp reminders. | Google Calendar connection to sync times and trigger messages. | | Google Calendar | Source of scheduled events and times. | Professional’s, salon’s, or clinic’s calendar. | | Reminder rules | Define when and how many reminders are sent before the appointment. | One reminder 24 hours before and another 1 hour before. | | WhatsApp (Cloud API) | Channel for sending reminder messages. | Message with date, time, and scheduled service, sent automatically. | | Flows (optional) | Lets you create custom message sequences for more complex scenarios. | Flow that sends confirmation, reminder, and post-appointment survey. | How automatic reminder sending works The logic is straightforward: whenever an appointment is created in your connected calendar, Papo automatically schedules WhatsApp reminders according to the rules you define. The message reaches the client as a normal WhatsApp notification, with no one needing to copy and paste texts. Papo’s Appointments module includes a follow-up tool that manages these reminders. You can configure multiple triggers (for example, one day before and one hour before) and customize the content of each. If the appointment was made by Papo’s AI during a conversation, the reminders are automatically queued. Step by step to set up reminders 1. Connect Google Calendar In the Appointments module, link your Google Calendar account. This integration allows Papo to read events and sync times in real time. 2. Define reminder rules Within the same module, access the follow-up settings. Choose the desired intervals — such as 24 hours, 2 hours, or 30 minutes before the appointment — and write the message that will be sent. You can use variables such as the client’s name and the service time. 3. Create the appointment (or let the AI create it) The appointment can be made manually by an agent or automatically by the AI during a WhatsApp conversation. In both cases, the appointment appears in Google Calendar and Papo schedules the reminders based on the configured rules. 4. Track the sends At the scheduled time, the message is sent from the WhatsApp number connected to your Papo account. The conversation history is recorded in the inbox, allowing your team to see whether the client replied or reacted. For more elaborate sequences — such as asking for confirmation, sending a last-minute coupon, or a satisfaction survey after the appointment — you can combine the Appointments module with Flows and Synapse. That way, the reminder becomes part of a complete relationship journey. Who this automation is for - Beauty salons, barbershops, and aesthetic clinics that depend on scheduled times and suffer from last-minute no-shows. - Medical offices and health professionals (dentists, psychologists, physiotherapists) who need to reduce absenteeism without overloading the secretary. - Freelance service providers (personal trainers, photographers, consultants) who manage their own calendars and want to automate communication. - Any business that schedules appointments via WhatsApp and wants to reduce the number of “no-shows” with a polite, automatic reminder. Frequently asked questions Do I necessarily need Google Calendar? Yes. The automatic reminder integration uses Google Calendar as the source of events. If you don’t use it yet, you need to create a free Google account and connect it to Papo. Can I customize the text of each reminder? Yes. In the follow-up rules, you write the message that will be sent, and you can include the client’s name, the scheduled service, and the time. It’s possible to have different texts for the 24-hour reminder and the 1-hour reminder, for example. Do reminders work if the client doesn’t have my number saved? Yes. Because sending is done through the official WhatsApp API (Cloud API), the message arrives normally, even if your number isn’t in the client’s contacts. Can I send more than one reminder per appointment? Yes. You can configure as many reminders as you want, each with its own lead time and message. What if the client reschedules or cancels? When the event is changed or removed in Google Calendar, Papo automatically adjusts the scheduled reminders, canceling those that are no longer needed. Can I use Flows to create a more complex sequence? Yes. For journeys that go beyond a simple reminder — such as asking for confirmation, sending a payment link, or a post-service survey — you can create a Flow and use Synapse to trigger it from Appointments module events. Read also - How to create a birthday automation to send a discount coupon via WhatsApp - How to create an automatic welcome flow for new WhatsApp contacts Start free now → The first user is free forever, allowing you to set up automatic reminders and reduce no-shows at no initial cost.

How to create an automated NPS survey on WhatsApp

To send an automated NPS survey on WhatsApp, use a Papo Flow (https://papo.global/) that starts after a service interaction ends, waits for a configurable interval, and sends the question "On a scale of 0 to 10, how likely are you to recommend our company?" The customer's response is recorded in the conversation and can be used to classify promoters, neutral customers, and detractors, guiding follow-up actions based on the contact's loyalty. NPS automation eliminates manual survey sending and ensures every customer receives the same standardized question. With the visual Flow builder, you build a journey that detects the end of the service interaction, applies a strategic delay, and collects the score, with the option to create branches to handle detractors or reinforce promoters. | Component | Main Function | Configuration Example | | :--- | :--- | :--- | | Automation Flow | Orchestrate the post-service journey. | Flow triggered when a conversation is closed or the "Service Completed" tag is applied. | | "Wait" Step | Introduce a pause before sending the survey. | Wait 30 minutes, 2 hours, or 1 day after the conversation is closed. | | "Send Message" Action | Send the NPS question. | Text message (within 24 hours) or Message Template (HSM) with the 0 to 10 scale. | | Condition Node | Classify the response as Promoter, Neutral, or Detractor. | Check whether the score is >= 9, between 7–8, or <= 6 to trigger different actions. | | Reports | Track score distribution and response rate. | Reports module (available on the platform) lets you view scores associated with conversations. | How NPS survey automation works The logic is similar to a CSAT survey, but adapted to measure loyalty. The Flow is triggered when the service interaction ends (you can use a closing event or the application of a tag). After an interval—so the customer can reflect on the experience—Papo sends the standard NPS question. The response (a number from 0 to 10) is recorded in the conversation. You can optionally use a condition node to automatically separate detractors (low scores) and route them to a supervisor, or send a special thank-you message to promoters. Step-by-step to create the NPS survey Flow 1. Access the Flows module: In Papo's main menu, go to Flows / Chatbot. 2. Create a new Flow: Give it a clear name, such as "Post-Service NPS Survey". 3. Define the trigger: Configure the Flow to start when a conversation is closed. You can use the application of a "Service Completed" tag or the conversation closing event itself. 4. Add the "Wait" step: After the initial node, insert a Wait node. Set the wait time—for example, 2 hours—so the customer can process the service interaction before responding. 5. Add the NPS question send: Next, include a Send Message action. Write the question, for example: "On a scale of 0 to 10, how likely are you to recommend our company to a friend or colleague? (0 = would not recommend, 10 = would definitely recommend)". If the delay exceeds 24 hours, you must use an approved Message Template (HSM). 6. Capture and classify the response (optional but recommended): Papo automatically records the customer's response. To act proactively, add a Condition node after sending. Configure three branches: - Promoter (9–10): send a thank-you message and, if desired, ask the customer to leave a public review. - Neutral (7–8): record it internally and, if applicable, trigger a light follow-up to understand improvement points. - Detractor (0–6): assign the conversation to an agent or supervisor, or send a request for a callback, such as "We're sorry about your experience. Someone from our team will contact you to understand better." 7. Save and activate the Flow: Review the logic, save it, and change the status to "Active". From that point on, every conversation that meets the trigger will receive the NPS question. Responses are available in the contact history and can be analyzed in reports. Who this automation is for - Companies that want to measure customer loyalty beyond point-in-time satisfaction. - Operations that use WhatsApp as their main channel and seek scalable feedback without relying on external links. - Customer success managers and CX leaders who need data to identify churn risks and brand ambassadors. - Sales and post-sale teams that want to nurture promoter customers with targeted campaigns. Frequently Asked Questions (FAQ) How does Papo record the NPS survey score? The platform associates the customer's response with the conversation. You can view the score in the contact history and in the available reports, allowing you to consolidate your base's Net Promoter Score. Do I need to use a message template (HSM) to send it? If the configured delay is less than 24 hours after the customer's last interaction, a regular text message is enough. For delays longer than 24 hours, a Message Template (HSM) approved by Meta is required. Can I send the survey immediately after the conversation ends, without a delay? Yes. Simply do not include the "Wait" node in the Flow. The question will be sent as soon as the conversation is closed. Is it possible to use a different scale than 0 to 10? The standard NPS is 0 to 10, but you can customize your message question and the condition logic for any scale that makes sense for your business, although the official metric requires the 11-point scale. How do I see NPS survey results? Response data is available in the Reports module. You can filter by period, agent, or tag to calculate the percentage of promoters, neutral customers, and detractors. What happens if the customer replies with text instead of a number? Papo stores the literal response. If the score is not converted automatically, you can adjust the Flow condition to interpret responses such as "9" or "ten". If numerical accuracy is critical, recommend in the message that the customer reply with only the number. Can I send a mass NPS survey to a contact list? Flow-based automation is triggered by an event (end of conversation). For proactive sends to a list, use the Broadcast module, scheduling the HSM send to the desired segment, within consent policies. Read also - How to create an automation to send a CSAT survey at the end of each service interaction - How to measure customer satisfaction (CSAT) in Papo? Start free → Create your free account and start sending automated NPS surveys on WhatsApp to measure customer loyalty without manual work.

How to run quick WhatsApp polls with reply buttons

To run quick WhatsApp polls with reply buttons in Papo (https://papo.global/), use interactive button messages to send satisfaction questions and capture the customer's choice right inside the conversation, with no external links. The answer is recorded in the contact's history and can be used to automate follow-up, centralize data, and reduce manual work for your support team. What polls with reply buttons are These are WhatsApp messages that present a question and clickable options, such as "Great", "Good", or "Fair". Instead of asking the customer to type a free-form answer, Papo structures the interaction into buttons. This increases the response rate, standardizes answers, and lets your team or your automation make decisions based on the chosen option. Papo keeps the entire conversation in the same inbox, with the contact's full history. Button answers appear in the chat timeline and can be combined with other platform modules, such as automations, CRM, and reports. Overview of the poll with reply buttons feature | Feature | Description | Main benefit | | :--- | :--- | :--- | | Interactive message with buttons | Question with clickable options inside WhatsApp. | One-tap answer, no typing. | | Structured collection | The customer's choice is recorded in the conversation history. | Organized data for tracking. | | Post-answer automation | The answer can trigger flows, classify the contact, or close the conversation. | Less manual work and faster follow-up. | | Centralized history | Answers appear on the same contact profile, alongside texts and files. | Complete context for the team. | How a quick poll works in practice The conceptual flow is simple and does not require the customer to leave WhatsApp. 1. Define the question and the options – Choose a clear question, such as "How do you rate our support?", and the answer options you want to offer. 2. Send the interactive message – In the contact's conversation, send the message with buttons using the WhatsApp channel connected to Papo. 3. The customer taps an option – The answer is automatically recorded in the conversation, with no need for the customer to type. 4. Track the answer in the history – The team sees the choice in the chat timeline, keeping the full context of the conversation. 5. Automate the next step – Based on the answer, you can trigger an automation, open a deal in the CRM, or send a follow-up. When to use quick WhatsApp polls Polls with buttons are useful in many support and sales scenarios. - Post-support satisfaction survey – Ask how the support experience was right after the ticket is resolved. - Lead qualification – Ask about interest, budget, or buying stage. - NPS and CSAT – Collect the customer's score through structured questions. - Preference confirmation – Find out the best time, channel, or service the customer wants. - Less noise in the conversation – Avoid open-ended answers that need to be interpreted manually. Who this feature is for Support, customer success, sales, and marketing teams that use WhatsApp as their main channel can use quick polls to measure satisfaction, prioritize leads, and automate repetitive tasks. Papo centralizes the answers in the same dashboard, which avoids parallel spreadsheets and loss of context between agents. Frequently asked questions Do I need an official WhatsApp number to use reply buttons? Yes. Interactive messages with buttons work in WhatsApp inboxes connected through the official API (Cloud API). To learn how to connect your number, see the WhatsApp connection guide or the Papo website. Is the customer's answer recorded in the history? Yes. The chosen option appears in the conversation timeline, alongside text messages, voice notes, and files. This lets any agent pick up the context without losing information. Can I use the answer to trigger an automation? Yes. Papo has automation and flow modules that can use the customer's answer to route, classify, schedule a follow-up, or close the conversation. The exact setup may vary depending on your use case and plan. What is the difference between a quick poll and an NPS or CSAT survey? A quick poll is a direct question with reply buttons. NPS and CSAT are specific survey use cases that can be run with the same type of interaction, by adapting the question and the options. Does the customer need to download an app? No. The customer answers by tapping the button inside WhatsApp itself, without installing anything extra. Can I use polls for sales and qualification? Yes. You can ask about interest, budget, timeline, or buying stage and use the answer to prioritize the conversation or move the contact through the funnel. Read also - How to create an automatic NPS survey on WhatsApp - How to create an automation to send a CSAT survey at the end of every conversation Start for free and create your first WhatsApp polls → Try Papo with the first user free and validate response collection in the same support dashboard.

How to personalize automated messages with the contact's name and data

To personalize automated messages with the contact's name and data in Papo (https://papo.global/), use contact variables in Flows and campaigns. That way, every WhatsApp or email message can automatically include the first name, company, subscribed plan or any custom attribute of the contact, without the team having to edit the text one by one. What contact variables are in practice Variables are dynamic placeholders that Papo fills with each contact's data at the moment of sending. Instead of writing "Hello, customer", you write a message with the name variable, and each person receives their own name. Papo's CRM already includes standard fields such as name, phone and email. In addition to those, you can create custom attributes to store information specific to your business, such as company, job title, plan, date of last purchase or student status. | Data type | Example content | Common use in personalization | | :--- | :--- | :--- | | Standard field | Name, phone, email | Opening greeting and contact identification | | Text attribute | Company, city, job title | Qualify the conversation and adapt the tone of the offer | | Number attribute | Total spent, age, order ID | Create progressive offers or business rules | | Date attribute | Date of last purchase, due date | Re-engagement, reminders and seasonal campaigns | | List attribute | Subscribed plan, student status | Routing and specific messages per profile | How to personalize a message in a Flow The Flows module lets you build visual sequences of messages and decisions. Personalization with contact data works mainly in the message-sending node. 1. Go to Flows / Chatbot and create a new flow. 2. After the initial node, add the Send Message action. 3. In the text field, write the message and insert the name variable. Example: Hello, {{ contact.first_name }}! I saw you were interested in our plan. 4. To include data such as company or job title, first create the attribute in Settings → Custom Attributes and fill in the contact's profile. 5. Use Condition nodes to send different messages depending on the attribute value. For example: if the plano_contratado attribute is "Pro", send advanced content; if it is "Basic", send upgrade tips. 6. Save and activate the flow. This format is especially useful in welcome, qualification and nurturing flows. The contact gets an individual experience, even when the message is sent automatically. How to use the same data in campaigns In WhatsApp and email campaigns, the logic is similar: you segment the audience based on CRM data and personalize the content of the send with the available fields. Custom attributes help you create dynamic segments. On the Contacts screen, you can filter by criteria such as "Subscribed Plan is equal to Gold" or "Date of Last Purchase is before 90 days". That segment can then be used as the audience for a campaign, ensuring the right message reaches the right people. | Step | What to do | Result | | :--- | :--- | :--- | | Create attributes | Define fields such as company, plan or purchase date | CRM enriched with business data | | Populate the data | Fill in manually, via a Flow or via integration | Contacts ready for personalization | | Create a segment | Filter contacts by attributes and save the segment | Dynamic, up-to-date audience | | Build the send | Use the contact's data in the campaign text | Individual messages at scale | Who this personalization makes sense for - B2B sales: include company, job title and market segment for more consultative approaches. - E-commerce: use last purchase, category of interest and total spent for relevant offers. - SaaS and education: personalize messages by plan, subscription status or enrolled course. - Support and after-sales: start conversations with the customer's name and route them through the right profile. Frequently asked questions Do I need custom attributes to include the contact's name? No. The standard name field can already be used directly in the message, for example with the variable {{ contact.first_name }}. Custom attributes are for data specific to your business, such as company, plan or date of last purchase. How do I fill in a custom attribute automatically? You can fill in attributes manually on the contact's profile, through a question asked in a Flow, or through integrations that send data from your system to Papo, such as API or webhooks. Can I use this data in WhatsApp and email campaigns? Yes. CRM data and custom attributes can be used to segment the audience and personalize campaign content. The exact fields available in each template can be checked on the website or in the official documentation. What happens if an attribute is empty? The message may be sent without that information. That's why it's worth populating the essential data before the contact enters the flow or before a campaign send. Can I personalize scheduled messages? Papo has a scheduled messages module. To confirm which variables are available for this type of send, check the official documentation or the platform's support team. Read also - How to create an automatic welcome flow for new WhatsApp contacts - How to create Segments to group contacts and conversations dynamically Start for free → Create your free account and start personalizing automated messages with your contacts' data.

How to debug an automation Flow that isn't executing the expected actions

To debug an automation Flow that isn't executing the expected actions in Papo (https://papo.global/), use the execution history to trace the path each contact took and check whether the action nodes were reached and executed. The visual editor, node-by-node history, and Automation Report help you distinguish a logic problem — the contact never reaches the node — from a configuration problem in the action node itself. A Flow that runs but doesn't execute the expected actions is often more challenging than a Flow that is completely stopped: it receives contacts, advances through some nodes, and only fails at specific steps. Instead of rebuilding everything, history- and log-based debugging lets you isolate exactly where execution deviates from the planned behavior. Diagnostic summary | Situation | Where to investigate | Typical fix | | :--- | :--- | :--- | | Action doesn't fire, but the contact reaches the node | Action node configuration | Review and correct the node parameters in the editor | | Action doesn't fire and the contact doesn't reach the node | Previous conditions or branched path | Expand conditions or add a default route | | Contact response isn't recognized | Response history of the question node | Include expected response variations | | Contact stops before the end | Node without continuity or question without exit | Add a default path and route to an agent | | Low completion rate even with an active Flow | Automation Report | Simplify the journey and compare metrics | First: confirm the Flow is Active Before analyzing the logic, check the Flow status. Only Flows with the Active indicator are ready to receive and process contacts. An inactive Flow can be edited, but it doesn't execute any action: contacts don't enter and already enrolled contacts don't advance. Second: use the node-by-node execution history The execution history is the main diagnostic source. Papo records every step a contact takes inside the Flow, letting you compare the actual path with the one designed in the editor. 1. Locate the specific execution Open the Flow and look for the execution history, which lists the contacts that went through it. If you already know which contact had the problem, open their profile in the CRM and check which stage of the flow they are in. 2. Compare the actual path with the planned one When you open the details of an execution, you see the exact sequence of nodes traversed. If the contact stopped before an expected action, the problem is in the previous logic. If they reached the node but the action had no effect, the problem is in the node's own configuration. 3. Validate question node inputs If the Flow uses the Ask a Question component, check in the history which response was captured. Variations such as "yes" instead of "Yes", words with different accents, or unexpected responses can make the condition fail to recognize the input and follow an unplanned path — or simply stop. 4. Check specific actions If an action such as Add Tag or Update Contact didn't work, confirm in the history whether the node was reached and executed. If it was skipped, correct the previous logic. If it was executed and the contact didn't change, review the node parameters in the visual editor. Third: fixing configuration errors After identifying where the Flow breaks, the most common fixes include: - Expand conditions: add expected response variations or a default path that captures any unexpected input. - Add an escape route: include a node that offers the option to talk to a human agent, to prevent the contact from getting stuck in a dead end. - Simplify the journey: remove unnecessary steps that may cause abandonment midway. - Review the Automation Report: compare the Contacts, Completed, and Completion Rate metrics to validate whether the fix worked in new executions. When a Flow has many questions and many nodes Flows with multiple questions — such as screenings, HR forms, onboarding, or surveys — are more susceptible to logic failures, because each branch depends on the previous answer. In these cases, node-by-node diagnosis is even more important: a single misconfigured condition can divert the entire flow to the wrong path. Review condition nodes in sequence, validating each captured response and the branch activated. When to start over from the Flow Library If the current flow is too complex or has logic that is hard to fix, the Flows module offers a Library with preconfigured automations that cover common scenarios such as welcome messages, lead qualification, and surveys. Starting from a working template reduces the chance of configuration errors and speeds up debugging. Frequently Asked Questions How do I know whether an action node was executed? Open the execution history and locate the contact. You'll see the exact sequence of nodes traversed. If the action node appears in the path, it was reached; if it doesn't appear, the contact was diverted by a previous condition. Why does the contact always stop at the same node? There is usually a condition that cannot be satisfied or an unrecognized response in the previous node. Check in the history which response was captured and whether the branching options cover all possible variations. Can an inactive Flow execute any action? No. Only Flows with Active status can receive and process contacts. Inactive Flows can be edited, but they don't execute any action or advance already enrolled contacts. How do I know whether the fix worked? Monitor the Automation Report after applying the fix. Compare the number of Contacts with Completed and observe the Completion Rate in new executions to validate the change. What should I do when the Completion Rate remains low? Simplify the journey, remove unnecessary steps, and ensure there is always an exit route to a human agent. If the flow is too long, consider splitting it into smaller Flows connected by the Enroll in another Flow action. Read also - How to create an automatic welcome flow for new WhatsApp contacts - How to use Lead Score in a Flow to send personalized offers Start for free and debug your Flows in practice → You can create a free Papo account and use the execution history to test and adjust your automations.

How to edit or remove documents from the AI Knowledge Base

To edit or remove documents from the Knowledge Base in Papo (https://papo.global/), you manage the files directly in the Knowledge Base module. The platform allows you to remove obsolete documents and replace them with updated versions, ensuring that the artificial intelligence always consults the correct content during interactions. Keeping the base clean and accurate is essential for assistants like Capitão to provide answers aligned with your company's current policies. Without this maintenance, the AI may rely on outdated information, generating inconsistencies and rework. | Action | Description | | :--- | :--- | | Delete document | Permanently remove a file from the base. The AI stops consulting it immediately after synchronization. | | Edit document | Editing is done by replacing the old file with a new version. Upload the updated document and delete the previous one. | | View document history | The base's document list shows name, upload date, and synchronization status, making it easier to identify what needs to be reviewed. | | Impact on responses | After deletion or replacement, the AI's responses will reflect only the content of the active documents in the base. | How document management works in the Knowledge Base Papo's Knowledge Base is the central repository where you store files that feed the AI assistants. Manuals, FAQs, commercial policies, or any text document can be uploaded and linked to an assistant like Capitão. The AI then uses these files to generate automatic responses on WhatsApp and other channels. The technical process behind this functionality is called Retrieval-Augmented Generation (RAG). When a customer asks a question, the AI searches for relevant excerpts in your reference documents and uses them as a basis to formulate a cohesive and contextualized response. When a document becomes obsolete — for example, a return policy that has changed — the ideal is to delete it and, if necessary, add the new corresponding document. This prevents the AI from mixing outdated information with current information. When to delete or edit a document - Policy or procedure change — Whenever a terms of service, warranty, or internal rule is changed, remove the old file and upload the revised version. - Error correction — If a file contains incorrect data, replace it with the corrected version so the AI does not reproduce the error. - Discontinued products — Remove materials for discontinued items to prevent the AI from recommending something that is no longer available. - Base reorganization — Over time, the base may accumulate documents on similar topics. Deleting duplicates helps the AI find the right content faster. General steps to delete and replace documents 1. Access the Knowledge Base module in the Papo panel. 2. Select the base you want to manage. You can have multiple bases organized by subject (e.g., “Products”, “Finance”). 3. Locate the document you want to delete. The list shows name, size, and synchronization status. 4. Remove the document using the delete option available on the file's row. 5. If necessary, upload the new version directly to the same base. The file will be processed and will be available to the AI in moments. The exact interface may vary depending on the platform version, but the logical flow is always the same: identify, remove, and replace. Frequently Asked Questions (FAQ) Can I edit a text directly on the platform? Papo does not offer a built-in text editor to change the content of an already uploaded document. To make an edit, delete the old file and upload a new version with the desired corrections. When I delete a document, do the AI's responses change immediately? Yes. As soon as the deletion is completed and synchronization is updated, the AI stops consulting that file. If you have already uploaded a replacement document, the AI will automatically start using the new content. What happens if I accidentally delete a document? Simply upload the same file to the base again. Since the synchronization process is fast, the AI will start using it again in a few moments. How many documents can I have in a single Knowledge Base? There is no strict published limit, but it is recommended... Start for free and keep your AI always up to date → With Papo, you manage your knowledge base simply so that the AI always responds with the most recent information from your company.

How to Automatically Archive Inactive Conversations After a Period

To automatically archive inactive conversations in Papo (https://papo.global/), you create an automation rule that closes and moves conversations with no interaction for a defined period to the archive, keeping the inbox organized and the team focused on active conversations. With this automation, you eliminate the manual work of reviewing and closing stalled conversations. The rule monitors inactivity and executes the archiving action without human intervention, ensuring that the service queue shows only what truly requires attention. Summary: manual archiving vs. automation rule | Method | How it works | When to use | | :--- | :--- | :--- | | Manual archiving | An agent reviews the inbox, identifies stalled conversations, and archives them one by one. | Operations with very low volume or when full control over each closure is needed. | | Automation rule | The system checks the time without interaction and automatically archives conversations that meet the configured criteria. | Medium- and high-volume operations, where keeping the inbox organized manually is unfeasible and prone to failure. | How the automatic archiving rule works The rule uses the conditional logic of Papo's Automation module. When a conversation reaches the set time without any new message (from the customer or the agent), the system executes the action to archive it. You define the inactivity period and, if desired, can filter by channel, tag, or other criteria, refining which conversations will be archived. Step by step to create the rule 1. Access the Automation module In Papo's dashboard, go to Settings → Automation. This is the hub where you manage all rules that automate actions in customer service. 2. Start a new rule Click the button to add a new rule. Give it a descriptive name, such as “Archive inactive conversations after 7 days,” for easy identification and future adjustments. 3. Define the inactivity trigger Choose the event that triggers the automation. For archiving by inactivity, select the trigger related to conversation without interaction. Then set the time condition: - Inactivity period: define how many days (or hours) the conversation must go without new messages to be considered inactive (e.g., 7 days). You can add extra conditions, such as a specific tag (e.g., “Resolved”) or a source channel, to limit which conversations enter the rule. 4. Choose the “Archive conversation” action In the actions section, select Archive conversation (or the equivalent closing action). From that point on, every conversation that reaches the defined inactivity time will be automatically moved to the archive, freeing up the inbox. 5. Save and activate the rule Review the settings, save, and make sure the rule is active. The inbox will stay organized without manual action. Who this automation is for - Teams with a high volume of customer service: dozens or hundreds of daily conversations don't pile up; only active ones remain visible. - Operations that prioritize SLA and productivity: agents don't waste time digging through inactive conversations and focus their effort on customers who are still waiting for a response. - Companies using multiple channels: you can create channel-specific rules (e.g., WhatsApp archives after 5 days, email after 10), adapting the archiving policy to each context. Frequently asked questions Does the rule archive conversations that are waiting for the customer's response? Yes, as long as the configured inactivity time is reached. If you want to preserve conversations where the customer has not yet replied, set a longer period or use a tag that excludes the conversation from the rule (e.g., “Waiting for customer”). Can I set different timeframes for different channels? Yes. Create one rule per channel, using the “Source inbox” condition and the respective inactivity period. This way, WhatsApp can archive after 3 days and email after 7. What happens if the customer sends a new message after archiving? The conversation is automatically reopened and appears again in the inbox as a new interaction. All previous history is preserved. Is it possible to archive only conversations that already have a specific tag? Yes. Add a tag condition to the rule. For example, you can archive only conversations with the “Resolved” tag that have been inactive for 5 days. Does the archiving rule affect reports or metrics? No. Archiving does not delete data. Conversations remain accessible in the archive and are still counted in historical reports. Only the inbox view is cleared. Do I need special permission to create automation rules? Permissions for accessing automation rules follow your Papo account's role configuration. Details about specific permissions can be checked directly on the platform. See also - How to set up rules to automatically assign conversations from a specific channel to a team - How to create an automation to send a CSAT survey at the end of each service interaction Start free and reduce noise in your inbox → The first user is free and no credit card is required — set up automatic archiving rules today.

How to create an Automation Rule to automatically tag conversations from a specific channel

To automatically tag conversations from a specific channel in Papo (https://papo.global/), you can create an automation rule with the trigger “New conversation”, the condition “Channel is equal to”, and the action “Add tag”. This setup eliminates the need to manually categorize each incoming message, ensuring that all conversations from that channel are organized from the very first moment. Summary: manual tagging vs. automation rule | Method | How it works | When to use | | :--- | :--- | :--- | | Manual tagging | An agent opens each conversation and applies the desired tag. | Very low volume or when the decision depends on case-by-case human analysis. | | Automation rule | The system detects a new conversation, checks the source channel, and applies the tag automatically. | Operations with medium to high volume, multiple channels, or the need for fast, consistent categorization. | How the tagging rule by channel works The rule uses Papo's Automation module, which works with "if-then" logic. You define a trigger (the event that fires the rule), a condition (the filter the conversation must meet), and an action (what the system executes). In the case of tagging by channel: - Trigger: "New conversation" — the rule is fired whenever a new conversation is created on the platform. - Condition: "Source inbox is equal to" — you select the desired channel (e.g., WhatsApp Sales, Instagram, Customer Service Email). - Action: "Add tag" — you choose the tag that will be applied automatically. You can combine multiple conditions (such as time or message content) and even add more than one action (for example, tag and assign to a team at the same time). Step-by-step to create the rule 1. Access the Automation module In the Papo panel, go to Settings and click Automation. This is the hub where you manage all rules that automate service actions. 2. Start a new rule Click the button to add a new rule. Give it a descriptive name, such as “Tag Instagram conversations as ‘Social’”. This makes it easier to identify and adjust later. 3. Set the “New conversation” trigger Choose the event that fires the automation. To tag as soon as the conversation arrives, select the trigger related to new conversation created (or "New conversation"). 4. Configure the channel condition Add a condition to filter by channel. Select the Source inbox option (or equivalent) and set the operator to "is equal to". Then choose the desired channel from the list of available inboxes (e.g., “WhatsApp Support”, “Instagram”, “Sales Email”). You can add other conditions, such as an existing tag or keywords in the first message, to further refine when the rule should act. 5. Choose the “Add tag” action In the actions section, select Add tag. Then choose the tag that will be applied. If the tag does not exist yet, you can create it on the spot or use one already registered in the account. If desired, add extra actions, such as “Assign to a team” or “Send automatic message”. 6. Save and activate the rule Review the settings, save, and make sure the rule is active. From that moment on, every new conversation that arrives through the defined channel will automatically receive the configured tag. Who this automation is for - Teams operating multiple channels: if you serve customers via WhatsApp, Instagram, email, and chat, you can create a rule for each channel, tagging conversations with the source (e.g., “WhatsApp”, “Instagram”, “Email”) and making organization and reporting easier. - Operations that need quick triage: the automatically applied tag can be used by other rules or flows to route service, prioritize, or trigger automatic responses. - Companies that use tags to analyze volume by channel: with tags, you can filter reports and understand the volume of each channel without manual work. Frequently asked questions Can I add more than one tag in the same rule? Yes. In the actions section, you can include multiple “Add tag” actions, each with a different tag. This way, an Instagram conversation can receive the “Social” and “Lead” tags at the same time. Does the rule work for conversations already in progress? No. Automation rules are applied only to new conversations that arrive in the inbox. To tag old conversations, use manual tag application in the conversation panel. Do I need to create the tag before setting up the rule? Not necessarily. While creating the rule, when you select the “Add tag” action, you can choose an existing tag or create a new one on the spot. The system allows you to create the tag directly in the configuration flow. Can I combine the channel condition with other conditions? Yes. You can add extra conditions, such as “the message contains the word X” or “the contact has the tag Y”. The rule will only run if all conditions are met at the same time. What happens if the conversation already has the tag applied? The tag is added without duplicating. If the conversation already has the tag, the system keeps the existing one and does not create a second identical one. The tag history remains organized. Can I create different rules for each channel? Yes. Create one rule for each channel, defining the corresponding inbox condition and the tagging action with the desired channel name or category. This allows granular organization of your service. See also - How to create a rule to automatically archive inactive conversations - How to set up rules to automatically assign conversations from a specific channel to a team Start free → Create your free Papo account and start automating conversation categorization in minutes, at no initial cost.

How to train Capitão to identify purchase intent and transfer to a salesperson

To train Papo's Capitão (https://papo.global/) to identify purchase intent and automatically transfer to a salesperson, create a Scenario with clear purchase trigger instructions and set the handoff action to the sales team. You can complement it with Tools to qualify or log the opportunity before the transfer. That way, messages like "I'd like a quote", "do you have a demo?" or "how do I sign up?" trigger the right salesperson instead of just getting an informational reply. | Customer situation | Example phrase | Suggested instruction in the Scenario | Result | | :--- | :--- | :--- | :--- | | Request for price or terms | "How much does the plan cost?" | If the customer asks for price, quote, value or payment terms. | Transfer to sales | | Request for a demo or meeting | "I want to see a demo" | If the customer asks for a demo, meeting or consultant. | Transfer or schedule | | Intent to close | "I want to sign up now" | If the customer says they want to hire, close, sign or start. | Transfer to salesperson | | Urgency or active comparison | "I'm comparing and I need to decide today" | If there is decision urgency or active comparison. | Transfer to salesperson | Why Scenarios are the way to handle purchase intent Scenarios are specific instructions you give Capitão so it knows exactly which action to take in predefined situations. Unlike FAQs, which are meant to answer questions, Scenarios are triggers for actions — such as transferring the conversation to a human agent. Purchase intent is a signal for action: the customer doesn't want a definition, they want to get the purchase done. That's why the Scenario is the right resource. When a new message arrives, Capitão analyzes the intent. If it matches the instruction of an active Scenario, it executes the associated action immediately, instead of trying to craft a reply from the knowledge base. This ensures a hot lead isn't treated like an ordinary question. Step by step to train Capitão to transfer to a salesperson 1. Go to the settings of the Capitão assistant you want to train. 2. Find the Scenarios area — also called Handoffs in some versions. 3. Create a new Scenario and give it a descriptive title, such as "Handoff to sales team". 4. Write the instruction with the purchase signals that should trigger the action. Use natural language and be specific. Examples: - When the customer shows purchase intent, asks for price, quote, demo or proposal, or says they want to sign up. - If the customer mentions the words "I want to close", "sign up now", "payment method" or "discount". - If the user asks to talk to a salesperson, consultant or specialist. 5. Set the handoff action to a sales agent or team. 6. Enable the Scenario. New Scenarios can be created disabled for testing; review the configuration and enable it so it works in real time in conversations. The exact menu names may vary, but the configuration logic is this: title, instruction and action. How to use Tools to qualify and log before the transfer Capitão's Tools are actions the assistant can perform to automate tasks and interact with other systems. They let the AI go beyond answering questions by carrying out practical actions during the conversation. There are two main types: | Tool type | Description | Example use in qualification | | :--- | :--- | :--- | | Native | Ready-to-use integration, no advanced technical setup required. | Check available slots in a salesperson's calendar and create the event automatically when the customer confirms. | | Custom | Connection to an external system via API, such as a CRM or ERP. | Log a new deal or lead in the CRM as soon as purchase intent is identified. | In practice, you can combine a Scenario and a Tool: the Scenario detects purchase intent and the Tool logs the opportunity or schedules the meeting. Then the handoff routes the conversation to the salesperson, who receives the full history with context already in place. Who this configuration makes sense for - Companies with a sales team separate from support. - Operations that receive quote and demo requests on the same channel. - B2B businesses that need to route hot leads without manual rework. - Teams that want the AI to capture purchase intent outside business hours and leave the conversation ready for the salesperson the next day. Frequently asked questions What's the difference between a purchase Scenario and a price FAQ? An FAQ replies with information, while a Scenario executes an action. If the goal is to transfer the conversation to a salesperson, set up a Scenario with a handoff, not an FAQ. Which signals can Capitão recognize as purchase intent? Anything described in the Scenario instruction: a request for price, quote, demo, consultant, urgency, or a mention of closing or signing up. Keep the instruction specific to reduce false positives. What happens when purchase intent is identified? Capitão stops searching the knowledge base for an answer and triggers the action configured in the Scenario, usually the handoff. The conversation is transferred with its history to the salesperson or sales team. Do I need technical knowledge to create a handoff Scenario? No. You define the title, instruction and handoff action. Custom Tools, which connect external systems via API, do require advanced technical knowledge. Can I schedule a meeting automatically before transferring? Yes. You can enable a native calendar Tool so Capitão checks available slots and creates the event. Then the Scenario can route the conversation to the salesperson with the appointment already booked. Is the Scenario active immediately? New Scenarios can be created disabled for testing. After reviewing the configuration, you enable it so it applies to new conversations. Read also - How to transfer a conversation to another agent during support - How to create a Deal directly from a WhatsApp conversation Start for free → Start for free to validate the flow with the first user for R$ 0.

How to test an automation Flow before activating it

To test an automation Flow before activating it for real customers in Papo (https://papo.global/), you can temporarily activate the Flow, manually enroll a test contact, and review the execution history to validate the complete journey. This process lets you debug the logic, verify actions, and fix issues without affecting any real lead or customer. Before releasing a Flow to the public, it is essential to ensure that every step — messages, questions, conditions, and actions — behaves exactly as planned. Building the visual flow is the first step; proving it works in practice is the next. Fortunately, Papo provides the resources needed for complete validation with a single control contact. Test process overview | Step | Action | What to watch | | :--- | :--- | :--- | | 1. Build the Flow | Design the flow with action, question, and condition nodes, and save it. | The visual logic is correct, with no loose nodes. | | 2. Temporarily activate | Change the Flow status to Active in the Flows module. | Only active Flows process contacts. | | 3. Enroll test contact | In the CRM, open a test contact and choose to manually enroll it in the active Flow. | The contact should be a test profile, not a real customer. | | 4. Monitor execution | Go to the Flow execution history and see each node the contact went through. | Conditions, captured answers, and executed actions. | | 5. Fix and repeat | If you find issues, deactivate the Flow, adjust the logic, and repeat the test. | Until the result is as expected. | How the test works, step by step 1. Build the Flow and save the draft Use the visual editor of the Flows / Chatbot module to build the journey. Add the necessary action nodes (send message, ask a question, add tag, etc.), define branches with conditions, and make sure there are no dead ends. When you finish, save the Flow — it will remain inactive until you decide to activate it. 2. Activate the Flow for testing An inactive Flow does not receive contacts or execute actions. Therefore, to perform the test, you need to change the status to Active. This activation is temporary while you validate operation; you can deactivate the Flow again if you find problems and then reactivate it when it is ready for production. 3. Create or choose a test contact To isolate the test from the real environment, use a contact created specifically for this purpose. Go to CRM, add a new contact with a fictitious name and phone number, for example, “Test Flow” with a number you control (such as your own personal WhatsApp, if you want to receive the messages). This contact will be the automation’s test subject. 4. Manually enroll the contact in the Flow With the Flow active, open the test contact’s profile in the CRM. Find the option to enroll it in a Flow and select the automation you want to test. This manual enrollment inserts the contact at the beginning of the journey, exactly as would happen with a real lead, but in a controlled way. 5. Monitor the execution history Immediately after enrollment, Papo begins processing the contact. Go to the same Flow, look for the execution history section, and find your test contact’s entry. When you open the details, you will see the exact sequence of nodes traversed — each message sent, each question answered, and each condition evaluated. Compare this real path with what you designed in the editor. 6. Analyze the results and iterate If the contact stopped at a condition node where it should not have stopped, check whether the captured answer was recognized correctly (pay attention to variations such as “yes” and “Yes”). If an action such as Add Tag or Update Contact was not executed, confirm in the history whether the node was reached. With the diagnosis in hand, deactivate the Flow, make the necessary adjustments, activate it again, and repeat the test with the same contact (or a new one) until everything flows perfectly. Who this process is for - Administrators creating their first automation and want to avoid surprises with real leads. - Marketing and sales teams that build qualification, survey, or welcome Flows and need to guarantee the customer experience. - Any user who customizes a Flow from the Library and wants to validate the adaptations before releasing it to the public. Frequently asked questions Can I test a Flow without activating it? No. Only Flows with the status Active can receive and process contacts. To test the logic, you need to temporarily activate the Flow. After testing, you can deactivate it without problems. Can my test Flow interfere with real customers? No, as long as you use an exclusive test contact. Until you enroll any real customer in the Flow — whether manually, in bulk, or through another automation — interactions remain restricted to the control profile. What should I do if the test contact does not advance past a node? Check the execution history: the stopping node indicates where the logic failed. It is usually a condition that did not recognize the contact’s answer (for example, the question expects “Yes” and the contact typed lowercase “yes”) or a dead-end path. Fix the condition or add a default route to capture unexpected answers. Do I need to create a new contact for each test? It is not mandatory. You can reuse the same test contact multiple times. Just enroll it again in the Flow (after restarting or adjusting) and monitor the new history. If you prefer to keep separate records, create a new contact for each test round. After testing, can I deactivate the Flow and reactivate it normally? Yes. You can switch the status between Active and Inactive at any time. After final adjustments, keep the Flow active to start receiving enrollments from real contacts. Does the execution history show the answers the contact sent? Yes. When inspecting an Ask a Question node in the history, you can view the exact input provided by the contact. This is essential to validate whether the correct branch was followed. See also - How to debug an automation Flow that is not working - What is the difference between Flow and Sequence in Papo? Start testing for free now → The first user is free and lets you validate automation Flows with no commitment, even before inviting your team.

How to list all Flows in your account and manually enroll contacts

To list all Flows in your account and manually enroll contacts in Papo, go to the Flows module. The platform shows the list of available flows and sequences with identifier, name, and type. After checking whether the flow is active, you can insert a specific contact at the beginning of the journey. Process summary | Step | What to do | Result | | :--- | :--- | :--- | | List the flows | Open the Flows module and view the available flows | You find the right flow by identifier, name, and type | | Check the status | Check whether the flow is Active or Inactive | Only active flows receive and process contacts | | Enroll the contact | Select the contact and insert them at the beginning of the flow | The contact starts moving through the designed journey | What you find in the Flows list In Papo's Flows module, the list shows the flows and sequences available in the account. For each item, you can identify three basic pieces of information: - Identifier: distinguishes one flow from the others. - Name: makes it easier to find the journey, for example, screening, onboarding, or follow-up. - Type: indicates whether the item is a generic flow or a sequence. This view lets you quickly find the flow you want to review, activate, or use to enroll a contact. Flows created from scratch or installed from the Flow Library may start as empty or inactive drafts; they only take effect after being saved and activated. Flow status: what changes between active and inactive The Flows list also shows the status of each journey. This status defines whether the flow is cleared to run: - Active: the flow can receive new contacts and process the configured steps. - Inactive: the flow is not cleared to receive or process contacts. It is the paused state of the journey. The main rule is simple: only Flows with Active status can receive and process contacts. If you try to enroll someone in an inactive flow, the journey does not advance. So, before any manual enrollment, confirm that the destination flow is active. How manual contact enrollment works An active Flow does nothing until contacts are enrolled in it. Enrollment is the act of inserting a contact at the beginning of the journey you designed. From that point on, the contact starts moving through the flow steps, such as messages, questions, conditions, and transfers. Manual enrollment is useful when you want to include a specific person in an existing flow without relying on an automatic trigger. For example: - A lead asked to enter a nurture sequence and you want to include them now. - A customer responded to an offer and needs to enter the welcome flow. - A contact was left out of a screening and should be added manually. In these cases, the action is individual: you select the contact and insert them into the desired flow. The platform preserves the contact's conversation history in the CRM; enrollment only defines the start of the new journey. When manual enrollment makes sense - Specific follow-up: a hot lead needs to enter a follow-up sequence without waiting for the next mass send. - Flow correction: a contact was not included by the original criterion and should now be handled. - Agreed service: the team spoke with the lead through another channel and wants to formalize the journey in Papo. - Controlled tests: you want to validate a flow with a real contact before releasing it to everyone. In all these scenarios, manual enrollment places the contact at the starting point of the flow, as long as the flow is active and has its steps configured. Frequently asked questions How do I know if a Flow is active? In the Flows list, the status appears next to each flow or sequence. The value Active indicates that the flow is cleared to receive and process contacts; Inactive indicates that the journey is paused. Can I enroll a contact in an inactive Flow? No. Only flows with Active status can receive and process contacts. Before enrolling, check that the destination flow is active; if it is inactive, activate it first. Does manual enrollment erase the contact's history? No. Enrollment inserts the contact at the beginning of the journey, but the conversation history and profile data in the CRM remain intact. What is the difference between listing Flows and listing Sequences? A Sequence is a specific type of Flow. In the list, the type field helps differentiate generic flows from continuous follow-up sequences. Manual enrollment works in both cases, as long as the item is active. Where do I find templates I haven't installed yet? Templates are in the Flow Library. When you install a template, it enters your account as an inactive flow, ready for review, activation, and contact enrollment. Read also - What is the difference between a Flow and a Sequence in Papo? - How to test an automation Flow before activating it Start for free and set up your Flows → Create your free account and use the first user at no cost to validate your Flows.

How to list all Sequences and track contact progress

To list all Sequences and track contact progress in Papo, go to the Flows module. The platform shows the list of available sequences, and from the contact profile you can see which cadence they are enrolled in, as well as pause or cancel future messages to that person. Managing follow-up sequences and automated messages is essential to avoid losing leads or annoying those who have already replied. Instead of looking for contacts one by one, you start from the overview of active cadences or the contact profile to make quick decisions. Summary of actions to list and manage sequences | Action | Where to do it | Result | | :--- | :--- | :--- | | List all Sequences | Flows module, filtering by Sequence type | You see the name, identifier, and status (Active/Inactive) of each cadence | | View a contact's progress | Contact profile in the Contacts module | See which active Sequences the contact is enrolled in | | Pause/remove a contact | Contact profile, in the active Sequences section | Stops future sends to this contact without affecting others | How to access and interpret the list of Sequences Sequences are part of Papo's Flows module. That's where all automated journeys — including follow-up cadences — are centralized. 1. Go to the Flows module in the platform panel. 2. In the list, identify entries whose type is Sequence. This distinguishes scheduled message cadences from other conversation flows. 3. For each sequence, the list shows: - Identifier: a unique code that identifies the cadence. - Name: the title you gave the sequence, such as "Post-quote follow-up" or "Trial welcome". - Status: indicates whether the sequence is Active (sending messages to enrolled contacts) or Inactive (paused for everyone). With this view, you can quickly check which cadences are in operation. An inactive sequence does not process any contacts; for it to work again, you need to activate it. Tracking contact progress in a Sequence To find out which stage a contact is at, the most direct path is their profile: 1. Go to the Contacts module and search by name or WhatsApp number. 2. Open the profile. The platform displays the Sequences or Flows the person is enrolled in. On this screen, you see the sequence name and participation status. If the contact is active in the cadence, the next scheduled messages will be sent according to the schedule. If you need to stop sending, the action is done from this same location. Removing or pausing a contact from a Sequence Within the contact profile, in the list of active sequences, you can remove the enrollment. This is equivalent to pausing the sequence for that specific contact. - What happens: future messages from the cadence stop being sent to that person. - What remains: the entire conversation history and messages already sent remain saved in the CRM. Other contacts continue receiving messages normally. - If you want to resume: since the enrollment is canceled, to send messages again you need to enroll the contact again in the sequence (or in a new one), starting the journey from the beginning. This removal is the fastest way to act when a lead has replied, asked for time, or no longer makes sense to continue in the cadence. Who this view is most useful for - SDRs and salespeople who manage multiple follow-up cadences and need to pause a contact as soon as they reply or schedule a meeting. - Operations managers who want to monitor which sequences are active and how many contacts are in each one. - Customer Success teams who use sequences for nurturing and onboarding and may need to stop sends during a parallel negotiation. Frequently asked questions How do I know a Sequence is actually working? In the Flows module, the Active status next to the sequence indicates it is enabled to send messages. If there are enrolled contacts, sends occur according to the programmed intervals. If it is Inactive, no sends are made. Can I see all contacts in a Sequence at once? The list in the Flows module shows the cadence's identifier, name, and status. To check specific contacts, the path is to check each profile or use conversion and campaign reports, if the sequence is linked to a sales funnel in Kanban. When I remove a contact from a Sequence, do they lose messages already received? No. Messages that have already been sent appear in the conversation history in the contact profile. Removal only blocks future sends scheduled for that enrollment. What is the difference between removing a contact and pausing the entire Sequence? Removing the contact stops sends only for that person, keeping other leads in the cadence. Pausing the entire Sequence deactivates the flow for all contacts at once. Is there a way to pause temporarily without canceling the enrollment? Yes. An alternative with custom attributes and conditions in the Flow allows you to pause and resume the cadence from where it stopped. Details on how to build this logic are in the article about temporary pause in the Help Center. Where can I see if a contact completed the Sequence? If the Sequence was completed, the participation no longer appears as active in the profile. For completion analyses, the platform reports and Kanban cards associated with the journey provide a complete view of lead progress. Read also - How to list all Flows in your account and manually enroll contacts - How to pause a message Sequence for a contact Start free with Papo → Create your follow-up sequences, enroll contacts, and manage progress — all in the same panel, at no initial cost.

How to duplicate an automation Flow to create a new version

To duplicate an automation Flow in Papo (https://papo.global/) and create a new version, go to the Flows module, find the original flow, and use the duplicate function. The copy preserves all nodes, connections, and settings, allowing you to adjust only what is needed for a new scenario — such as a different campaign, a lead segment, or a service channel. Summary: creating from scratch vs. duplicating a Flow | Method | Effort | When to use | | :--- | :--- | :--- | | Create a Flow from scratch | High — you need to build the entire structure of nodes, conditions, and actions | Completely new flows with no similarity to existing ones | | Duplicate an existing Flow | Low — the structure is copied and you only adjust what changes | Variations of the same journey (e.g., different products, audiences, or channels) | How Flow duplication works Duplication creates an independent copy of the selected Flow. The new flow inherits all components from the original — action nodes, conditions, messages, tags, and branches — but it does not carry enrolled contacts or execution history. From the copy, you can rename, edit, or remove any element without affecting the source Flow. This feature is especially useful when you already have a validated automation and need to replicate it for another context. For example, a welcome Flow for Instagram leads can be duplicated and adapted for WhatsApp leads, changing only the initial message and the tag applied. Step by step to duplicate a Flow 1. Go to the Flows module In the Papo dashboard side menu, click Flows / Chatbot. This is the hub where all your visual automations are listed. 2. Find the Flow you want to duplicate In the Flows list, find the flow that will serve as a template. You can use the search or available filters to speed up finding it. 3. Start the duplication Select the duplicate action associated with the Flow. The platform will immediately create a copy, usually with the original name plus a suffix such as “(copy)”. 4. Rename the new Flow Click the copy’s name and set a title that reflects the new purpose. A clear name helps the team quickly identify the automation’s purpose (e.g., “Instagram Welcome” vs. “WhatsApp Welcome”). 5. Edit the nodes as needed Open the duplicated Flow’s visual editor and make adjustments: - Change message texts to suit the new audience or channel. - Modify conditions (e.g., expected responses, checked tags). - Replace actions such as “Add tag” or “Update contact” with the correct values for the new scenario. - Remove or add nodes to include specific steps of the new journey. 6. Activate the Flow After reviewing the changes, save and activate the flow. Only Flows with Active status can receive and process contacts. Who Flow duplication is recommended for - Marketing and sales teams managing multiple campaigns with similar structures (e.g., product launches, seasonal promotions). - Multichannel operations that want to replicate a successful journey across different channels (WhatsApp, Instagram, email) with minor adaptations. - Agencies and partners that set up automations for multiple clients and need to reuse validated flows as a starting point. - Any team that wants to test variations of a Flow without risking breaking the original version in production. Frequently asked questions Does duplication also copy already enrolled contacts? No. The duplicated Flow is created with no enrolled contacts. You will need to enroll the desired contacts manually, in bulk, or via another automation. Can I duplicate a Flow that is inactive? Yes. Duplication works regardless of the original Flow’s status. You can duplicate both active and inactive flows. Do changes I make to the original Flow affect the copy? No. After duplication, the two Flows are completely independent. Any edit to one is not reflected in the other. Can I duplicate the same Flow multiple times? Yes. You can generate as many copies as you want, as long as you respect your plan’s Flow limit. Each duplication counts as a new Flow in your account. What happens to webhooks or integrations configured in the original Flow? Webhook settings and actions that call external services are copied along with the nodes. However, it is recommended to review the URLs and parameters in the duplicated Flow to ensure they point to the correct endpoints for the new scenario. Does duplication consume any account limit? The duplication action itself has no additional cost, but the created Flow takes up a slot in your plan’s Flow limit. Check the amount available in your account. Read also - How to list all Flows in your account and manually enroll contacts - How to test an automation Flow before activating it Start for free and create your automation Flows → With Papo, you duplicate and adapt service journeys in minutes, without having to start from scratch.

How to configure Capitão to check order status in real time

To configure Papo's Capitão (https://papo.global/) to check order status in real time, use a custom Tool via HTTP. It connects the assistant to your ERP, e-commerce, or CRM endpoint and lets the AI interpret the customer's message, send the order number to the external system, and return the status during the WhatsApp conversation. With this setup, Capitão is no longer just an assistant that answers questions and starts performing practical actions. The customer can send “where is my order 12345?” and receive an automatic reply, such as “Order in transit,” without human intervention. This reduces queues, speeds up service, and frees the team for complex cases. Configuration summary | Stage | What to do | Note | | :--- | :--- | :--- | | 1. Access Capitão | Open the assistant settings in the Papo platform. | Exact menu names may vary. | | 2. Create a custom Tool | Use the HTTP type to connect Capitão to an external system with an API. | Requires technical knowledge of APIs and HTTP requests. | | 3. Define the endpoint | Enter the API URL of your ERP, e-commerce, or CRM. | Example: https://api.meusistema.com/v1/pedidos. | | 4. Configure method and authentication | Use GET to fetch data and include security headers, such as an authentication token. | Example: Authorization: Bearer sua_chave_secreta. | | 5. Send the order number | Allow Capitão to use the information provided by the customer in the conversation. | The action becomes dynamic according to the order provided. | | 6. Link to WhatsApp | Activate the assistant in the WhatsApp inbox. | The lookup starts working during support. | How order status lookup works with Capitão The flow is a cycle that connects the customer's message to a concrete action: 1. The customer asks about the order: Sends something like “Where is my order 12345?”. 2. Capitão understands the intent: Based on the instructions and descriptions of the configured tools, it identifies that it should check the status. 3. The Tool is triggered: The assistant sends the number “12345” to your external system endpoint. 4. The system returns the status: The API returns a result, such as “Order in transit” or “Order delivered.” 5. Capitão replies to the customer: The AI uses this return to formulate a natural and complete response on WhatsApp. All this happens in real time, without a human agent needing to copy the order number and open another system. Prerequisites for using a custom HTTP Tool Before configuring, it's important to have: - A Capitão assistant created on the Papo platform, with a name and description defined. - An external system with an available API, such as an ERP, CRM, or e-commerce platform. - The endpoint URL that will be called to check the order. - Authentication credentials, such as an API key or access token, to keep the request secure. - Knowledge of APIs and HTTP requests, since configuring custom Tools is advanced. If you don't have an API in your system or don't know what data it accepts, the technical details should be validated with the development team or the vendor's documentation. Native vs. custom Tools for orders Papo offers two types of Tools for Capitão: | Type | Description | Example use | Technical knowledge | | :--- | :--- | :--- | :--- | | Native | Preconfigured integrations ready to use on the platform. | Schedule a demo in Google Calendar. | Low, just activate and connect the account. | | Custom (HTTP) | Custom actions that connect to external systems or Papo's API. | Check the status of an order in the company's e-commerce. | Advanced, requires APIs and HTTP requests. | For checking order status in real time, the most common path is the custom HTTP Tool, since the order is usually in the company's own system. There is no automatic native integration for any ERP or e-commerce; each system needs to be connected through the corresponding endpoint. Configuring the custom Tool for order status The conceptual process to enable order lookup follows a clear logic: 1. Access the assistant settings: In the Papo panel, open the Capitão configuration area where Tools are managed. 2. Create a custom HTTP Tool: Choose the option that allows connecting to an external system via API. 3. Fill in the request data: Enter the endpoint URL, the HTTP method (GET to fetch data), and the authentication headers, such as Authorization: Bearer sua_chave_secreta. 4. Define how the order number will be sent: The Tool can use data from the conversation to make the action dynamic. Thus, when the customer provides order 12345, Capitão sends this value in the call. 5. Clearly describe when the Tool should be used: Capitão selects the appropriate tool based on the instructions and descriptions you configured. Include examples of messages, such as “check status,” “where is my order,” or “track order.” 6. Link the assistant to an inbox: Activate Capitão in the WhatsApp inbox or another channel connected to Papo. From then on, it will respond to status requests within the conversation. Exact screen and button names may vary within the platform. If you have questions about where a field is located, consult the Help Center or Papo support. When this automation makes sense Configuring real-time order status lookup is useful for operations that: - Receive many repeated questions about orders on WhatsApp. - Sell through e-commerce or work with delivery deadlines. - Want to reduce first response time in post-sale conversations. - Need to free the team for cases that require human analysis, such as exchanges, refunds, or complaints. - Already have an external system with an order lookup API. If your operation doesn't yet have an order API, full customization may not be possible immediately. In that case, it's worth evaluating whether a native Papo Tool covers another stage of support, such as scheduling a demo in Google Calendar. Frequently asked questions Do I need to know how to code to configure order status lookup? Yes, for custom HTTP Tools, advanced knowledge of APIs and HTTP requests is required. Native Tools have simpler configuration, but they don't automatically cover all external systems. Can Capitão check more than one system? Yes. You can configure multiple custom Tools so Capitão can communicate with different APIs, such as ERP, CRM, or e-commerce platform. How can I ensure the lookup is secure? Use authentication headers, such as an access token, so the request is secure and authorized. For example, Authorization: Bearer sua_chave_secreta. Do not share keys publicly, and validate access only to the necessary endpoint. Can I test before activating it on WhatsApp? Yes. Ideally, validate the API call with an external testing tool before activating the assistant in the inbox, avoiding incorrect responses to the customer. What happens if the external system returns an error? Capitão can only respond based on the API return. If the endpoint is down or returns an error, the response may not contain the status. That's why it's important to monitor the availability of the integrated system. Read also: - How to train Capitão to identify purchase intent and transfer to a salesperson - How to test a Papo webhook connection before activating automation Start for free and configure order lookup with Capitão → Centralize WhatsApp, Instagram, email, and SMS in one inbox with Papo's AI replying in 2 seconds.

How to create a Capitão Tool to fetch freight quotes

To create a Capitão Tool that fetches freight quotes, use a custom HTTP Tool on the Papo platform (https://papo.global/). It connects the assistant to your carrier's or logistics platform's API and allows the AI to interpret data such as destination postal code (CEP), weight, and dimensions sent by the customer, query the external system, and automatically return the freight cost during the WhatsApp conversation. With this setup, Capitão is no longer just an assistant that answers questions and starts performing practical logistics actions. The customer can ask “How much is freight to CEP 02030‑040?” and immediately receive a response such as “Freight to this CEP is R$ 18.90 with a delivery time of 3 business days,” without human intervention. This reduces purchase friction, speeds up service, and frees the team for more complex negotiations. Configuration summary | Step | What to do | Note | | :--- | :--- | :--- | | 1. Access Capitão settings | Open the assistant settings in Papo. | Exact menu names may vary. | | 2. Create custom Tool | Use the HTTP type to connect Capitão to a carrier API. | Requires technical knowledge of HTTP requests. | | 3. Define the endpoint | Enter the URL of the API that calculates freight (e.g., Correios, Braspress, Intelipost). | Example: https://api.transportadora.com/v1/cotacao. | | 4. Configure method and authentication | Use POST or GET with security headers, such as an access token. | Example: Authorization: Bearer your_api_key. | | 5. Send quote data | Allow Capitão to extract from the conversation the postal code (CEP), weight, format, or other parameters required by the API. | The action becomes dynamic according to what the customer provides. | | 6. Connect to WhatsApp | Enable the assistant in the WhatsApp inbox. | The quote starts working in real time during service. | How freight quoting works with Capitão The flow connects a common customer question to a real-time calculation: 1. The customer requests a quote: Sends a message such as “What is the freight to CEP 04123‑000?”. 2. Capitão interprets the intent: Based on the instructions and Tool descriptions, it recognizes that it should use the quote Tool. 3. The Tool is triggered: The assistant sends the values extracted from the conversation (CEP, product weight, etc.) to the carrier's endpoint. 4. The API returns the cost and delivery time: The external system returns a JSON with the available freight options. 5. Capitão responds to the customer: The AI translates this data into a natural response, such as “We have PAC for R$ 22.90 (7 days) and Sedex for R$ 39.50 (2 days).” The entire process happens without a human agent needing to copy the CEP and open another system. Prerequisites for setting up the quote Tool Before creating the Tool, you need to have: - A Capitão assistant already created on the Papo platform, with a defined name and description. - An available carrier API, whether from Correios, a logistics platform (Intelipost, Frenet, SuperFrete), or your own delivery management system. - The endpoint URL that will be called to calculate freight. - Authentication credentials (API key, token, or specific header) so the request is secure. - Knowledge of APIs and HTTP requests, as custom Tool configuration is advanced. If your carrier does not offer a public API, you will need to develop an intermediary or use a quoting solution that already exposes an endpoint. Technical specifications must be validated with the development team or the provider's documentation. Native vs. custom Tools for logistics Papo offers two types of Tools for Capitão: | Type | Description | Use example | Technical knowledge | | :--- | :--- | :--- | :--- | | Native | Preconfigured integrations ready to use on the platform. | Schedule a demonstration in Google Calendar. | Low, just enable and connect the account. | | Custom (HTTP) | Custom actions that connect to external systems via API. | Check real-time freight quotes on WhatsApp. | Advanced, requires APIs and HTTP requests. | To fetch freight quotes, the recommended path is the custom HTTP Tool, because the calculation depends on each carrier or logistics operator. There is no generic native integration for any carrier; you will need the specific endpoint of the quoting service you use. Setting up the custom Tool for freight quoting The conceptual process for enabling this automation follows clear logic, without needing to memorize screen or button names: 1. Access the assistant settings: In the Papo panel, go to the Capitão configuration area where Tools are managed. 2. Create a custom HTTP Tool: Choose the option that allows connecting to an external system via API. 3. Provide the request data: Fill in the carrier endpoint URL, the HTTP method (usually POST, because you send the CEP and receive the calculation), and authentication headers, such as Authorization: Bearer your_key. 4. Define dynamic parameters: The big advantage is that Capitão can use information the customer provides in the conversation. You can configure the Tool to send the CEP that appeared in the message, the product's default weight, or the package format. The Tool will interpret these values automatically. 5. Instruct Capitão on how to respond: In the Tool description or the assistant guidelines, you can guide the AI to present the cost and delivery time in a friendly way, such as “Your freight to CEP … costs R$ … and arrives in …”. 6. Connect the assistant to the inbox: Make sure Capitão is enabled for WhatsApp (or another channel) where customers will ask the question. After configuration, any message that mentions “freight,” “quote,” or “shipping” will trigger the flow—as long as Capitão recognizes the intent. The final response comes from the carrier's real data, without an employee needing to intervene. Frequently asked questions Do I need to have an API from my carrier to use this Tool? Yes. The custom HTTP Tool works by calling an endpoint that returns the freight cost. If your carrier does not offer an API, you can use a quoting service (such as Intelipost, Frenet, or SuperFrete) that consolidates multiple carriers and provides a single endpoint. How does Capitão know which CEP or weight to use? During configuration, you can indicate that the CEP value should be extracted directly from the customer's message. The Tool uses dynamic variables to capture this data. The assistant can also confirm with the customer if the information is incomplete, depending on the instructions you provide. Is a developer required to configure it? Yes. Creating a custom HTTP Tool requires knowledge of APIs, HTTP methods, and authentication. Usually, a developer or someone experienced with integrations performs this step. Papo does not provide the carrier endpoint, only the infrastructure to connect it. Is the freight quote recorded in Papo's CRM? The conversation history and the generated response are saved in the inbox. Papo does not automatically create a quote record in a specific module (such as a deal in Kanban), but you can complement the flow using other automations, such as webhooks, to create a lead or task from this interaction. Can I use this same Tool to check delivery times? Yes. Usually the freight API response already includes the estimated delivery time. The Tool can be configured to display both the cost and the delivery time in Capitão's response. Is there an additional cost for using custom Tools? Custom Tools are part of Capitão's features for accounts that have access to this resource. For details on availability by plan, consult the Papo website. Read also - How to configure Capitão to check order status in real time - How to generate a public key and a secret key for the Papo API Start for free and create your assistant with smart tools → Centralize customer service and automations on WhatsApp with AI, without multiplying tools.

How to configure AI to automatically mark conversations as “Resolved” based on keywords

To configure the automation that marks conversations as “Resolved” based on keywords in Papo (https://papo.global/), create an Automation Rule in the Automation module, set the incoming message trigger, add the keyword condition (e.g., “thanks”, “resolved”), and configure the actions to apply the “Resolved” tag and archive the conversation. This keeps the inbox clean and prevents the team from having to manually close every conversation that has already ended. Summary: manual marking vs. keyword automation | Method | How it works | When to use | | :--- | :--- | :--- | | Manual marking | The agent right-clicks the conversation and selects the option to mark it as resolved. | Low volume, cases requiring human judgment, or exceptions. | | Automation Rule by keyword | The system identifies configured terms in the message and executes automatic actions, such as applying a tag and archiving. | Standardized interactions where expressions like “thanks” indicate closure. | Papo keeps conversation history and tags, so status remains traceable even after archiving. How the keyword rule works In the Automation module, rules follow trigger + conditions + actions logic. In the keyword-closing scenario, the trigger identifies a new message; the condition compares the text with configured terms; the actions perform tagging and archiving. You can restrict the rule to an inbox, a current tag, or other criteria, reducing false positives. For example, the rule can apply only to Support WhatsApp conversations that do not yet have the “Resolved” tag. Step-by-step setup 1. Go to the Automation module In the Papo dashboard, go to Settings and click Automation. This is where you manage the rules that automate service actions. 2. Create a new rule Start a new rule and give it a descriptive name, such as “Mark as Resolved by keyword”. This makes it easier to identify and adjust the automation later. 3. Set the message trigger Choose the event that triggers the automation when a new message arrives in the conversation. If the dashboard presents separate triggers by channel or message type, select the one that matches your flow. 4. Configure the keyword condition In the condition, specify that the message content must contain one of the closing terms: “thanks”, “thank you”, “resolved”, “closed”. Add variations with and without accents to prevent the rule from missing messages written differently. 5. Add the tag and archive actions Configure the action to apply the “Resolved” tag and the action to archive the conversation. If your account interface displays only one action per rule, you can chain complementary rules or review the options available in your dashboard version; current details are on the Papo website. 6. Save and activate Review the settings, save, and make sure the rule is active. From that point on, conversations that meet the conditions will be tagged and archived automatically. Practical examples of use | Example | Suggested keywords | Recommended action | | :--- | :--- | :--- | | Support closed | “thanks”, “resolved”, “all good” | Apply “Resolved” and archive | | Post-sale confirmed | “received”, “installed” | Apply “Customer confirmed” and keep follow-up | | Formal email resolved | “I appreciate it”, “it's resolved” | Apply “Resolved” and archive | Choose terms that appear naturally at the end of the interaction. Very generic words increase the risk of closing conversations that are still active. Who this automation is for - High-volume support teams: eliminates the repetitive task of reviewing and closing conversations one by one. - Operations that already use standardized tags: keeps classification consistent without depending on each agent remembering the right tag. - Companies with multiple channels: you can create one rule per channel, respecting the vocabulary and pace of each inbox. - Teams that want to reduce administrative tasks: after automatic closing, the agent is free for the next conversation. FAQ Can the rule apply the tag and archive in the same conversation? Yes, you can configure both actions in the same rule. If your account presents the actions separately, organize the rule to apply the tag before archiving. Can I use more than one keyword? Yes. Add the terms that indicate closure and variations, such as “thanks”, “thank you”, “resolved”, and “closed”. A more specific list reduces the chance of marking conversations that are still active. Does the rule work for old messages or only new ones? The automation is triggered by new messages or configured events. For conversations already open, use manual marking in the conversation dashboard. What happens if the customer writes again after archiving? The conversation is automatically reopened and appears again in the inbox, preserving the previous history. Can I limit the rule to a channel or tag? Yes. You can add conditions such as source inbox or current tag, so the rule only acts in specific contexts. Do I need to have the “Resolved” tag created beforehand? Papo allows you to create the tag in the action if it does not yet exist. To maintain a standard, it is worth reviewing the tag list and the team's flow before activating the rule. Read also - How to use an Automation Rule to close inactive conversations and send a CSAT survey - How to use tags to signal priority in service By automating the identification of closure by keywords, you turn the end of the conversation into a predictable step: tag applied, inbox organized, and team free for the next customer. Adjust the terms and conditions according to your audience's vocabulary and monitor reports to validate whether the rule is classifying correctly. Start for free and automate conversation closing → Create your Papo account and configure automation rules to tag and archive conversations based on keywords, keeping your inbox organized.

How to Customize Capitão's Tone of Voice for Different Audiences

To customize the tone of voice of Papo's Capitão assistant (https://papo.global/) for different audiences, you use the Prompt and AI module to define rules for formality, empathy, and vocabulary, creating specific agents or instructions that adapt responses to each customer profile. Capitão is the artificial intelligence assistant that automates service on your channels, but the tone he uses to communicate can make all the difference in the customer experience. While the knowledge base (documents, FAQs, and scenarios) defines what the assistant answers, the Prompt and AI module controls how it answers — from more technical and corporate language to a relaxed, approachable style. Summary of approaches to customize tone | Approach | How to do it | When to use | | :--- | :--- | :--- | | Adjust the main agent | Edit the general rules (general_rules) of the default agent with tone instructions. | When a single tone serves most customers, with minor variations. | | Create sub-assistants | Create secondary agents with their own rules and vocabulary and link them to specific inboxes. | To serve very different audiences (e.g., formal B2B and young B2C) on separate channels. | | Combine rules and context | Add conditional instructions to the prompt based on tags, channel, or customer profile. | When the same channel receives varied profiles and the assistant needs to adapt tone dynamically. | How the Prompt and AI module controls tone of voice The Prompt and AI module is the brain that defines Capitão's behavior during conversations. In it, you configure the general rules (general_rules) and AI settings (ai_settings) that determine the assistant's communication style. These rules are applied to all responses generated by Capitão, regardless of the knowledge base it consults. For example, you can include instructions such as: - “Use formal language, avoid slang, and address the customer as ‘sir’ or ‘madam’.” - “Be empathetic and welcoming, showing understanding before offering a solution.” - “Adopt a light tone and use emojis occasionally for a young audience.” In addition to the main agent, you can create sub-assistants (secondary agents) with their own rules. Each sub-assistant can be linked to a different inbox, allowing the sales WhatsApp to have a more commercial tone, while the support website chat maintains a more technical and patient tone. Step by step to define Capitão's tone of voice 1. Access the Prompt and AI module In the Papo platform, find the Prompt and AI module. This is where you manage agents and their behavior settings. 2. Choose the agent you want to customize You can edit the main agent (which serves the account as a whole) or create an inactive sub-assistant as a draft to test a new tone before activating it. Sub-assistants are useful for isolating audiences without affecting general service. 3. Edit the general rules (general_rules) General rules are the field where you enter natural-language instructions about tone of voice. Write clearly and directly, as if you were guiding a human agent. Practical examples: - For a corporate audience: “Maintain a professional and objective tone. Use ‘Dear customer’ in the greeting and avoid abbreviations.” - For a young audience: “Be friendly and use casual language. You can use ‘Hi!’, ‘Thanks!’ and emojis like 😊 when appropriate.” - For empathetic service: “Always show empathy first. If the customer reports a problem, start with ‘I understand how frustrating this can be’ before proposing a solution.” 4. Adjust the AI settings (ai_settings), if necessary In addition to text rules, the module lets you refine behavior with AI parameters. Although the technical details vary, you can use this space to control the level of creativity or adherence to rules, ensuring the tone does not deviate from what is expected. 5. Apply the prompt and test After defining the rules, apply the prompt so the agent starts using them. Send a few test messages in the linked inbox to check whether Capitão is responding with the desired tone. If you need adjustments, simply edit the rules again. 6. Link the agent to the correct inbox If you created a sub-assistant for a specific audience, associate it with the corresponding inbox (for example, the WhatsApp number used for B2B sales). This way, the customized tone will be automatically applied to all conversations on that channel. Who this customization is useful for - B2B and B2C companies operating together, needing a formal tone for corporate clients and a more relaxed one for end consumers. - Brands present on multiple channels, where Instagram calls for more visual and casual language, while email requires more formality. - Businesses serving different age groups, adapting vocabulary and empathy according to the customer profile. - Support teams that want to humanize service, injecting warmth and patience during complaints. Frequently asked questions Can I have a different tone of voice for each channel? Yes. Create a sub-assistant for each channel (WhatsApp, Instagram, website chat) with specific tone rules and link each one to its respective inbox. Capitão will automatically adopt the tone configured for that channel. Can Capitão automatically adapt the tone based on the customer profile? The assistant follows the rules you defined. If you include conditional instructions in the prompt — for example, “if the customer uses formal language, respond at the same level” — Capitão can adjust the tone based on the conversation context. However, fully automatic adaptation without explicit rules is not guaranteed. Do I need to create a separate assistant for each tone? Not necessarily. You can keep a single main agent with rules that cover most situations. Creating sub-assistants is recommended when audiences are very different and you want to completely isolate behaviors. Do tone rules affect responses based on FAQs and documents? Yes. Prompt and AI rules are applied to any response generated by Capitão, whether it is extracted from a document, an FAQ, or a scenario. The content remains the same, but the packaging (tone, formality, empathy) follows your instructions. How can I test whether the tone is appropriate before activating it for real customers? You can create an inactive sub-assistant (draft) and temporarily link it to a test inbox. Send messages simulating different customer profiles and evaluate the responses. Adjust the rules until you achieve the desired tone, and only then activate the agent for the real audience. Can I use the same Capitão tone of voice in Disparador campaigns? The Prompt and AI module controls the assistant's behavior in interactive conversations. For bulk message campaigns, the tone is defined directly in the message content you create in Disparador. The agent rules do not apply to these sends. Read also - How to train Capitão to answer your company's frequently asked questions - How to configure AI to automatically mark conversations as “Resolved” based on keywords Start now for free → With the first free user, you can already configure Capitão's tone of voice and test different service styles.

How to create a Flow with a condition based on a contact's custom attribute

To create a Flow that automatically branches a contact's journey based on custom attributes — such as contracted plan, city, or interest — use the Papo Flow editor and insert a condition node configured to check “Contact attribute.” Based on the values stored in the CRM, the flow can send specific messages, add tags, or route the lead to the right team, all automatically and without manual intervention. With this automation, each contact receives a unique treatment based on information your company already has. This increases communication relevance, improves customer experience, and optimizes the work of marketing, sales, and support teams. Flow summary with attribute condition | Element | Function | Example | | :--- | :--- | :--- | | Flow (Flows/Chatbot) | Visual sequence that automates interactions | “Onboarding by plan” Flow | | Custom attribute | Extra field in the contact profile (text, number, date, or list) | plan (Basic, Pro, Enterprise), city (SP, RJ, BH) | | “Contact attribute” condition node | Checks an attribute value to decide the path | If plan equals “Pro,” follow the special offer branch | | Actions per branch | What happens in each condition result | Send WhatsApp template, apply “VIP” tag, notify manager | How it works: step by step 1. Prepare the custom attribute Before creating the Flow, the data you want to use must be available in the CRM. Go to Settings → Custom Attributes and create fields such as “Contracted Plan” (List type), “City” (Text), or “Last Purchase” (Date). Then populate the contact values — manually, by import, or via integrations (API, webhooks). 2. Create the Flow with the right trigger In Papo's side menu, click Flows / Chatbot and add a new flow. Replace the default trigger with the event that will start the journey, such as “Contact created,” “Message received,” “Tag added,” or another available option. 3. Insert the condition node After the trigger, click the “+” icon and choose the condition option. Inside the condition, select the “Contact attribute” type. In the configuration, choose the desired custom attribute and define the rule: “equals,” “contains,” “is greater than,” depending on the field type. You can combine more than one rule using “AND” and “OR” operators. 4. Create the journey branches For each value or range of values you want to treat differently, configure a branch. Practical example: - If plan is “Pro” → path A (premium customer) - If plan is “Basic” → path B (standard customer) - If the attribute is empty → fallback path to request a profile update 5. Add actions to each branch In each branch, insert the appropriate action nodes: Send Message (WhatsApp, email), Add tag, Create Kanban task, Send internal notification, etc. Personalize the messages using contact variables, including the attribute itself (e.g., {{contact.city}}). 6. Activate the Flow Review the entire logic, save, and change the status to Active. The flow will process all contacts that meet the trigger, branching instantly based on the actual values of the CRM attributes. Whenever a branch includes a pause longer than WhatsApp's 24-hour window, the next message must be a Message Template (HSM) approved by Meta, which Papo lets you select directly in the send node. Who is this strategy for? - SaaS or subscription companies that want different onboardings according to the contracted plan. - E-commerce businesses that segment offers by city, interest, or customer spending range. - Educational institutions that personalize communications by course or student campus. - B2B sales teams that prioritize leads from a certain region or with a certain job title in the CRM. - Any operation that already collects structured data and wants to turn it into automatic, relevant communication. Frequently asked questions Do I need to create the attribute before building the Flow? Yes. The custom attribute must exist in Settings → Custom Attributes before it can be used as a condition. Contacts must also have the field filled in — manually, by import, or via integrations. What comparison types can I use? They depend on the attribute type. For Text, you can use “equals” or “contains.” For Number, “equals,” “greater than,” “less than.” For Date, “is before,” “is after.” List fields let you directly select one of the predefined options. How do I personalize the message in the branch with the attribute value? Inside any action node, use the contact variable corresponding to the attribute (e.g., {{contact.plan}}) in the text field. The system automatically replaces it with each lead's actual value. What if the contact doesn't have the attribute filled in? You can configure a specific branch for “empty value” in the condition. This way, those contacts follow an alternative route, such as receiving a message asking them to complete their profile or being routed to an agent. Can I combine multiple attributes in the same condition? Yes. The condition node supports multiple rules with the logical operators “AND” and “OR,” allowing segmentations such as “city contains São Paulo” AND “plan equals Pro.” Does it work with data updated automatically via integration? Yes. Attributes populated by API, webhook, or any Papo module are reflected in real time in the Flow. When the trigger is fired, the condition will use the contact's most recent value. Read also - How to automatically trigger a Flow when a new contact is created in the CRM - How to list all Flows in your account and manually enroll contacts Start creating Flows with custom attributes → The first Papo user is free and doesn't require a credit card — ideal for testing attribute conditions in your Flows before scaling.

How to set up automatic follow-up for leads who haven't replied in X days

To set up an automation that sends a follow-up when a lead doesn't reply in X days, use a Papo Flow (https://papo.global/) with an inactivity condition and the Send Message action. The flow waits for the defined period, checks whether there was no interaction, and only then triggers the follow-up, allowing you to re-engage silent leads without relying on manual reminders. The logic is simple: the lead enters the automation, the Flow pauses for the set time, and at the end a condition checks whether there was a reply. If there was no interaction, the message action sends the follow-up. If the lead replied, the automation can stop or route to the team. Everything runs inside Papo, with history recorded in the conversation and in the CRM. Automation summary | Step | Papo resource | Configuration | Result | | :--- | :--- | :--- | :--- | | Pause the journey | Wait step in the Flow | Set X days, hours, or minutes | Waits the ideal period before follow-up | | Detect silence | Inactivity condition in the Flow | Check whether the lead has not interacted | Advances only for leads who did not reply | | Re-engage | Send Message action | Text or HSM template | Triggers the automatic follow-up | | Keep the conversation within the rules | Message Template | Required after 24h on WhatsApp | Avoids blocks and keeps delivery valid | How the inactivity follow-up automation works In Papo's Flow builder, you combine two moments: a wait and a check. The Wait step introduces a pause of minutes, hours, or days. Right after, an inactivity condition evaluates whether the contact remained without interacting during that period. Only when the condition is true does the Flow advance to the Send Message action. This sequence is useful for sales cadences, lead nurturing, and reopening conversations after a proposal, quote, or material has been sent. It can also be used to remind customers of deadlines, confirm interest in a negotiation, or follow up with contacts who went unanswered in the funnel. You can start the Flow in different ways, depending on the operation: when a new contact is created in the CRM, when a tag is applied, when the card enters a Kanban stage, or when a rule detects a conversation with no interaction. The important thing is that the automation is activated for the right audience and with valid opt-in. Step by step to set up the follow-up 1. In the Papo dashboard, go to Flows / Chatbot and create a new Flow or edit an existing one. 2. At the point in the journey where the lead should start waiting, add the Wait step. 3. Set the value and time unit: for example, 3 days, 48 hours, or 30 minutes. 4. Add an inactivity condition right after the Wait step. Configure it so the flow continues only if the lead has not interacted during the period. 5. Connect the Send Message action after the condition. Write the follow-up briefly and with context, such as a question about the proposal or a reminder of the next step. 6. If the send happens after WhatsApp's 24-hour window, use a Message Template (HSM) previously approved by Meta. 7. Save the Flow and make sure it is active to process contacts. A variation to keep the inbox organized is to use a rule with the conversation with no interaction trigger. In this setup, the rule detects inactivity, archives the conversation, and the archiving triggers a Flow that performs the pause and sends the follow-up. This option is interesting when the team wants to clear the queue and, at the same time, not lose the opportunity to re-engage the lead. Who this automation is for - Sales teams that receive leads on WhatsApp, Instagram, or other channels and don't want to lose opportunities due to lack of follow-up. - Operations that use Kanban and CRM to track proposals but need an automatic touch when the lead goes silent. - Businesses with high volume and a small team that can't manually remember every follow-up. - Companies that already separate leads by tag, stage, or channel and want standardized re-engagement cadences. Frequently asked questions Is the follow-up canceled if the lead replies during the wait? The safest way is to configure the inactivity condition after the Wait step. This way, if there is a reply, the flow can be ended or routed to human support instead of sending the automatic message. The final behavior depends on the condition you define in the Flow. Do I need to use a Message Template on WhatsApp? Yes, if the follow-up is sent after WhatsApp's 24-hour service window. In that case, the message must be a Message Template (HSM) approved by Meta. For shorter pauses, a regular text message can be used within the allowed window. Can I restrict the automation to a specific group of leads? Yes. You can add conditions in the Flow, such as tag, channel, or funnel stage, so that only leads that meet those criteria receive the follow-up. This avoids messages to contacts outside the campaign context. Can I use this logic for email or other channels? The Flow logic works for automations within Papo, but each channel has its own sending rules. On WhatsApp, the 24-hour window and the use of HSM are central points. For email, delivery follows the account's campaign and sequence settings. Check the platform to validate the rules of the chosen channel. Is the follow-up recorded in the contact history? Yes. Because the automation runs within Papo, the message is linked to the conversation and the contact in the CRM. The team can view the complete history, including when the follow-up was sent and what the lead replied afterward. Do I need to configure anything besides the Flow for the automation to work? After creating and saving the Flow, you need to keep it active. If the automation is started by another event, such as contact creation or an inactivity rule, also check that this trigger is active and has the correct conditions. The specific details of your account can be checked on the platform or in Papo's help center. Read also - How to use an Automation Rule to close inactive conversations and send a CSAT survey - How to trigger a Flow automatically when a new contact is created in the CRM Start for free and automate your follow-ups → Create your free account and use Papo Flows to re-engage silent leads on WhatsApp without manual work.

How to use Bússola to discover business opportunities in your history

Papo (https://papo.global/) offers the Bússola module, an intelligence tool that combines customer radar and semantic search across history to find hot leads, upsell and cross-sell opportunities, and customers at risk of churn — turning past conversations into immediate business actions. Over time, your base of conversations on WhatsApp, Instagram, and other channels becomes a goldmine of information. Without the right tool, phrases like “how much does the Gold plan cost?”, “I’m thinking about canceling,” or “do you also do X?” get lost in the daily volume. Bússola solves this by using artificial intelligence to scan, interpret, and organize these interactions, showing exactly who deserves your attention now and letting you search for any term across your entire chat archive. | Feature | What it does | What it is for | | :--- | :--- | :--- | | Customer Radar | Lists contacts that need attention, ordered from most critical to least. | Prioritize follow-ups, avoid churn, and automatically identify hot leads or dissatisfied customers. | | History Search | Lets you search keywords, phrases, or meanings across all conversations. | Find mentions of products, competitors, purchase intent, or complaints that were hidden in messages. | How customer radar reveals opportunities Radar does not rely on manual searches. It works proactively: Bússola's AI continuously analyzes the content of all conversations that arrive in your unified inbox and highlights the contacts that require action. The list is ordered from most urgent to least critical, helping your team start the day with what really matters. Signals the AI considers include: - Purchase interest – mentions of quotes, higher-tier plans, or questions about advanced features. - Churn risk – phrases like “cancel,” “dissatisfied,” “I’m already talking to another provider.” - Delayed follow-up – leads that asked an important question and have not received a response for a long time. - Recurring issues – complaints that indicate a customer is about to abandon the service. When checking the radar, your sales or customer success team can act before an opportunity goes cold or a problem escalates, contacting the right person at the exact moment. How history search uncovers hidden opportunities While radar shows what is urgent, Bússola's semantic search lets you investigate history with your own questions. Instead of looking for a contact by name, you search the content of conversations — a way to do “data mining” in your own chat archive. The table below suggests targeted searches for each type of opportunity: | Goal | Suggested search terms | Possible action | | :--- | :--- | :--- | | Upsell | “upgrade,” “higher plan,” “I need more,” “feature X” | Approach the customer with a personalized offer for a larger plan. | | Cross-sell | “do you also have,” name of another product, “besides that” | Offer a relevant complementary service. | | Lead reactivation | “quote,” “proposal,” “how much does it cost,” “I’ll think about it” | Follow up with a new condition. | | Churn prevention | “cancel,” “dissatisfied,” “problem,” “doesn’t work” | Start proactive service to retain the customer. | | Competitive intelligence | Competitor name | Adjust your argument and highlight differentiators. | Step by step to perform a strategic search in Bússola Using history search is simple and does not require technical knowledge. The process below helps turn a suspicion into a list of actionable contacts. | Step | Action | Detail | | :--- | :--- | :--- | | 1. Access Bússola | In the Papo panel, locate the Bússola module (intelligence area). | It is the same environment that brings together customer radar and search. | | 2. Choose the feature | In the history search tab, type the keyword or phrase. | Use exact terms or variations the customer would use in the conversation. | | 3. Analyze the results | The system displays a list of conversations where the term was found, with an excerpt of the dialogue. | Click any result to open the full conversation and see the context. | | 4. Act on the opportunity | Based on the information, take the appropriate sales or support action. | You can start a new message, create a task in Kanban, assign it to an agent, or trigger an automatic follow-up. | Bússola is useful for any operation that already accumulates a reasonable volume of interactions and wants to stop losing money in its history. Sales, customer success, and marketing teams can act more precisely because decisions become based on what contacts actually said — not on guesswork. Frequently Asked Questions Does Bússola replace Papo's CRM? No. Bússola complements the CRM. While the CRM organizes contacts, pipelines, and activities, Bússola analyzes the content of conversations to reveal what each contact is saying and how urgent an action is. How does radar decide that a customer is “critical”? The artificial intelligence evaluates signals in messages, such as the tone of the conversation (positive, negative, or neutral), the presence of keywords related to opportunities or problems, and the time since the last interaction. The combination of these factors generates the ordered priority list. Do I need to configure anything for radar to work? No. Radar operates automatically on conversations that are already in your unified inbox. You just need the Bússola module active in your Papo account. Can I search for whole phrases, or only isolated words? Bússola's search is semantic. You can type phrases like “I want to upgrade my plan” or “delivery problem” and the system will find conversations that contain that meaning, not just the exact word. Are radar results updated in real time? Yes. Whenever a new message enters the platform, the AI analyzes it. If it contains a relevant signal, the contact can appear or move up in the radar immediately. Is there a history limit in search? Search covers the entire conversation history available in your account. Specific parameters can be consulted on the website or directly in the platform. Read also - How to use Kanban to manage cross-sell and upsell opportunities - How to set up automatic follow-up for leads who haven't responded in X days Start for free now → The first user on the platform is R$ 0; connect your WhatsApp and activate Bússola to reveal hot leads and churn risks hidden in your history.

How to Train Capitão with Old Call Transcripts to Improve Understanding

To train Capitão, Papo's artificial intelligence assistant (https://papo.global/), with old call transcripts, export the audio from your call history, convert it to text with an external transcription tool, and add the text files as documents to the assistant's Knowledge Base. This process leverages real conversations already handled by your team to teach the AI the terms, questions, and tone your customers actually use, refining Capitão's understanding and increasing the automatic response rate. Summary of the transcript training process | Step | Action | Expected result | | :--- | :--- | :--- | | Export audio | Download recordings from Papo's call history (when available). | Raw material with customers' real voices. | | Transcribe | Convert audio to text using services such as Google Speech-to-Text, Azure, or Whisper. | Searchable text ready for analysis. | | Prepare documents | Organize transcripts into .txt or .pdf files, removing sensitive data. | Clean, anonymous content for the AI. | | Add to Knowledge Base | Upload the documents in Capitão's Documents section. | The assistant starts consulting real speech to formulate responses. | | Create FAQs from patterns | Identify frequent questions in the transcripts and register them as FAQs. | Direct answers to the most common questions that appear in calls. | Why use old call transcripts? The phone conversations your team has already handled are a gold mine for training Capitão. Unlike static documents or FAQs written by a manager, transcripts show how customers really express themselves: the words they use, the questions they repeat, and even the objections they raise before closing a purchase. By feeding this material into the Knowledge Base, you teach the AI to recognize intentions more naturally, reducing the number of messages that need to be transferred to a human agent. This method is especially useful for businesses that already have a large volume of recorded calls and want to accelerate the assistant's learning curve without relying only on assumptions or hypothetical questions. How call history works in Papo Papo's Calls module stores the history and statistics of all calls made through the platform. You can view the list of calls with information such as direction (inbound/outbound), status, and method. If your account has call recording enabled, the audio files are linked to each record and can be exported for external use. If recording is not available, you can still use notes and conversation summaries as a training source, although audio transcription provides the highest level of detail. Step by step to train Capitão with transcripts 1. Export the call audio. Go to the call history in the Papo dashboard and find the conversations you want to use. If recording is enabled, download the audio files to your computer. 2. Convert audio to text. Use an external transcription tool (such as Google Speech-to-Text, Microsoft Azure Speech, or Whisper) to generate the text corresponding to each call. Review the transcript to correct any errors and remove sensitive personal data (document numbers, card numbers, etc.). 3. Format the transcripts as documents. Save each transcript in a .txt or .pdf file. Group conversations by topic (sales, support, billing) if you want to better organize the base. 4. Add the documents to Capitão's Knowledge Base. In the Papo dashboard, go to the settings of the Capitão assistant you want to train. In the Knowledge Base section, upload the transcript files as documents. The platform processes the content and makes it available for the AI to consult. 5. Create FAQs based on the patterns found. Analyze the transcripts and identify the most repeated questions. Register these question-and-answer pairs in Capitão's FAQs area. Example: if several calls ask "What is the deadline for the boleto to become due?", create an FAQ with that question and the correct answer. 6. Track improvement in reports. After a few days, check the service reports to see whether Capitão's automatic response rate has increased and whether the number of transfers to humans has decreased. Adjust the documents and FAQs as needed. Who this training is recommended for - Companies that already have a large volume of recorded calls and want to speed up assistant setup. - Operations that want to capture customers' real vocabulary, including slang, synonyms, and regional expressions. - Teams that notice Capitão does not understand some questions because they are phrased differently than expected. - Businesses that want to reduce the AI's ramp-up time by using historical data instead of creating FAQs from scratch. Frequently asked questions Does Papo automatically transcribe calls? The Papo platform stores call history and recordings, but audio-to-text transcription must be done with an external tool. After conversion, you can send the text as a document to Capitão's Knowledge Base. Can I use any transcript file format? The most common formats accepted as documents in the Knowledge Base are .txt and .pdf. Check the upload section to see which extensions are currently supported. Do I need to edit the transcripts before sending them? Yes. It is recommended to review the text to correct speech recognition errors and, above all, remove personal or sensitive information that should not be stored in the AI's base. Do transcripts replace FAQs? No. Transcripts work as context material that Capitão consults to better understand customer language. FAQs, on the other hand, are direct answers to specific questions. The ideal is to combine both resources: documents with transcripts to broaden understanding and FAQs to ensure accurate answers. How long does it take for Capitão to start using the transcripts? After upload, the documents enter processing and, once synchronized, can already be consulted by the AI in the next conversations. The exact time depends on the volume of files. Do I need special permission to export recordings? Access to call history and recordings follows your Papo account permissions. Consult an administrator if you cannot see the download option. Read also - How to customize Capitão's tone of voice for different audiences - How to use AI sentiment analysis to generate proactive alerts in Papo Start free → The first user is free — take the opportunity to test how real call transcripts make Capitão sharper.

How to view performance statistics for the Capitão assistant

To view performance statistics for the Capitão assistant in Papo (https://papo.global/), go to the Reports module and open the Agents Report. Capitão appears as an agent in the list, showing metrics such as resolved conversations, average response time, CSAT, and SLA. This way you track the AI's effectiveness in real time and identify opportunities to improve the knowledge base. Combining automated service volume with customer-perceived quality is essential to adjust Capitão's training. Papo organizes these metrics in the same report used for human agents, allowing you to compare the AI's performance with the team's and make data-driven decisions. Summary of metrics available for Capitão | Metric | What it shows | How to interpret | | :--- | :--- | :--- | | Resolved conversations | Total conversations completed by Capitão in the period. | Indicates the volume of automatic responses and the workload the AI absorbed. | | Average response time | Minutes Capitão took to send the first message to the customer. | Reflects the AI's speed; it is usually under a second. | | CSAT | Average score from 1 to 5 given by customers to Capitão's conversations. | Measures satisfaction with automatic responses. | | SLA | Percentage of conversations answered within the deadline defined by the operation. | Shows whether the AI is meeting service-level agreements. | How to access Capitão statistics The Agents Report is the starting point for analyzing the AI assistant's performance. It lists all agents — including Capitão — with the main metrics side by side. 1. Go to the Reports module – In Papo's side menu, click Reports. This hub brings together all operation metrics, from conversation volume to sales performance. 2. Open the Agents Report – Within the reports hub, find the Agents section. The list shows the complete ranking, and Capitão appears identified as “IA Papo” (or the name you gave the assistant). 3. Apply the desired period filter – Use the predefined ranges (Today, Last 7 days, Last 30 days) or set a custom period. To evaluate the impact of a recent knowledge base update, for example, select the days after the change. 4. Analyze the performance columns – Look at the Resolved conversations, Average response time, CSAT, and SLA columns. The number of resolved conversations indicates how many conversations Capitão completed without human intervention. By comparing this number with the total conversations in the period, you calculate the automatic response rate. 5. Export the data – In the report itself, there is an option to export the content as PDF (for presentations) or CSV (for detailed analysis in spreadsheets). The generated file respects the applied period filter, delivering exactly the slice you defined. Who this monitoring is most useful for - Customer service managers who need to measure the return on investment in AI and justify expanding Capitão's use. - Quality analysts who want to compare customer satisfaction with automatic responses versus human service. - Operations teams seeking to identify bottlenecks in the knowledge base — a drop in Capitão's CSAT may signal that FAQs or documents need updating. - White label resellers who monitor AI performance across their clients' accounts to offer improvement suggestions. Frequently asked questions Does Capitão always appear in the Agents Report? Yes. Whenever the Capitão assistant is active and linked to at least one inbox, it will be listed as an agent in the Agents Report, with the same metrics as human agents. How do I calculate Capitão's automatic response rate? The automatic response rate can be obtained by dividing the number of conversations resolved by Capitão by the total conversations in the period. For example, if Capitão resolved 300 conversations and the total was 500, the rate is 60%. This calculation can be done quickly with the data exported to CSV. Is Capitão's CSAT measured the same way as human agents'? Yes. The satisfaction survey is sent to the customer after the conversation ends, regardless of who handled it. Capitão receives a score from 1 to 5, and the average appears in the CSAT column of the report. Can I see Capitão's statistics separated by inbox? The Agents Report shows the assistant's consolidated performance across all inboxes where it operates. For a channel view, you can cross-reference the exported data with the channel volume reports available in Papo. How often are Capitão's metrics updated? The metrics in the Agents Report are updated in real time, reflecting the most recent interactions processed by the assistant. What should I do if Capitão's CSAT is below expectations? A low score generally indicates that the knowledge base needs reinforcement. Review the FAQs and documents linked to the assistant, add frequently asked questions that were not covered, and consider training Capitão with transcripts of old calls to improve understanding of customers' language. Read also - What is the AI Knowledge Base and how to add documents to train Capitão - How to train Capitão with transcripts of old calls to improve understanding Start free → With Papo, you activate the Capitão assistant in minutes and start measuring AI performance at no initial cost.

How to train Capitão to respond in different languages

To train Capitão, Papo's (https://papo.global/) artificial intelligence assistant, to respond in different languages, you configure the assistant's prompt to instruct the AI to detect the language of the message and reply in the same language, and you also add FAQs and documents to the Knowledge Base in the languages you want. You can also create separate Capitão assistants for each language and link them to specific inboxes, ensuring each audience is served in their native language. Capitão's multilingual capability does not depend on a standalone language selector, but rather on the combination of the prompt instructions (the assistant's "brain") and the content you provide in the Knowledge Base. With proper configuration, the same assistant can switch between Portuguese, English, Spanish and other languages, or you can create specialized assistants for each language and link them to different channels. | Strategy | How it works | When to use it | | :--- | :--- | :--- | | Single multilingual prompt | You create a Capitão assistant with a prompt that instructs the AI to "reply in the same language as the question". The Knowledge Base is fed with FAQs and documents in the supported languages. The assistant detects the language and responds accordingly. | Ideal for operations with moderate international volume, where a single team handles all conversations and the experience does not require full separation by language. | | Separate assistants per language | You create a Capitão assistant for each language (e.g., "Capitão EN", "Capitão ES") with a prompt and Knowledge Base exclusive to that language. Each assistant is linked to a specific inbox or widget (e.g., English website, WhatsApp with a dedicated number). | Recommended when there is a high volume of conversations in each language, teams segmented by language, or when answer accuracy requires fully isolated content. | How Capitão processes different languages Capitão uses the same AI engine as Papo's Prompt & AI. When a customer sends a message, the flow is as follows: 1. The message arrives in the inbox (WhatsApp, website, Instagram, etc.). 2. The Capitão assistant linked to that inbox receives the text and consults its prompt and the Knowledge Base. 3. If the prompt contains the instruction to detect the language and reply in the same language, the AI identifies the language of the question and generates the appropriate response. 4. If the Knowledge Base has FAQs or documents in the detected language, Capitão prioritizes that information to compose the answer. Therefore, the key to multilingual support lies in two pillars: - A well-structured prompt: with clear commands such as "always reply in the same language as the customer's message" or "if the customer writes in English, reply in English; if they write in Spanish, reply in Spanish". - A multilingual Knowledge Base: FAQs and documents in the languages you want to support. Without that content, Capitão may understand the language but will not have accurate answers based on your business. Step by step to configure a multilingual Capitão Before you start, check whether you already have a Capitão assistant created. If you don't, create a new one in the Capitão module by providing a name and description. 1. Open the Capitão assistant you want to train. In the Papo dashboard, go to the Capitão module and select the assistant from the list. 2. Edit the prompt (the AI instructions). Include clear guidance on multilingual behavior. Example: "You are a customer service assistant. Always reply in the same language as the customer's question. If the question is in Portuguese, reply in Portuguese; if it is in English, reply in English; if it is in Spanish, reply in Spanish." 3. Add FAQs in the languages you want. In the assistant's Knowledge Base section, create question-and-answer pairs for each language. For example: - "Qual o horário de funcionamento?" / "Atendemos de segunda a sexta, das 9h às 18h." - "What are your business hours?" / "We are open Monday to Friday, 9 AM to 6 PM." - "¿Cuál es el horario de atención?" / "Atendemos de lunes a viernes, de 9 a 18 h." 4. Upload documents (PDFs, text files) that contain relevant information in each language. Capitão extracts knowledge from these files and uses it to answer questions that are not covered by the FAQs. 5. Test the behavior. Send test messages in different languages through the inbox linked to the assistant and check that the replies are in the correct language and contain the expected content. 6. Adjust the handoff scenarios if needed. Create scenarios to transfer the conversation to a human agent when Capitão is not confident about the answer or when the customer asks to speak to a person — regardless of the language. If you choose separate assistants per language, repeat the process for each assistant, with exclusive prompts and knowledge bases, and link each one to the corresponding inbox (e.g., English website widget, WhatsApp number for Spanish-speaking customers). Who this configuration makes sense for - E-commerces that sell abroad and need to serve customers in English, Spanish or other languages without hiring bilingual teams for the first contact. - Travel agencies, hotels and tour operators with a multinational audience that reaches out via WhatsApp or website chat. - Technology and SaaS companies with a global user base that want to offer automated support in each customer's native language. - Businesses that already use Capitão in Portuguese and want to expand automation to new markets without duplicating their support structure. Frequently asked questions Do I need to create a different Capitão assistant for each language? Not necessarily. A single assistant can handle multiple languages if the prompt instructs automatic detection and the Knowledge Base contains content in the desired languages. The choice between one multilingual assistant or several dedicated assistants depends on your conversation volume and how your team is organized. Does Capitão automatically detect the customer's language? Yes, as long as the prompt contains the proper instruction. Papo's AI is able to identify the language of the message and reply in the same language, using the Knowledge Base content available in that language. Can I add FAQs in Portuguese and English to the same knowledge base? Yes. You can register the same question in different languages as separate FAQs. Capitão will look for the answer that matches the language detected in the conversation. How do I get Capitão to reply in Portuguese for customers in Brazil and in English for customers in the US? Configure the prompt with the instruction to reply in the language of the question and feed the Knowledge Base with FAQs and documents in Portuguese and English. If you use different WhatsApp numbers or widgets for each country, you can also create separate assistants and link them to each inbox. Does multilingual training work with PDF documents? Yes. You can upload PDF files, text documents and other formats in the languages you want Capitão to understand. The AI extracts the content and uses it to answer questions, respecting the language of the document. How do I check whether Capitão is replying in the correct language? Use the Capitão assistant's Statistics area to track the volume of automatic replies and, most importantly, run manual tests by sending messages in different languages through the linked inboxes. Analyzing real conversations also helps identify whether the language is being respected. Read also - What the AI Knowledge Base is and how to add documents to train Capitão - How to view the performance statistics of the Capitão assistant Start for free → Create your Papo account and configure Capitão to serve customers in any language, at no initial cost.

How to use Capitão to create automatic conversation summaries and send them by email

Yes, you can use Papo's Capitão (https://papo.global/) to automatically generate a summary of each conversation and send it by email the moment the interaction is closed. The platform combines a custom HTTP Tool in the Capitão module with the artificial intelligence that already analyzes the dialogue, delivering an instant report to the manager – without copying, pasting, or opening another system. | Step | Action | Where to configure | | :--- | :--- | :--- | | 1 | Create a custom HTTP Tool in Capitão (email endpoint) | Capitão module → Custom Tools | | 2 | Create a Flow with a closed conversation trigger | Flows module | | 3 | Add an AI node to generate the conversation summary | Action node (AI) in the Flow | | 4 | Add a node that calls the Capitão Tool, passing the summary | Action node (Tool) in the Flow | | 5 | Activate and test the flow | Flows module | This automation ensures that as soon as an agent closes an interaction, Capitão creates an intelligent summary and forwards it by email – with no manual intervention. How the automation with Capitão works Papo already uses AI to score leads, categorize topics, and provide insights. Capitão acts as a copilot capable of executing programmed actions. By combining a custom HTTP Tool (which sends the summary to an email service) with a Flow that triggers when the conversation closes, the process becomes fully automatic. 1. Trigger: The Flow starts whenever a conversation changes to the “closed” status — whether manually by the agent or by an automatic rule. 2. AI summarization: An artificial intelligence node reads the conversation history and generates structured text with the main points: subject, decisions, pending items, lead score, and sentiment. 3. Call to the Capitão Tool: The summary is sent as a parameter to the configured HTTP Tool, which forwards the message to your email service (your own SMTP, SendGrid webhook, Make, Zapier, etc.). 4. Email delivered: In seconds, the manager receives a concise report in their inbox with everything that matters from that interaction. This way, no information is lost; every closed conversation becomes an immediate record for monitoring or follow-up. Step by step to configure the automatic summary 1. Create the custom HTTP Tool in Capitão Go to the Capitão module and find the custom tools section (Custom Tools). Create a new HTTP tool with the following parameters: - Name: identify the tool (e.g., “Send summary by email”). - Endpoint: URL of the service that will receive the request and send the email (for example, a Make webhook, a SendGrid endpoint, or your own API). - Method: POST. - Headers and Body: configure according to your email provider's API. In the request body, a variable will receive the summary text (e.g., {“summary”: “{{summary}}”}). This variable will be filled by the Flow at execution time. Be sure to test the tool with a sample request to validate email delivery. 2. Create the Flow with a closed conversation trigger In the Flows module, create a new flow and name it (e.g., “Conversation summary on close”). Add a Trigger node and select the event related to closing a conversation. This way, every time an interaction is closed on the platform, the flow will be triggered automatically. 3. Add the AI node to generate the summary Include an Action node that uses Papo's artificial intelligence. In the instruction field (prompt), define exactly what the summary should contain. Example: “Analyze the recently closed conversation and create a summary with: main subject, decisions made, pending items, contact lead score, and overall sentiment. Keep the text in Portuguese, objective, and up to 250 words.” The output of this node will be the summary text, available as a variable for the next steps. 4. Add the action that calls the Capitão Tool After the AI node, insert a new Action node of the type that invokes an external tool – in this case, the HTTP Tool you created in Capitão. Select the “Send summary by email” tool (or the name used) and map the variable with the summary generated by the AI to the expected parameter. This way, whenever the flow runs, the summarized text will be sent to the configured endpoint. 5. Activate and test Save the flow and activate it. To test, close a test conversation (or use a real conversation that will not cause impact) and check whether the email arrives at the configured recipient with the AI summary. Adjust the prompt or the tool as needed. Who this automation is useful for - Service supervisors who need to monitor each interaction without constantly opening Papo. - Sales teams who want instant feedback on the progress of negotiations. - Companies with quality auditing, automatically recording each resolution. - Lean operations, where the manager takes on multiple roles and needs quick reports without manual work. Frequently asked questions Do I need an external email service to use the tool? Yes. Capitão's HTTP Tool sends the data to an endpoint you control. This endpoint can be an API from your email provider (SendGrid, Amazon SES, etc.) or a webhook from integration services such as Make or Zapier, which in turn sends the email. Can I use the Flow's native email action instead of a Capitão tool? Yes. If you prefer to avoid configuring an HTTP tool, you can add a “Send email” node directly to the Flow (using an SMTP credential already configured in Papo). In this scenario, the summary generated by the AI goes straight to the inbox, without depending on the Capitão module. The difference is that, with the custom tool, you gain extra flexibility (such as attaching files, using HTML templates, or integrating external flows). Does the summary include the entire conversation message history? The AI analyzes all messages exchanged in the interaction and extracts the most relevant information according to the configured prompt – subject, decisions, sentiment, and points of attention. You can adapt the prompt to provide more or fewer details, according to the manager's needs. Does the automation trigger when the agent closes the conversation or when the system closes it due to inactivity? The Flow will be triggered whenever the conversation status changes to “closed”, whether by the agent's action, by an automation rule, or by automatic closure configured on the platform. Can I send the summary to more than one email address? Yes. Simply configure the tool's endpoint to accept multiple recipients (or create an email group in your provider) and, in the Flow, pass the summary normally. The HTTP tool will send it to all defined contacts. See also - What is the AI Knowledge Base and how to add documents to train Capitão - How to use AI sentiment analysis to generate proactive alerts in Papo Start free → The first user is free and does not require a credit card — automate your conversation summaries with Papo's AI.

How to Use Bússola to Identify Customers at Risk of Cancellation

Papo's Bússola (https://papo.global/) allows you to identify customers at risk of cancellation by combining the Customer Radar, which prioritizes contacts showing signs of dissatisfaction and churn, and search in conversation history, which locates mentions such as “cancel,” “problem,” or “dissatisfied” across all interactions. With these two features, your team can act proactively on retention before the customer asks to cancel. | Feature | What it shows | How to use it to prevent churn | | :--- | :--- | :--- | | Customer Radar | Lists contacts that need attention, from most critical to least. | Prioritize follow-ups with customers who showed frustration or mentioned cancellation. | | History Search | Lets you search terms and phrases across all conversations. | Find conversations with risk signals such as "cancel," "doesn't work," or a competitor's name. | | Sentiment analysis | Classifies messages as positive, negative, or neutral. | Feeds the radar to increase the criticality of persistent negative conversations. | How Bússola turns conversations into churn alerts Bússola automatically analyzes messages received on channels connected to Papo, such as WhatsApp, Instagram, and others. Instead of relying on an agent to read every conversation, the artificial intelligence identifies risk signals and organizes them for the team. The Customer Radar works as a daily priority list. Criticality is calculated from factors such as dissatisfaction, complaints, cancellation intent, delayed follow-up, and recurring problems. A customer who wrote “I'm thinking about canceling” or “I'm already talking to another provider” tends to appear at the top of the list. History search is the investigative complement: you search strategic terms and get a list of conversations where they were mentioned, with an excerpt from the dialogue. This allows you to do a proactive scan even on contacts that did not appear in the radar. | Goal | Search in Bússola | Recommended action | | :--- | :--- | :--- | | Identify imminent cancellation | "cancel," "I want to cancel," "I don't want it anymore" | Reach out, understand the reason, and offer a solution. | | Detect dissatisfaction | "dissatisfied," "problem," "doesn't work," "disappointed" | Escalate to retention and resolve the cause. | | Monitor competition | Competitor's name | Reinforce differentiators and adjust the proposal. | | Recover stalled customers | "quote," "proposal," "I'll think about it" | Follow up with a new condition. | Step by step to use Bússola for retention | Step | Action | Detail | | :--- | :--- | :--- | | 1 | Access the Bússola module | In the Papo dashboard, open the intelligence area where the radar and search are located. | | 2 | Check the Customer Radar | Review the list ordered from most critical to least. | | 3 | Refine with history search | Search terms such as "cancel," "dissatisfied," or a competitor's name. | | 4 | Analyze the context | Open the full conversation to understand what triggered the signal. | | 5 | Act to retain | Start a message, assign it to an agent, create a Kanban task, or trigger a follow-up. | When it makes sense to use Bússola for retention The routine is useful for operations that already accumulate a reasonable volume of conversations and want to stop losing customers due to lack of follow-up. Customer success, support, and sales teams can use the same signals to: - Prioritize the day based on real criticality, not just arrival order. - Find customers who never opened a formal ticket but showed frustration in chat. - Measure retention effectiveness by reviewing conversations before and after actions. If the operation is small and volume is low, manual reading may be enough. Bússola starts to pay off when there are many messages per day and risk signals get lost among ordinary interactions. Frequently Asked Questions Does Bússola replace Papo's CRM? No. Bússola complements the CRM: it analyzes conversations and points out who needs attention, while the CRM maintains contacts, history, and pipeline. How does the radar define that a customer is critical? The artificial intelligence considers signals in conversations, such as negative sentiment, cancellation words, recurring problems, and delayed follow-up. Customers with these signals appear earlier in the list. Do I need to configure anything to use Bússola? Bússola is already part of Papo's intelligence module. You can start by checking the radar and searching history; account configuration details are in the platform. What terms should I search to find cancellation risk? We recommend starting with "cancel," "I want to leave," "dissatisfied," "problem," "doesn't work," "disappointed," and competitor names. Does search consider only recent messages? Search goes through the conversation history of channels connected to Papo, allowing you to find old and recent mentions. For period limits, check the platform. Does Bússola analyze sentiment automatically? Yes, sentiment analysis classifies messages as positive, negative, or neutral and is one of the factors used by the radar to prioritize customers. Read also - How to use AI sentiment analysis to generate proactive alerts in Papo Start for free and identify customers at risk before they cancel → Start for free to unify WhatsApp, Instagram, and other channels in a single AI-powered dashboard and use Bússola to act before cancellation.

How to customize Live Coach suggestions for specific sales scripts

To customize Live Coach suggestions for a specific sales script in Papo (https://papo.global/), you need to configure the sales Playbook with detailed context about your products, map the most common objections with counter-arguments, define your brand's tone of voice, and select the inboxes where the assistant will operate. Ongoing adjustments based on performance reports further refine the real-time guidance, ensuring every interaction follows your sales operation's exact script. Live Coach (also known as Captain) acts as an AI copilot during sales conversations. For it to avoid generic suggestions and instead follow the script you use every day, it is essential to customize the Playbook — the knowledge base that "teaches" the AI how to communicate, make its case, and handle objections your way. Components of Live Coach customization Customization relies on four pillars that you adjust directly in Papo. Each of them steers the assistant's behavior in the inboxes you choose. | Component | What it customizes | Impact on suggestions | | :--- | :--- | :--- | | Product context | Description of your product/service, differentiators, pricing, and technical details. | The AI can craft accurate answers about features and prices, exactly as the script dictates. | | Tone of voice | Choose between consultative, direct, friendly, or formal. | The style of the suggestions reflects your brand identity, keeping the message consistent. | | Objection mapping | List of frequent objections and ready-made counter-arguments. | When a customer raises an obstacle, Live Coach suggests the reply you defined, not an improvisation. | | Inboxes | Select which channels (WhatsApp, Instagram, etc.) Live Coach should monitor. | Suggestions appear only in the designated sales conversations, avoiding interference in support channels. | How to adjust the Playbook to follow your sales script The key to having Live Coach deliver suggestions truly aligned with your process is filling in every Playbook field with the depth and language of your sales team. Follow this roadmap: 1. Describe the product in rich detail. Don't just say "We sell the Premium plan." Include what the plan covers, its monthly price, who it serves, and the main differentiators versus the competition. The more information, the more relevant the suggestions will be when a lead asks "What does this plan include?" or "Why should I buy from you?". 2. Define your team's exact tone of voice. Papo offers four styles: - Consultative (ideal for complex sales, where the salesperson educates the customer) - Direct (transactional sales, focused on call-to-action) - Friendly (close relationship, more informal language) - Formal (corporate communication, more conservative) Choose the one your script requires. 3. Map objections and create counter-arguments your script already uses. List the 5 to 10 most common objections (e.g., "It's too expensive," "I need to think about it," "I use competitor X"). For each one, write the reply your best salesperson would give — without skimping on words. Live Coach will use these texts as the basis for real-time suggestions. 4. Enable the Playbook only in sales inboxes. During setup, choose the channels where suggestions should appear (for example, only "WhatsApp Sales" and "Instagram Sales"). This prevents the assistant from interfering with support or post-sale inquiries, keeping the focus entirely on the sales script. Whenever any of these settings is changed, Live Coach starts generating suggestions based on the new rules, with no need for manual retraining. Refining suggestions over time Customizing Live Coach is not static work. After enabling the Playbook, monitor Papo's reports to validate whether the suggestions are actually driving sales metrics. Compare conversion rate, first response time, and the volume of closed deals before and after customization. If a particular objection is still causing difficulties, revise the counter-argument you registered. If the tone of voice seems off, switch between the available options. Combining report data with constant Playbook adjustments is the path to turning Live Coach into a copilot that faithfully reflects your sales process. Who needs this customization - Sales leaders who need to standardize the whole team's approach and reduce reliance on individual knowledge. - Agencies and white label resellers who want to offer a customized AI experience for each client, with exclusive scripts per operation. - Companies with complex products or consultative sales, where the right argument and objection handling make all the difference in conversion. Frequently asked questions Can I have a different Playbook for each channel? Yes. You can select different inboxes when configuring each Playbook. That way, you can create distinct scripts for WhatsApp, Instagram, or other channels, respecting each one's particularities. How do I adjust the tone of voice so the suggestions sound like a real salesperson? The platform offers four options: consultative, direct, friendly, and formal. Just select the one that best represents your team's communication. The AI will build that style into every suggestion it generates. Do I need to map all objections at once? No. You can start with the 3 to 5 most frequent objections and add others as the team gains maturity with the tool. The Playbook allows unlimited edits. Can Live Coach suggest scripts for different funnel stages (prospecting, negotiation, closing)? Yes. The product context and objection mapping cover everything from first contact to closing. Suggestions adapt automatically to the moment in the conversation, as long as the information has been registered with the necessary depth. How do I know if the suggestions are really following the script and improving results? Track the evolution of your conversion rate and other metrics in Papo's Agent Report and Sales Report. Comparing the periods before and after customization shows whether the script is being followed and generating a positive impact. Can I test the Playbook before enabling it for the whole team? Since there is no isolated testing environment, the common approach is to enable the Playbook in a single inbox or for a restricted group of agents and monitor the results before expanding. Adjustments can be made at any time. Read also - How to configure Captain to use different knowledge bases per inbox Start for free and customize your team's sales scripts → Papo offers a free trial so you can configure the Playbook and see Live Coach suggesting replies exactly as your sales process requires.

How to integrate Voice Assist with a telephony system for voice support

To integrate Papo's Voice Assist (https://papo.global/) with a telephony system and automate voice call handling with AI, you connect a phone number via Wavoip and associate the AI assistant that answers calls using the same knowledge base and prompt configured on the platform. This way, every incoming call is handled by an AI that understands context, looks up information, and performs actions such as scheduling meetings, without human intervention. Papo lets you unify voice and text support in a single inbox. By connecting a VOIP provider such as Wavoip, you turn Voice Assist — originally created for the website widget — into an intelligent telephone attendant capable of interpreting the customer's speech, responding accurately, and even creating events in Google Calendar. Integration summary | Component | Role in call automation | | :--- | :--- | | Active Wavoip account | Provides the phone number and voice traffic. | | Voice inbox in Papo | Channel that receives and manages calls. | | Wavoip API token | Connects Papo to the telephony provider. | | AI prompt (Voice Assist) | Defines the assistant's behavior and knowledge. | | Knowledge Base | Feeds the AI with company information. | | Google Calendar connection (optional) | Allows the AI to schedule meetings during the call. | How AI voice support works When a customer calls the connected number, Papo receives the call via Wavoip and routes it to the AI assistant. Voice Assist transcribes speech in real time, interprets intent with the same engine as Super Chat, and responds naturally, as if it were a human attendant. If the request requires an action — such as scheduling an appointment — the AI checks Google Calendar and creates the event automatically, without the customer needing to type anything. The flow follows three steps: 1. Call reception: the customer dials the number and the call is forwarded to Papo's voice inbox. 2. Interpretation and response: the AI transcribes the audio, looks up the answer in the Knowledge Base, and responds by voice. 3. Action (if needed): the AI schedules meetings, transfers the call to a human, or sends a summary by email, as configured. Step by step to integrate Before you start, make sure you have an active Wavoip account and a configured phone number. Then follow the process in the Papo dashboard: 1. Get the Wavoip token: in the Wavoip dashboard, copy the API token. 2. Create the voice inbox: in Papo, go to Settings → Inboxes and click Add inbox. Choose the Voice channel and select Wavoip as the provider. 3. Enter the token and save: paste the Wavoip token, give the inbox a name (e.g., “Voice support”), and add the agents who can monitor calls. 4. Associate the AI prompt: in the same inbox, enable automatic AI support and select the prompt you want to use. This prompt can be the same one you already use in the website Voice Assist. 5. Enable actions (optional): if you want the AI to schedule meetings, connect Google Calendar and check the “Schedule a meeting” action in the inbox settings. 6. Save and test: after saving, call the Wavoip number and check that the AI answers and responds correctly. From that point on, all incoming calls will be handled by Voice Assist, with the history recorded in Papo. Who this integration is for Automating telephone support with AI is useful for businesses that receive a high volume of repetitive calls or that need to offer a fast, qualified first response: - Service companies that schedule consultations, visits, or demonstrations. - E-commerce businesses that answer questions about orders, deadlines, and exchanges. - Support operations that want to filter calls before passing them to a human. - Local businesses (clinics, salons, real estate agencies) that want 24-hour support without hiring extra staff. Frequently asked questions Can Voice Assist answer calls like a robot? Yes. With the Wavoip integration and the configured AI prompt, the assistant answers the call, understands the customer's speech, and responds with voice, like a human attendant. Do I need a physical phone number? No. Wavoip provides a virtual number that works over the internet. You can use it to receive and make calls directly in Papo. Can the AI schedule meetings during the call? Yes. If the “Schedule a meeting” action is enabled and Google Calendar is connected, Voice Assist checks available times and creates the event automatically, confirming the details by voice. Can I use the same knowledge base as the text chat? Yes. Voice Assist uses the same prompt and the same Knowledge Base that you already configured for chat support, ensuring consistency in responses. What happens if the AI doesn't know how to answer? You can configure the inbox to transfer the call to a human attendant when the AI cannot find a suitable answer or when the customer asks to speak to a person. Does this integration work with other VOIP providers? Papo offers native integration with Wavoip. For other providers, consult the API documentation or contact support. Read also - How to integrate Wavoip into Papo to make and receive voice calls - How to enable and configure Voice Assist in your website chat? Start free and automate voice support with AI → With Papo's free account, you connect your Wavoip and configure Voice Assist to answer calls automatically, without needing extra equipment.

How to use Macros in Papo to apply quick actions in conversations

With Papo (https://papo.global/), Macros let you run a set of repetitive actions — such as tagging, transferring, and archiving — on a conversation with a single click, reducing manual work and ensuring consistency in service. Macros are shortcuts that apply a predefined sequence of management actions to one or more conversations. Instead of making three or four clicks to organize a service interaction, the agent triggers a single Macro and the platform executes everything at once. Summary: Macros, Quick Replies, and Automation Rules | Tool | What it does | When to use | | :--- | :--- | :--- | | Macros | Runs a set of manual actions with one click (tag, assign, close). | To apply a standard workflow to a specific conversation during human service. | | Quick Replies | Inserts a predefined block of text into the conversation. | To answer frequently asked questions quickly, without typing the same text multiple times. | | Automation Rules | Runs actions automatically based on triggers (e.g., new message). | To automate processes that do not need manual intervention, such as assigning a new lead to a team. | What are Macros? Macros are “recipes” of actions that can be applied to one or more conversations with a single command. They are designed to automate manual and repetitive tasks that agents perform inside the inbox. While Quick Replies focus on speeding up communication (by inserting text), Macros focus on speeding up management (by executing actions on the conversation). The most common actions that can be included in a Macro are: - Add or remove tags (labels): To categorize the conversation (e.g., “Hot Lead”, “Tier 2 Support”, “Feedback”). - Assign the conversation: To forward the service to a specific agent or team (e.g., “Sales”, “Finance”). - Change the conversation status: To mark a conversation as “Pending” or “Resolved” (closed). - Add an internal note: To leave a comment for other team members. By combining these actions, you create fast, error-proof workflows for common everyday situations. How to create a Macro in Papo Macro creation is centralized, allowing administrators to configure shortcuts for the entire team. 1. Go to Settings: Navigate to the settings panel of your Papo account. 2. Find the Automations section: Look for the “Macros” option under “Automations”. 3. Create a new Macro: Click the button to add a new Macro. You will need to define: - Macro name: A clear, descriptive name that helps agents understand its purpose. For example: “Qualify and send to Sales” or “Close and mark as resolved”. - Actions: Select from the list of available actions what the Macro should do. You can add multiple actions that will be executed in sequence. For example, for the Macro “Qualify and send to Sales”, the actions could be: 1. Add tag: “Qualified Lead” 2. Assign to team: “Sales” 4. Save the Macro: After configuring the actions, save the Macro so it becomes available to agents. How to apply a Macro to a conversation Once Macros are created, agents can easily apply them during service without leaving the conversation screen. 1. Open the conversation: Select the conversation to which you want to apply the actions. 2. Find the Macros menu: Inside the conversation interface, usually near the tag or assign options, there will be a button or menu for “Macros”. 3. Select the desired Macro: Click the button and a list of all available Macros will appear. Choose the one that matches the action you need to perform. 4. Confirm the application: The platform will immediately execute all actions configured in the selected Macro. For example, when you click the Macro “Close and mark as resolved”, the conversation will be tagged and its status will automatically change to “Resolved”. This feature can also be used in bulk. In some platform views, you can select multiple conversations and apply the same Macro to all of them at once, further optimizing the management of a large volume of service interactions. Who Macros are useful for - Service agents who need to execute standardized routines quickly, without wasting time on repetitive clicks. - Supervisors who want to ensure that triage, categorization, and closing processes are followed consistently by the entire team. - High-volume conversation operations, where agility in organizing the queue directly impacts response time and customer satisfaction. Frequently asked questions What is the difference between Macros and Quick Replies? The main difference is in what each one does. Quick Replies insert blocks of text into the message box to speed up typing. Macros execute a set of management actions on the conversation, such as tagging, assigning to a team, or closing the service. One does not replace the other — they complement each other. Can I apply a Macro to multiple conversations at the same time? Yes. In some Papo screens, you can select multiple conversations and apply the same Macro to all of them at once. This is especially useful for bulk archiving, tagging a group of service interactions, or redistributing conversations among agents. In what order are the Macro actions executed? Actions are executed in the order in which they were configured during Macro creation. For example, if you define “Add tag” first and then “Assign to team”, the tag will be applied before the transfer. Can I edit a Macro after creating it? Yes. You can access the list of Macros in settings, select the one you want to change, and modify its name or actions. The change takes effect immediately for all agents. Do Macros work in conversations from any channel? Yes. Because Papo’s inbox unifies channels such as WhatsApp, Instagram, Facebook, Telegram, SMS, and email, Macros can be applied to conversations from any origin, without distinction. Is it possible to undo the application of a Macro? Applying a Macro is immediate and cannot be undone automatically. However, you can manually reverse each action (remove the tag, reassign the conversation, reopen the status) if necessary. Read also - What is Super Chat and how to use commands to speed up tasks in the CRM - How to configure conversation assignment by agent skill with teams and rules in Papo Start using for free → The first user is free and no credit card is required.

Automations in Papo: Rules, Macros, Flows, and Webhooks – Which One to Use?

To automate customer service and sales processes, Papo (https://papo.global/) offers four complementary tools: Rules, Macros, Flows, and Webhooks. Use Rules to define the conditional logic that triggers actions; Macros to apply quick actions in conversations; Flows to create interactive journeys; and Webhooks to integrate Papo with external systems. Each resource addresses a different type of need. While Rules and Macros handle one-off and immediate actions, Flows build guided conversations and Webhooks connect the platform to your tool ecosystem. The following table summarizes the main differences. | Tool | Main Purpose | When to Use | Practical Example | | :--- | :--- | :--- | :--- | | Rules | Conditional “if-then” logic to trigger and route automations | Automatic actions based on events, contact segmentation, branching inside Flows | IF a contact receives the “Hot Lead” tag, THEN start the demo scheduling Flow | | Macros | Execute a set of predefined actions with one click | Speed up repetitive tasks during service, such as sending a greeting, adding a tag, and moving the card on the Kanban | With a shortcut, send “Hello, how can I help?”, apply the “Support” tag, and assign the conversation to the technical team | | Flows | Create interactive journeys and multi-step processes | Qualification chatbots, satisfaction surveys, customer onboarding, any flow that requires user interaction | A chatbot that asks the reason for contact and, based on the answer, routes to sales or support | | Webhooks | Send HTTP notifications to external URLs when events occur | Integrate Papo with CRMs, ERPs, spreadsheets, or automation tools such as n8n | Send a new lead’s data to the company CRM as soon as it is created in Papo | What are Rules in Papo? Rules are the platform’s logic engine. They work with an “if-then” structure: you define a condition (the “if”) and an action (the “then”), and the system executes automatically whenever the condition is met. Rules are not a standalone tool, but they appear in three main contexts: - Inside a Flow: to create branches. Example: “IF the contact’s answer contains the word ‘price’, THEN send the price list”. - As an automation trigger: to start a Flow or Sequence. Example: “IF a contact receives the ‘Qualified Lead’ tag, THEN start the scheduling Flow”. - When creating Segments: to filter contacts based on attributes. Example: “Contacts whose ‘Plan’ attribute equals ‘Pro’”. Use Rules when you need immediate automatic actions, without user interaction, or to segment your contact base dynamically. What are Macros in Papo? Macros are shortcuts that execute a sequence of preconfigured actions in a conversation with a single click or keyboard command. They were created to eliminate repetitive tasks during service, such as sending the same greeting message, applying tags, moving the card on the Kanban, or assigning the conversation to a specific team. For example, you can create a macro called “VIP Welcome” that, when triggered, sends the message “Hello, welcome!”, adds the “VIP Customer” tag, and transfers the conversation to the premium service team. Macros are ideal for standardizing responses and ensuring that no step in the process is forgotten, especially in operations with a high volume of conversations. What are Flows in Papo? Flows are Papo’s most versatile automation tool. Through a visual editor, you build complete journeys for the contact, step by step. Each step is a “node” that can be a message, a question, an action (such as adding a tag or updating an attribute), or a condition. Unlike Rules and Macros, Flows allow user interaction: the contact responds and the flow makes decisions based on those responses. They are perfect for creating lead qualification chatbots, satisfaction surveys (CSAT/NPS), onboarding processes, and any scenario that requires a guided conversation. To speed up creation, Papo offers a Flow Library with ready-made templates for the most common use cases. Just install the template, customize the messages, and activate it. What are Webhooks in Papo? Webhooks are HTTP notifications that Papo sends to an external URL when certain events happen on the platform. They are the bridge to integrate Papo with other systems, such as CRMs, ERPs, spreadsheets, or automation tools (e.g., n8n, Zapier). For example, you can configure a webhook so that whenever a new contact is created, Papo automatically sends the data (name, phone, email) to your CRM. Or, when a card is moved to the “Won” column in the Kanban, a webhook triggers a notification in the team’s Slack. Webhooks are ideal when you need Papo to “talk” with other tools in real time, without manual intervention. How to choose the right tool? The decision depends on the type of action you want to automate. Use the quick guide below: - Need a one-off, immediate action without interaction? → Rules or Macros. If the action depends on an event (e.g., tag added), use Rules. If you want a shortcut to execute several manual actions at once, use Macros. - Need an interactive conversation with the customer? → Flows. They allow you to capture responses and branch the service. - Need to send data from Papo to another system automatically? → Webhooks. They are the outbound channel for external integrations. In practice, these tools work together. A Rule can trigger a Flow, which in turn uses a Webhook to send data to a CRM. An agent can use a Macro inside a conversation that is already being guided by a Flow. The combination allows you to automate everything from simply archiving conversations to complete sales and post-sale journeys. Frequently Asked Questions (FAQ) Can I use a Rule without a Flow? Yes. Many Rules execute direct actions, such as applying a tag, archiving a conversation, or sending a notification, without the need for a Flow. Are Macros and Rules the same thing? No. Macros are triggered manually by an agent during a conversation, while Rules are automatic and fire based on events, without human intervention. Do I need to know how to program to use Webhooks? No. Configuring Webhooks in Papo is done through the panel, by entering the destination URL and selecting the events. Implementing the receiver (the system that receives the data) may require technical knowledge, but sending from Papo is simple. Can I use a Flow to send a sequence of messages over time? Yes, but for linear cadences with fixed intervals (e.g., a 5-message follow-up over 2 days), Papo also offers Sequences, which are optimized for this type of communication. Flows are more suitable when there is interaction and branching. How many Macros can I create in my account? The platform allows you to create multiple macros, with no strict published limit. For updated details, consult the official documentation or the Papo panel. Do Papo Webhooks work with any external tool? Yes, as long as the external tool has a URL capable of receiving HTTP requests (a receiving webhook). Tools such as n8n, Zapier, Make, and CRMs with webhook support are compatible. Read also - How to use macros in Papo to apply quick actions in conversations - How to use the Papo API to manage automation Flows Start automating your customer service now → The first user is free forever, no credit card required.

How to install templates from the Flow Library in Papo

To install ready-made automations and speed up your processes, Papo (https://papo.global/) provides the Flow Library, a collection of workflow templates that you can add to your account in a few clicks. This lets you quickly implement lead qualification journeys, satisfaction surveys, and customer onboarding without starting from scratch. The Flow Library was created so teams of all skill levels can take advantage of automation, saving time and reducing complexity when creating workflows. | Step | Action | Description | | :--- | :--- | :--- | | 1. Access | Go to the Flow Library | Inside the Flows module on the Papo platform, find the option to access the template library. | | 2. Explore | Find and preview a template | Browse categories (e.g., Sales, Support) and click to preview the complete flow structure. | | 3. Install | Install the flow in your account | With one click, a copy of the template is added to your list of flows, with the status Inactive by default. | | 4. Customize and Activate | Edit and publish the flow | Review the messages, adjust the actions, and when ready, activate the flow to start using it. | What is the Flow Library? The Flow Library is a collection of preconfigured automation flows provided by Papo. Think of it as an app store for your automations. Each template in the library was designed to solve a specific business need, allowing you to start automating tasks in minutes instead of hours. Some examples of templates you may find include: - Lead Qualification: A chatbot that asks initial questions to filter and prioritize new contacts before sending them to the sales team. - New Customer Welcome: A sequence of messages to guide newly arrived customers through the first steps with your product or service. - Customer Satisfaction Survey (CSAT/NPS): A flow that automatically sends a survey after a support interaction ends and takes actions based on the score received. - Demo Scheduling: A flow that collects lead information and directs it to a salesperson's calendar. The main benefit is speed. Instead of designing a complex flow from scratch, you can install a tested model and focus only on fine-tuning, such as customizing message text or defining which team a task will be sent to. How to install an automation from the Library step by step Implementing an automation from a template is a simple and safe process. The flow only starts working after you review, customize, and manually activate it. 1. Accessing the Library Within the Papo dashboard, go to the Flows module, where you manage all your automations. On the Flows screen, look for the option that leads to the Flow Library or a templates section. 2. Exploring and choosing a template In the library, you will find several templates organized by categories or use cases, such as Sales, Marketing, Support, or Human Resources. - Browse the options: Explore the available templates to find one that fits your needs. - Preview the flow: Before installing, you can click a template to see a preview of its visual structure. This shows all the “nodes” (the steps) of the flow, such as messages, conditions, and actions, allowing you to understand exactly what that automation does before even adding it to your account. 3. Installing the template Liked a template? The next step is to install it. When you click the install button, a complete copy of that flow is added to your list of automations in the Flows module. A crucial detail is that, by default, the flow is installed as INACTIVE. This is a safety measure to ensure an automation does not start running accidentally before it is configured specifically for your business. 4. Customizing and activating the flow With the flow installed and inactive, you have complete freedom to adjust it to your needs. 1. Open the newly installed flow in your list of automations. 2. Review each node: Read the text messages and adjust the language to reflect your brand's voice and policies. 3. Configure the actions: Check actions such as “add a tag to the contact,” “create a task in the Projects module,” or “send a webhook” and configure them according to your operation's rules. 4. Test the flow: Use the platform's testing tools to simulate the customer journey and ensure everything is working as expected. 5. Activate the flow: When you are satisfied with the customizations, simply change the flow status from “Inactive” to “Active.” From that moment on, it will be ready to be triggered by the triggers you defined. Who is the Flow Library for? This feature is especially useful for: - Companies starting with automation: It offers a clear, functional starting point, reducing the learning curve. - Marketing and Sales teams: To quickly implement qualification campaigns, lead nurturing, and follow-up. - Support operations: To automate satisfaction surveys, ticket triage, and answers to frequently asked questions. - White-label resellers: It speeds up configuring automations for different clients, allowing best practices to be replicated with few adjustments. Frequently asked questions Is the installed flow already active? No. Every template from the Flow Library is installed with the status Inactive. You need to review it, make the necessary customizations, and manually activate it before it starts processing contacts. Can I edit a template after installing it? Yes. After installation, the flow becomes an independent copy in your account. You can modify any node, add or remove steps, change texts and conditions, exactly as you would with a flow created from scratch. Is the Flow Library free? The Flow Library is a feature available on the Papo platform. Access to the templates and the ability to install them are part of the Flows module features. For details about plans and limits, see the pricing page on the official website. Can I create my own templates and make them available in the Library? The Flow Library contains official templates curated by Papo. You can create custom flows from scratch and save them in your account, but they are not automatically added to the public library. If you are a white-label reseller, you can replicate flows between your clients using the platform's management tools. What happens if an installed template conflicts with my existing automation rules? Because the flow is installed inactive, you have the opportunity to review its entire structure before activating it. During customization, check whether the template's conditions and actions do not interfere with other automations already running. Adjust the triggers and rules as needed to avoid unwanted behavior. Do I need to know how to code to use the Flow Library? No. Installing and customizing templates is done visually, with no code required. Just click to install and then edit the nodes by dragging and configuring options in Portuguese. Read also - Automations in Papo: Rules, Macros, Flows, and Webhooks – which one to use? Start for free and install your first flow → Create your free Papo account and get immediate access to the Flow Library to automate support and sales in minutes.

How to train Prompt and AI with real conversation examples

To train Papo's artificial intelligence agent (https://papo.global/) with real conversation examples, export transcripts of previous calls or chats, convert them into text documents, and add them to the assistant's Knowledge Base in the Prompt and AI module. This process teaches the AI to recognize the vocabulary, questions, and tone your customers actually use, increasing the automatic response rate and reducing transfers to humans. Summary of the training process with real conversations | Step | Action | Expected result | | :--- | :--- | :--- | | Obtain conversations | Export audio from recorded calls or copy chat texts (WhatsApp, Instagram, etc.). | Raw material with customers' real language. | | Transcribe (if needed) | Convert audio to text with tools such as Google Speech-to-Text, Azure, or Whisper. | Searchable text ready for analysis. | | Prepare documents | Clean sensitive data and save as .txt or .pdf. | Anonymous, organized content for the AI. | | Add to the Knowledge Base | Upload the documents in the agent's Documents section in the Prompt and AI module. | The assistant starts consulting real speech to formulate responses. | | Create FAQs from patterns | Identify frequent questions in the transcripts and register them as FAQs. | Direct answers to the most common questions that appear in conversations. | | Track improvement | Check in the service reports whether the automatic response rate has increased. | Continuous adjustments based on real data. | Why use real conversations to train the AI? The interactions your team has already handled — whether by phone, WhatsApp, Instagram, or chat — reveal exactly how customers express themselves. Unlike static documents or hypothetical questions, transcripts show slang, synonyms, common objections, and even typos that the AI needs to interpret. By feeding the Knowledge Base with this material, you reduce the number of messages the assistant does not understand and that need to be transferred to a human agent. This method is especially useful for businesses that already have a significant volume of recorded conversations and want to accelerate the AI agent's learning curve without relying only on assumptions. How to obtain real conversation examples Papo offers different ways to access interaction history, depending on the channel: - Phone calls: If your account has call recording enabled, you can export the audio files directly from the call history in the Calls module. Then use an external transcription tool (such as Google Speech-to-Text, Microsoft Azure Speech, or Whisper) to convert the audio to text. - Chat conversations (WhatsApp, Instagram, Facebook, etc.): You can manually copy the text of relevant conversations directly from the inbox. For larger volumes, Papo's API allows you to extract message history programmatically, making it easier to collect many examples. - Email and other channels: The same principle applies: copy the body of messages exchanged with customers and save them as text. Regardless of the source, the important thing is to gather interactions that represent real service, sales, or support situations. Step-by-step to train the AI agent with real examples 1. Gather the conversations. Select interactions that cover your business's main topics: product questions, complaints, quote requests, appointments, etc. Prioritize conversations in which the customer used natural language, with regional variations or informal terms. 2. Convert to clean text. For audio, transcribe and review errors. For chats, copy the text and remove sensitive personal information (document numbers, cards, addresses). Save each conversation or group of conversations in a .txt or .pdf file. 3. Add the documents to the Knowledge Base. In the Papo panel, access the Prompt and AI module, locate the agent you want to train (usually the Captain) and go to the Knowledge Base section. Upload the transcription files as documents. The platform processes the content and makes it available for the AI to consult during automatic responses. 4. Create FAQs based on the patterns found. Analyze the transcripts and identify questions that repeat frequently. Register these question-and-answer pairs in the same agent's FAQs area. Example: if several conversations ask “What is the delivery time for the countryside?”, create an FAQ with that exact question and the correct answer. 5. Refine the agent's prompt (optional). In the Prompt and AI module, you can review the main prompt text to incorporate common expressions that appeared in the conversations. This helps the AI adopt the tone and vocabulary your customers expect. 6. Track the results. After a few days, check the Prompt and AI module statistics and the service reports. See whether the automatic response rate has increased and whether the number of transfers to humans has decreased. Adjust the documents and FAQs as needed. Who this training is for - Companies that already have a large volume of recorded calls or historical chats and want to accelerate the assistant's setup. - Operations that want to capture customers' real vocabulary, including slang, synonyms, and regional expressions. - Teams that notice the AI agent does not understand some questions because they are phrased differently than expected. - Businesses that want to reduce the manual work of writing hypothetical FAQs and go straight to what customers actually ask. Frequently asked questions Can I use chat conversations instead of calls? Yes. The method works with any text from real interactions: WhatsApp, Instagram, Facebook, email chats, or even notes from in-person service. The important thing is that the content reflects customers' natural language. Do I need to manually transcribe every conversation? No. For audio, use automatic transcription tools. For chats, simply copy the text. If the volume is very large, Papo's API can help extract message history programmatically. Does the AI learn automatically from conversations in real time? Papo's AI agent uses the Knowledge Base and FAQs you provide. It does not learn on its own from each new conversation; training is done by you when you add documents and adjust the prompt. That is why feeding the base with real examples is essential. How many conversations are needed to see improvement? There is no fixed number. Even a few dozen well-selected transcripts covering the main topics can already make a difference. Ideally, include a variety that represents the different types of interaction in your business. Can I use conversations from channels other than WhatsApp? Yes. The Knowledge Base is shared by the AI agent, regardless of the inbound channel. Conversations from Instagram, Facebook, Telegram, email, or SMS are equally valid for training the assistant. How do I know the training worked? Track the Prompt and AI module statistics and the service reports. An increase in the percentage of successful automatic responses and a reduction in transfers to human agents indicate that the AI is understanding questions better. Read also - How to optimize the Knowledge Base to improve the Captain's RAG search - How to view the Prompt and AI Module statistics Start free → The first user on the Silver plan is free forever, no credit card required — start training your AI agent with real conversations today.

How to configure Capitão to collect customer data and fill the CRM automatically

To configure Papo's Capitão (https://papo.global/) to collect customer data and fill the CRM automatically, you need to create an assistant with strategic questions and a custom tool (HTTP custom tool) that calls the contacts API. This way, during the conversation, Capitão asks for name, email and interest, and writes that data to the CRM with no manual intervention. With this automation, your sales or support team stops typing repetitive information and keeps the CRM always up to date. Papo lets Capitão ask questions in sequence, store the answers and, through an HTTP tool, create or update a contact automatically. | Component | Role | Example | | :--- | :--- | :--- | | Assistant questions (FAQs) | Collect customer data during the conversation | "What is your name?" / "What is your email?" | | Scenario (handoff) | Trigger the tool after the data is collected | Hand off to the contact creation tool | | Custom HTTP tool | Call Papo's contacts API to create or update the record | Send name, email and interest to create a new contact | | CRM | Store the information automatically | Contact appears in the contact list with the fields filled in | How automatic data collection with Capitão works Capitão is Papo's AI assistant, which can be configured with questions, documents and tools. When a customer starts a conversation on a connected channel, the assistant can follow a script of pre-defined questions to capture essential information. At the end of the interaction, a custom tool sends that data to the contacts API, recording everything in the CRM. For example, you can configure Capitão to: - Ask for the customer's full name. - Request the contact email. - Ask about their interest (e.g., "which product are you looking for?"). - Confirm the data and then create the contact automatically. The whole process happens inside the conversation, without the customer noticing the technical complexity. A human agent only steps in if needed, and the CRM stays filled with accurate, standardized information. Step-by-step configuration 1. Create a dedicated Capitão assistant. In the Capitão module, create a new assistant with a descriptive name, such as "Lead Capture". This assistant will be responsible only for capturing data and feeding the CRM. 2. Add the questions (FAQs). Register the questions the assistant will ask the customer. For example: "What is your name?", "What is your email?", "What is your interest?". These questions are stored as approved FAQs and will be used in the conversation. 3. Configure an HTTP custom tool. Create a custom HTTP tool that points to the contact creation endpoint of Papo's API. This tool must be able to receive the collected data (name, email, interest) and send it in the format the API expects. Papo's documentation provides the details needed for this configuration. 4. Create a scenario (handoff). Define a scenario that will be triggered after the answers are collected. This scenario should hand control over to the HTTP tool, which will actually create the contact. The scenario can be configured with a clear instruction, such as "After collecting name, email and interest, call the contact creation tool". 5. Link the assistant to the inbox. Associate the assistant with the desired inbox (WhatsApp, Instagram, website chat, etc.). From that moment on, every conversation arriving on that channel will be handled by the configured Capitão. 6. Test the automation. Send a test message to the channel and check that the assistant asks the questions correctly. After the interaction, open the CRM and confirm that the contact was created with the information provided. Who this configuration is useful for - Sales teams that receive many leads via WhatsApp and need to qualify them quickly, without filling in spreadsheets. - Customer support that wants to automatically record the data of whoever gets in touch, speeding up ticket creation. - Marketing operations that capture interests to segment future campaigns. - White label resellers that offer CRM automation as a differentiator for their clients, letting each one configure their own data collection assistant. Frequently asked questions Do I need to know how to code to configure the HTTP custom tool? It is not mandatory, but you do need to understand the basics of REST APIs and follow Papo's documentation. The platform provides the interface to create the tool; you just fill in the URL, headers and request body as instructed. Can Capitão update an existing contact? Yes. The HTTP tool can be configured to look up a contact by email or phone and, if found, update the fields instead of creating a new record. This prevents duplicates in the CRM. Is it possible to collect more than three pieces of information? Yes. You can add as many questions as needed and map the corresponding fields in the API call. The limit is set by your qualification needs and by the fields available in the CRM. Does this automation work on every channel? Yes, as long as the assistant is linked to the corresponding inbox. You can have the same assistant handling WhatsApp, Instagram and website chat, or create different assistants for each channel. How can I make sure the collected data is correct? Capitão can be configured to confirm the information with the customer before saving it. For example, after collecting the data, it can repeat back what it understood and ask for confirmation, reducing typing errors. Can I use this configuration for white label reselling? Yes. Each client of your reseller business can have its own assistant and custom tools, letting you deliver a CRM automation solution under the client's brand. Read also - How to configure Capitão to check product stock in real time Start for free and automate data collection in your CRM → With Papo, you reduce manual work and keep your CRM always up to date automatically.

How to configure Capitão for upsell and cross-sell

To configure Papo's Capitão (https://papo.global/) for upsell and cross-sell, you can create scenarios that trigger offers automatically when the customer mentions certain terms, and custom tools that query your catalog of complementary products in real time. This way, the assistant suggests upgrades and additional items during the conversation, without human intervention. Capitão is the AI assistant that operates within the chat, supporting service or responding directly to customers. With the combination of scenarios (handoffs) and tools, it not only answers questions but also acts proactively to increase the average ticket, offering the right product at the right time. Configuration summary | Step | Action | Where | | :--- | :--- | :--- | | 1. Identify opportunities | Use Synapse to find customers who have already shown interest in upgrades or complementary products. | Synapse module | | 2. Create offer scenario | Define a scenario in Capitão with a title and instruction that guides the assistant to suggest a product when the customer talks about a specific need. | Capitão assistant settings | | 3. Configure custom tool (optional) | If you need to fetch catalog data in real time, create an HTTP Tool that queries your product database and returns the most suitable suggestion. | Capitão custom tools | | 4. Link to channel | Associate the assistant with the WhatsApp inbox (or another channel) so that suggestions appear automatically in conversations. | Inbox settings | | 5. Test and adjust | Simulate interactions and refine the scenario instructions or tool parameters until the offers are relevant and natural. | Test chat | How Capitão can do upsell and cross-sell Capitão understands the customer's intent based on the conversation context and the settings you define. There are two main ways to activate automatic offers: - Scenarios (handoffs): You create a scenario with a title like "Upgrade offer" and an instruction such as: "When the customer asks about advanced features or mentions they need more resources, suggest the Pro plan and explain the benefits." Capitão follows this guidance whenever the conversation matches the trigger. - Custom tools (HTTP): If the suggestion depends on external data (such as stock, price, or compatibility), you can create a Tool that queries an API from your system. Capitão extracts the current product or the customer's need from the conversation, calls the API, and returns the personalized recommendation in natural language. Both approaches can be combined: a scenario defines when to act, and the tool fetches what to offer. Step-by-step to configure automatic offers | Step | Action | Details | | :--- | :--- | :--- | | 1. Map triggers | List words or phrases that indicate openness to upsell or cross-sell. | Examples: "do you have something more complete?", "I need an add-on", "do you also sell…". Use Synapse to validate these terms in the real conversation history. | | 2. Create the scenario | In Capitão settings, create a new scenario. Give it a descriptive title and write a clear instruction. | Instruction: "If the customer shows interest in more features, offer the Enterprise plan mentioning benefits X, Y, and Z." | | 3. (Optional) Create the custom tool | If necessary, configure an HTTP Tool. Provide your product API endpoint, method (GET/POST), and authentication headers. | Capitão will send parameters such as current_product or category and receive a JSON with the suggestion. | | 4. Associate the assistant with the inbox | Link Capitão to the WhatsApp inbox where customers interact. | From that moment on, the assistant starts applying the scenarios and tools in conversations. | | 5. Test with real cases | Send messages simulating a customer and check if the offer appears naturally. | Adjust the scenario instruction or tool parameters until the suggestion is contextual and useful. | Examples of scenarios and tools - Cross-sell scenario for e-commerce: "Whenever a customer buys a cell phone, suggest a protective case and a fast charger. Explain that they are the items most bought together." - Upsell scenario for SaaS: "When the lead asks about advanced reports, offer the Pro plan and highlight that it includes custom dashboards and PDF export." - Custom tool for recommendation: Configure a Tool that queries your recommendation system and returns the ideal complementary product based on the customer's profile or the item they are viewing. Capitão turns the API response into a message like "Customers who bought this product also bought Kit X, which is 10% off today." Who it is for This configuration is useful for businesses that already have a catalog of products or services and want to increase the average value per customer without overloading the team. E-commerces, SaaS, distributors, and service providers with tiered plans are the main beneficiaries. If you already use Kanban to manage upsell opportunities, Capitão acts at the initial stage, generating these opportunities automatically. Frequently asked questions Can Capitão suggest products based on customer history? Yes, if you integrate a custom tool that queries the CRM or purchase history. Capitão can send the customer identifier to the API and receive personalized recommendations. Do I need technical knowledge to create custom tools? Yes, configuring HTTP Tools requires familiarity with APIs and requests. If you don't have that knowledge, you can start with scenarios only, which are text-based and do not depend on external integrations. How does Capitão know when to offer an upsell? You define triggers in the scenario instructions. For example, "when the customer asks about limits of the current plan" or "when they mention competitors." Capitão interprets the conversation and acts according to these rules. Can I use Capitão for cross-selling services? Yes. Just create scenarios with instructions aimed at complementary services. If you are an accounting firm, you can configure the assistant to offer a tax BPO service whenever a customer asks about ancillary obligations. What are scenarios in Capitão? Scenarios are handoffs that define specific assistant behaviors. Each scenario has a title and an instruction in natural language that guides Capitão on when and how to act, without the need for code. Is it possible to integrate with my product catalog? Yes, through a custom HTTP tool. You need an endpoint that returns complementary products (or the catalog itself) and configure the Tool in Capitão with the URL and authentication. Read also - How to configure Capitão to collect customer data and fill the CRM automatically - How to configure Capitão to generate and send PDF quotes automatically via WhatsApp Start free and configure your Capitão to sell more → No commitment, you test the AI that automatically suggests complementary products on WhatsApp.

How to configure Voice Assist to serve in multiple languages on the website widget

To configure Papo's Voice Assist (https://papo.global/) to serve in multiple languages on your website widget, you associate an AI Agent (prompt) that understands and responds in the desired languages. The voice assistant inherits the behavior of Super Chat: if the prompt is configured to respond in Portuguese, English, or Spanish, voice recognition and responses will follow that guidance, with no need for extra per-language settings. Voice Assist turns your website's chat widget into a voice contact point, and its "brain" is exactly the same AI Agent you use for text support. This means multilingual capability depends directly on how you structure the prompt (assistant instructions). With a single comprehensive prompt or multiple widgets for each language, you can offer voice support in Portuguese, English, and Spanish simply and consistently. | Strategy | How it works | When to use | | :--- | :--- | :--- | | Single multilingual prompt | You create an AI Agent that understands voice commands in Portuguese, English, and Spanish and responds in the same detected language. The Voice Assist associated with this prompt will serve all three languages in the same widget. | Ideal for websites with low international traffic or when the user experience does not require language separation. | | Widgets per language | You create a Website type inbox for each language (e.g., Portuguese page, English page). Each widget receives a Voice Assist configured with a prompt exclusive to that language. | Ideal for websites with translated sections, where each audience should speak with an assistant that already speaks their language natively. | How Voice Assist handles different languages Voice Assist does not have a separate language selector. Instead, it follows the same AI flow as text chat: 1. The user taps the microphone and asks a spoken question. 2. Papo transcribes the audio to text. 3. The text is sent to the same AI Agent that responds to text conversations (Super Chat). The prompt defines the rules, including the language(s) the assistant should understand and use to respond. 4. The generated response is displayed as text in the chat window — and, if the audio response option is enabled, Papo synthesizes the audio in that language using the available voices. Therefore, the key to multilingual voice support is the prompt: - If the prompt instructs the agent to respond in Portuguese, Voice Assist will only understand and respond in Portuguese. - If the prompt includes the instruction "respond in the same language as the question," and the agent has the knowledge to do so, it can switch between Portuguese, English, and Spanish automatically. - If you create a specific prompt "AI Agent – English," the Voice Assist associated with it will serve only in English. Step-by-step configuration Before you begin, make sure the website widget is already installed. Configuration is done per inbox. 1. Go to Settings → Inboxes in the Papo dashboard. 2. Select the Website type inbox you want to adjust or create a new one. 3. In the Voice Assist section, enable the feature. 4. In the Prompt field, choose the AI Agent that contains the language instructions you prepared (e.g., "Multilingual Agent" or "English Attendant"). 5. If you want the assistant to also respond with audio, check that this option is enabled in the Prompt & AI module (audio inherits the language of the text response). 6. Save your changes. If you chose separate widgets per language, repeat the process for each Website type inbox, associating the corresponding prompt with each one. Who this makes sense for - Online stores selling abroad that need visitors from different countries to receive voice assistance in their own language. - Tourism, hospitality, and event companies with a multinational audience browsing the site in more than one language. - Corporate websites or support portals with translated versions (e.g., /en, /es). - Businesses already using Super Chat in multiple languages that want to extend this capability to Voice Assist without duplicating effort. Frequently asked questions Do I need to create a special Voice Assist for each language? No. You keep the same feature, but you can associate a different AI Agent with each website widget. Voice Assist inherits the language from the prompt, so you just need prompts (Agents) configured for the languages you want to serve. Does Voice Assist automatically understand if the customer speaks English or Spanish? The language detection capability depends on the AI Agent's prompt. If you configure the assistant to "respond in the same language as the question," and the AI has sufficient knowledge, it will be able to identify the language and respond appropriately. For greater predictability, the safest method is to have separate prompts per language and use them in different widgets. Can I use the same Knowledge Base for all languages? Yes. The Prompt & AI module allows you to upload reference documents (PDFs, texts) in multiple languages. The AI will use this content to respond according to the detected language, as long as the prompt is instructed to search the base. Does Voice Assist work with audio responses in all languages? Yes. When the audio response option is enabled, Papo synthesizes the voice in the language of the response generated by the prompt. You can choose from the available voices, which support Portuguese, English, and Spanish. Do I need to change the widget code for each language? No. The widget script is the same. Language separation is done by creating different inboxes in the dashboard and associating each one with the widget installed on the corresponding page of your website. Is it possible to switch languages during the same conversation? If the prompt is configured to detect and respond in the language of the question, the assistant can switch languages with each interaction. However, the consistency of the experience depends on the quality of the prompt and the knowledge base. Read also - How to train Voice Assist to answer phone calls in Portuguese and English Start free and enable multilingual Voice Assist on your site → With Papo's free account, you connect your channels and offer voice support in Portuguese, English, and Spanish without needing multiple tools.

How to configure human review for Capitão responses in sensitive conversations

To configure human review for Capitão responses in Papo (https://papo.global/) sensitive conversations, you can combine an automation rule that detects terms like cancellation or complaint with actions to tag and assign the conversation to a human team. The goal is to prevent the assistant from continuing to respond on its own and ensure that an agent takes over the case quickly. Summary: automation vs Capitão handoff | Component | Role in human review | Practical example | | :--- | :--- | :--- | | Automation rule | Detects new conversation with sensitive term and applies tag + assignment | Trigger “New conversation”, content condition, and actions to tag and assign | | Tag | Marks priority or sensitivity of the service | “Cancellation”, “Complaint”, or “Retention” | | Capitão handoff scenario | Transfers an ongoing conversation to a human agent | Scenario with instruction “transfer when customer talks about canceling or complaining” | | Assignment to agent or team | Ensures a human takes over the conversation | Assign to retention or support team | How human review works in Capitão In Papo, Capitão can act as a copilot in chat and answer questions based on approved FAQs and documents. In sensitive conversations, however, it's best that it doesn't continue responding automatically. Human review can be configured at two moments: 1. At conversation entry: an automation rule with trigger “New conversation” identifies the content of the first message. If there are terms like cancellation, complaint, refund, or threat of churn, the rule applies a tag and already assigns the conversation to a human agent or team. 2. During Capitão's service: an assistant handoff scenario can transfer control to a human agent when the conversation goes off-script or becomes sensitive. Thus, Capitão stops responding and an agent takes over the case with the full history. Step-by-step to configure review with automation rule 1. Access the Automation module In the Papo panel, go to Settings and click Automation. This is the hub where you manage the rules that automate actions in service. 2. Create a new rule Give it a descriptive name, such as “Human review – cancellation or complaint”. This makes it easy to identify and adjust later. 3. Define the trigger “New conversation” Choose the event that triggers the automation. To capture the conversation as soon as it arrives, select the trigger related to new conversation created or “New conversation”. 4. Add the sensitive content condition Include a condition to filter the message content. You can use keywords like “cancel”, “cancellation”, “complaint”, “refund”, or common expressions from your business. You can also combine other conditions, such as source channel or time, to refine when the rule should act. 5. Add the actions: tag and assign In the actions section, select Add tag and choose a tag like “Sensitive” or “Retention”. Then add a second action to Assign to a team or agent responsible for retention. This way, the conversation arrives already marked and directed to a human. 6. Save and activate Review the settings, save, and activate the rule. From that moment on, every new conversation that meets the conditions will receive the tag and be assigned to the human team, reducing the chance of Capitão responding alone in a delicate case. How to configure Capitão handoff for ongoing conversations In addition to the entry rule, you can create a handoff scenario in the Capitão module. This scenario should contain a clear instruction, such as: “When the customer shows intention to cancel, complain, or request a refund, do not respond with an FAQ. Immediately transfer the service to a human agent and inform that the team will help.” The scenario is linked to the Capitão assistant responsible for the inbox. When the conversation reaches this point, control passes to a human agent, who sees the full history in the panel. Who this configuration is for - Retention and support teams that need to prioritize cancellations and complaints before any automatic response. - High-volume operations on WhatsApp and other channels, where not every contact should be handled by AI. - Companies that use Capitão as first-line service, but want human control in risky conversations. - White label resellers that offer automation with security for their clients, maintaining human review in sensitive cases. Frequently asked questions Can Capitão be paused automatically when detecting cancellation? Yes, in two layers. For new conversations, the automation rule tags and assigns the case to a human before Capitão takes over. For ongoing conversations, the Capitão handoff scenario can transfer the service. The exact configuration may vary by account; additional details are available on the Papo website. Do I need to create a tag before configuring the rule? Not necessarily. During rule creation, when selecting the Add tag action, you can choose an existing tag or create a new one in the configuration flow itself. Does this automation work for all channels? Yes. You can use the Source inbox condition to apply the rule to specific channels, such as WhatsApp, Instagram, or email, or combine with the content condition to cover multiple channels at once. What if the customer complains after Capitão has already responded? In that case, the Capitão handoff scenario is the appropriate resource to transfer the conversation to a human agent mid-service. The agent takes over with the previous history visible in the panel. Can I add more than one action in the same rule? Yes. You can include multiple actions, such as Add tag and Assign to a team, in the same rule. Thus, the conversation arrives already tagged and directed to the responsible human. Read also - How to configure conversation escalation between support levels (N1, N2, N3) in Papo - How to configure Capitão to collect customer data and fill the CRM automatically Start for free → Create your free account on Papo and centralize WhatsApp, Instagram, email, and SMS in a single inbox with AI and CRM in the same panel.

How to create an automatic post-service reminder via WhatsApp

To create an automatic post-service reminder via WhatsApp, use the Flows module in Papo with a condition triggered when the conversation changes to the closing status. The flow sends a follow-up message using an official Meta template (HSM), keeping the relationship with the customer alive without manual work. Unlike a pre-service reminder, which notifies about an upcoming appointment, the post-service follow-up runs after the conversation is completed. It can include a satisfaction survey, after-service instructions, a request for a review, or a comeback offer. Everything is triggered automatically through WhatsApp, without an agent having to copy and paste messages. Overview of the post-service follow-up automation | Component | Role in the automation | Example of use | | :--- | :--- | :--- | | Flows | Orchestrates the sequence of checking and sending after the service ends. | Flow "Post-service follow-up". | | Status condition | Checks whether the conversation has been closed, preventing messages for active conversations. | If the conversation status indicates closure, continue the flow. | | "Send Message" action | Triggers the follow-up template via WhatsApp. | Template with the customer's name and a satisfaction question. | | Official Meta template (HSM) | Message approved for proactive sending outside the customer interaction window. | "Hi, {{contact.first_name}}! How was your service experience?" | | Synapse (optional) | Connects the status change or another event to the Flow automatically. | When the conversation is closed, enroll the contact in the follow-up Flow. | How the post-service follow-up works The logic is simple: when an agent or the AI closes the conversation, the status is updated in the Papo dashboard. The Flow monitors this condition and, if it is true, triggers the sending of a WhatsApp message. Since the customer may not have sent a new message at that moment, the send is proactive — which is why using an official Meta template approved in the WhatsApp inbox is mandatory. This follow-up keeps the relationship active after the issue has been resolved or the sale has been completed. Instead of letting the contact go cold, the automation brings the customer back for a survey, a recommendation, or a next step, without relying on the team's memory. Step by step to set up the automatic reminder 1. Define the closing status Before building the flow, confirm which status your team uses to indicate that the conversation has ended. The names may vary depending on the account configuration. Use that same status as the reference in the condition. 2. Create a new Flow In the Flows / Chatbot module, add a new flow and give it a clear name, such as "Post-service follow-up". 3. Configure the status condition After the initial node, insert a Condition. Choose a rule that compares the current conversation status with the closing status defined in step 1. This way, the flow will not send messages for conversations that are still in progress. 4. Add the sending action On the branch where the condition is true, add the Send Message action. Select an approved Meta template available in your WhatsApp inbox. Write the text with variables such as {{contact.first_name}} to personalize the message. 5. Activate the Flow and monitor it Save the changes and change the flow status to Active. From that moment on, the automation starts monitoring closed conversations and triggers the follow-up according to the configured condition. 6. Connect Synapse if needed If the status change is not available as a direct trigger in your flow, use Synapse to connect the closing event to the Flow. It automates contact enrollment and reduces manual configuration. Important: since the message is sent proactively, you must use a Message Template (HSM) approved by Meta. Papo lets you select approved templates directly in the WhatsApp Cloud API inbox. Who this automation is recommended for - Clinics, medical offices, and beauty salons that want to know how the service went and encourage customers to return. - Service providers who want to send after-service instructions or ask for a review without manual effort. - E-commerce stores and shops looking to increase repeat business with a post-sale message. - Support teams that want to reduce the number of repeated tickets and keep the customer informed after resolution. Frequently asked questions Do I need an approved template (HSM)? Yes. For proactive follow-up sending, Meta requires an approved Message Template. Papo lists the approved templates from your WhatsApp Cloud API inbox. Can I use the same flow for Instagram? Yes, as long as your Instagram inbox is connected to Papo and the approved template is available for the channel. The Flow logic can be adapted for different channels. How do I avoid sending a follow-up for a conversation that is still active? The status condition handles that. The flow only triggers when the conversation is in the status defined as closing. While the conversation is active, the condition is not met. Can I include a satisfaction survey in the message? Yes. In the template content, include a rating question or a link to the survey. The message can contain variables such as the contact's name to personalize the experience. Do I need Synapse for this automation? Not necessarily. If your flow already accepts the status condition as a trigger, Synapse is optional. It becomes useful when you need to connect the status change across different modules of the platform. Read also - How to set up an automatic NPS survey flow when a deal is closed - How to create a seasonal automatic greeting on WhatsApp and Instagram Start for free → The first user is free so you can test the automation on your company's number.

How to configure Voice Assist to answer FAQs on the website and transfer to a human?

To configure Papo's Voice Assist (https://papo.global/) to answer frequently asked questions on the website and transfer to a human, associate an AI Agent trained with your FAQs and set the "Talk to an agent" action in the widget settings. This way, the assistant answers by voice and, when the customer asks for a person or the AI cannot find the answer, transfers the conversation to an agent in the Papo chat. Voice Assist is a feature that adds voice interaction to your website's chat widget. It uses the same AI Agent (prompt) and the same Knowledge Base as Super Chat, so the FAQs you already prepared for text support can be used by voice. The transfer to a human is done within the conversation itself, without switching channels or losing context. | Step | Action | What to configure | | :--- | :--- | :--- | | 1. Access Inboxes | Go to Settings → Inboxes. | Manage communication channels. | | 2. Select the website widget | Choose the inbox of type Website or create a new one. | Voice Assist is available for this type of inbox. | | 3. Enable Voice Assist | Turn on the feature in the dedicated section within the widget settings. | A simple toggle turns on the feature. | | 4. Associate an AI Agent with FAQs | In the Prompt field, select the agent trained with your Knowledge Base. | Voice responses come from the same AI as the text chat. | | 5. Configure the transfer action | In the assistant's actions, check the Talk to an agent option. | This allows the AI to forward the conversation to a human agent. | | 6. Save and publish | Save the changes to update the widget. | The microphone icon will appear on the website. | How Voice Assist answers FAQs and transfers to a human When a visitor clicks on the chat widget and taps the microphone icon, Voice Assist follows this flow: 1. The user asks the question out loud. 2. Papo transcribes the audio to text in real time. 3. The text is sent to the associated AI Agent, which queries the Knowledge Base (PDFs, texts, company FAQs). 4. The AI formulates a response and displays it in the chat — as text or also as audio, depending on the configuration. 5. If the user explicitly asks to speak to a person, or if the AI cannot find a reliable answer in the FAQs, the assistant offers the Talk to an agent option. 6. When the customer accepts, the conversation is transferred to Papo's human support queue, keeping the dialogue history in the same inbox. For the transfer to happen naturally, it's worth instructing the prompt (in the Prompt & AI module) to recognize phrases like "I want to talk to an agent", "I need help from a person" or "I want to cancel", and to trigger the corresponding action instead of trying to answer on its own. Step-by-step configuration 1. Access Inboxes In the Papo dashboard, click Settings and then Inboxes. 2. Select the website inbox Choose the inbox that corresponds to the widget installed on your website. If there isn't one yet, click Add inbox and select the Website option. 3. Enable Voice Assist Within the website inbox settings, find the Voice Assist section and turn on the toggle. 4. Associate the AI Agent with your FAQs In the Prompt field, select the AI Agent that contains your company's frequently asked questions and answers. This same agent can be used in the text chat and Voice Assist. 5. Configure the transfer actions In the interaction options, check Talk to an agent. You can also offer Start a WhatsApp conversation or Schedule a meeting, if it makes sense for your support. 6. Save and test After saving, open the widget on your website, tap the microphone, simulate a frequently asked question, and then ask to speak to an agent. Check that the transfer reaches the correct queue. Who it makes sense for - Support teams that receive many repeated questions and want to free up agents for complex cases. - Online stores and services with a high volume of questions about deadlines, exchanges, and availability. - Companies that want to offer voice support on their website without abandoning the human support flow. - Businesses that already use Super Chat with FAQs and want to enable the same knowledge by voice. Frequently asked questions Can Voice Assist transfer to a phone? No. The transfer configured in Papo's Voice Assist sends the conversation to a human agent within the Papo chat or, if you enable it, to WhatsApp. For phone support, you need to use another channel or platform feature; details are on the Papo website. How does the AI know when to transfer instead of answering? You can instruct the AI Agent's prompt to recognize the intent to speak to a person. Additionally, if the AI cannot find a reliable answer in the Knowledge Base, it offers the Talk to an agent option, provided that action is enabled. Can I use the same knowledge base as the text chat? Yes. Voice Assist uses exactly the same AI Agent and the same Knowledge Base as Super Chat. There's no need to create separate FAQs for voice. Do I need to configure anything additional for the transfer to keep the history? The transfer occurs within the same conversation and the same inbox, so the human agent sees the dialogue context. If you need specific rules about history or queues, check your account options on the Papo website. Does Voice Assist understand any accent or language? The recognition and response capability depends on the AI Agent's prompt. To support multiple languages, configure a multilingual prompt or associate different prompts with separate widgets per language. Read also - How to configure Voice Assist to support multiple languages in the website widget - What is Papo's artificial intelligence and how does it work? Start for free and enable Voice Assist → With Papo, you set up the voice assistant at no initial cost and the AI answers FAQs by voice, transferring to a human when needed.

How to configure Capitão to send automatic follow-up if the lead doesn't respond in 24 hours

To automatically send a follow-up message when a lead doesn't respond in 24 hours, the most direct path in Papo (https://papo.global/) is to configure a Flow with a 24-hour wait, an inactivity condition, and the Send Message action. Capitão comes in as a support layer: it can help qualify the lead, suggest the message tone, or give the team context, while the flow ensures the follow-up goes out at the right time, without relying on a manual reminder. Automation summary | Step | Resource | Configuration | Result | | :--- | :--- | :--- | :--- | | Pause the journey | Wait step in Flow | Set 24 hours | Waits the ideal period before re-engaging | | Detect silence | Inactivity condition in Flow | Check that the lead did not interact | Advances only leads without a response | | Send the follow-up | Send Message action | Short text or Message Template | Triggers the re-engagement message | | Personalize the content | Capitão as copilot | Suggest tone and context before saving | Message better suited to the conversation | How the inactivity follow-up automation works In Papo's Flow builder, you combine two moments: a wait and a check. The Wait step pauses the journey for minutes, hours, or days — in this case, 24 hours. Right after, an inactivity condition evaluates whether the contact remained without interacting during that period. Only when the condition is true does the Flow advance to the Send Message action. If the lead responded in that interval, the Flow can be ended or routed to an agent, avoiding out-of-context messages. This sequence is useful for sales cadences, lead nurturing, and reopening conversations after a proposal, quote, or material sent. It also helps confirm deadlines, resume stalled negotiations, or remind the customer of the next step. The automation can be started when a new contact is created in the CRM, when a tag is applied, when the card enters a Kanban stage, or when a rule detects a conversation without interaction — always respecting valid opt-in. Capitão does not execute the wait or the inactivity condition; those are Flow functions. But you can use it before activating the flow to review the message, suggest variations, or qualify the leads that will enter the follow-up. How to configure the 24-hour follow-up 1. In the Papo dashboard, go to Flows / Chatbot and create a new Flow or edit an existing one. 2. At the point where the lead should start waiting, add the Wait step. 3. Set 24 hours as the interval. 4. Add an inactivity condition right after the Wait step. Configure it so the Flow continues only if the lead has not interacted in the period. 5. Connect the Send Message action after the condition and write a short follow-up, with context and a clear question. 6. If the send occurs after WhatsApp's 24-hour window, use a Message Template (HSM) approved by Meta. 7. Optional: use Capitão as a copilot to review the text or suggest a more personalized version before saving. 8. Save, activate, and test the Flow with a sample contact. A variation to keep the inbox organized is to use a rule with the conversation-without-interaction trigger. In this configuration, the rule detects inactivity, archives the conversation, and the archiving triggers a Flow that performs the wait and sends the follow-up. This option is interesting when the team wants to clear the queue and, at the same time, not lose the opportunity to re-engage the lead. Who this automation makes sense for - Sales teams that receive leads on WhatsApp, Instagram, or other channels and don't want to lose opportunities due to lack of follow-up. - Operations that use Kanban and CRM to track proposals and need an automatic touch when the lead goes silent. - Businesses with high volume and a small team, which can't manually remember every follow-up. - Companies that separate leads by tag, stage, or channel and want standardized re-engagement cadences. - White label resellers that want to offer follow-up cadences as part of customer onboarding. Frequently asked questions Is the follow-up canceled if the lead responds during the wait? The safest way is to configure the inactivity condition after the Wait step. Thus, if there is a response, the flow can be ended or directed to human service, instead of sending the automatic message. The final behavior depends on the condition you define in the Flow. Do I need to use a Message Template on WhatsApp? Yes, if the follow-up is sent after WhatsApp's 24-hour service window. In that case, the message must be a Message Template (HSM) approved by Meta. For shorter waits, a regular text message can be used within the allowed window. Can I restrict the automation to a specific group of leads? Yes. You can add conditions in the Flow, such as tag, channel, or funnel stage, so that only leads that meet the criteria enter the wait and receive the follow-up. Does Capitão send the follow-up by itself? No. Automatic sending by inactivity is executed by the Flow, with the Send Message action. Capitão can support lead qualification and the construction of more contextual text, but it does not replace the flow's inactivity condition. Can I use this follow-up on channels other than WhatsApp? The wait and inactivity logic can be applied to channels connected to Papo. Sending rules, such as the requirement of a Message Template after 24 hours, apply to WhatsApp; on other channels, check the platform's sending policies. Read also - How to configure Capitão for upsell and cross-sell - How to create an automatic post-service reminder via WhatsApp Start for free and activate your automatic follow-up on WhatsApp → The first user is free, no card required, to test inbox, CRM, and AI on your number.

How to configure Voice Assist to automatically schedule appointments from your website

To automatically schedule appointments from your website with Papo's Voice Assist (https://papo.global/), you need to enable the voice assistant in the widget, associate it with an AI Agent, connect your Google Calendar, and activate the scheduling action — which can also use services and professionals defined in the platform's Appointments module. | Requirement | Role in voice scheduling | | :--- | :--- | | Website chat widget | Displays the microphone icon so visitors can start a voice command. | | Voice Assist enabled | Allows voice commands to be received and processed in the widget. | | Associated AI Agent | Interprets the spoken request, identifies the scheduling intent, and guides the conversation. | | Google Calendar connection | Checks busy times and automatically creates the event in Google Calendar. | | "Schedule a meeting" action enabled | Unlocks the scheduling flow for the voice assistant. | | Appointments module (optional but recommended) | Defines services, durations, and professionals, allowing scheduling to follow these rules. | How voice scheduling works with the Appointments module Voice Assist turns a spoken request into a real appointment. When a visitor taps the microphone in the widget and says something like “I want to schedule a consultation tomorrow at 2 p.m.”, the platform transcribes the audio and uses the same AI Agent as Super Chat to interpret the request. From there, Papo checks the connected Google Calendar, verifies available times, and creates the event with the date, time, and guest email. If you also use the Appointments module — which lists services, professionals, and durations — the AI can cross-reference this information with the calendar. For example, it can check the chosen professional's availability, suggest a time that fits the service duration, and include the appointment type in the event. This makes the flow even more precise for clinics, practices, studios, and businesses that organize schedules by specialist. The complete process in three steps: 1. Voice input: the visitor speaks the request in the website widget. 2. Interpretation and lookup: the AI understands the intent, checks Google Calendar availability, and, if applicable, the Appointments module data. 3. Event creation: Papo schedules the appointment without the visitor needing to type or navigate through multiple screens. Step-by-step setup Before you begin, confirm two points in your Papo dashboard: - The account is connected to Google Calendar (the platform shows the connection status). - If you want to schedule specific services with professionals, set up the services and team in the Appointments module (duration, price, and linked professionals). Then follow these steps: 1. Go to Settings → Inboxes. 2. Select an existing Website inbox or create a new one. 3. In the Voice Assist section, enable the feature. 4. Associate Voice Assist with an AI prompt already configured in your account (the same Agent that handles text chat). 5. In the interaction modes, check the Schedule a meeting option. 6. Save your changes so the widget receives the microphone icon. You do not need to change the widget installation code. Once saved, any visitor who accesses the website from a mobile phone or computer will see the microphone in the chat and can schedule by voice immediately. Who this is for - Health clinics and practices (psychologists, nutritionists, physiotherapists) that rely on appointment scheduling through their website. - Beauty salons, barbershops, and aesthetic studios that organize schedules with different professionals and services. - Pilates, yoga, and gym studios with private classes. - Service providers (lawyers, architects, consultants) that offer an initial consultation or demo. - Stores that schedule product demos or technical visits. Frequently asked questions Does Voice Assist schedule on its own or only suggest times? With the scheduling action enabled and Google Calendar connected, the assistant can check available times and automatically create the event in Google Calendar. The AI confirms the details with the visitor before finalizing. Do I need to connect Google Calendar before enabling Voice Assist? Yes. Without this connection, the assistant cannot check busy times or create the appointment. Do I need to use the Appointments module to start scheduling? It is not required. Without the module, Voice Assist schedules a generic event on the calendar. But with the module configured, the appointment is created with the correct service and, if desired, the chosen professional — which makes team organization easier. Can the AI choose a specific professional? Yes. If professionals are registered in the Appointments module and their calendars are reflected in Papo's Google Calendar connection, the assistant can check each one's availability and suggest times with the professional the customer requests. Can I use voice scheduling on WhatsApp or other channels? No. Voice Assist is exclusive to the Website chat widget. Inboxes for other channels, such as WhatsApp, Instagram, or Telegram, follow each channel's own rules. What if the visitor wants to reschedule or cancel later? Voice Assist can handle rescheduling or cancellation requests by voice, as long as the AI Agent prompt includes instructions for this. The AI interprets the command and updates the event in Google Calendar according to the conversation, keeping your operation up to date without manual intervention. Read also - How to configure Voice Assist to answer FAQs on your website and transfer to a human? - How to configure Voice Assist to support multiple languages in the website widget? Start for free and schedule appointments by voice on your website → With Papo's free account, you connect your Google Calendar and let visitors automatically schedule appointments by voice command, without switching tools.

How to use Bússola to find conversations that mention a specific product

Bússola da Papo (https://papo.global/) is the intelligence module that lets you search by keywords or phrases across your entire conversation history. With it, you quickly find every interaction in which a customer mentioned a specific product, even if they didn't use the exact name. That turns your chat archive into an active base for sales follow-up and support. Instead of relying on the team's memory or outdated spreadsheets, Bússola's semantic search automatically scans messages from channels such as WhatsApp, Instagram and email. Just type the product name (or common variations) to see who said it, when they said it and in what context — and act right away. Quick summary: from search to follow-up | Action | How to do it | Result | | :--- | :--- | :--- | | Open Bússola | In the Papo dashboard, open the Bússola module | Access to intelligent search and the customer radar | | Search for the product | Type the product name or related terms into the search field | A list of conversations where the term appears, with highlighted excerpts | | Analyze the context | Click each conversation to read the full dialogue | Understand whether it was interest, a question, a complaint or a comparison | | Tag and assign | Apply a tag (e.g., "interest-product-X") and assign the conversation to a salesperson | The opportunity becomes visible in the owner's queue, with an internal alert | | Follow up | Resume the conversation with a personalized message referencing what the customer said | Seize the right moment to move the sale forward or overcome an objection | How semantic search finds product mentions Bússola's search is not limited to finding the exact product name. It interprets the meaning of sentences, locating indirect mentions such as "the most advanced model", "the premium version", "the one with AI" or even nicknames customers use. This is possible because the platform continuously indexes the content of every conversation that arrives in Papo's unified inbox. In practice, when you search for "Gold Plan", the system may also return messages such as "your most expensive package" or "the plan that includes reports". This intelligence eliminates the manual work of imagining every variation and ensures no opportunity goes unnoticed. Step by step to find conversations about a specific product | Step | Action | Practical detail | | :--- | :--- | :--- | | 1. Open Bússola | In the Papo dashboard, click the Bússola module | Inside Bússola, locate the conversation history search area | | 2. Define your search terms | Type the product name and common variations | Include codes (e.g., "SKU-123"), nicknames, descriptions and even typical objections ("expensive", "doesn't have feature X") | | 3. Analyze the results | Bússola displays a list of conversations with excerpts of the mention | Click each one to see the full dialogue, the channel and the date | | 4. Create an internal alert | Apply a tag (e.g., "interest-product-X") and assign the conversation to an agent or salesperson | The tag lets you filter later; the assignment sends a notification so the owner can act | | 5. Start the follow-up | Write a contextualized message that picks up on what the customer said | E.g., "I saw you mentioned Product X last month. Can I help with details or a demo?" | Who this search is useful for - Sales teams and SDRs who want to recover leads that showed interest in a specific product but didn't move forward. - Customer success teams monitoring mentions of new features, upgrades or renewals. - Marketing and product teams that want to collect real feedback about a launch or feature. - Operations with a high volume of conversations, where reading every interaction manually would be impossible. Bússola also helps build a product intelligence base: instead of guessing, your team analyzes what customers actually said, allowing you to adjust offers, sales arguments and development priorities. Frequently asked questions Does Bússola find indirect mentions of the product, such as nicknames or descriptions? Yes. The search is semantic and interprets meaning, so terms like "the most expensive one", "the full version" or "the one with AI" can surface relevant conversations even without the exact product name. Do I need to configure anything before using product search? No. Bússola automatically indexes messages from the channels connected to Papo (WhatsApp, Instagram, email and others). Just open the module and search. Can I combine product search with other filters? The search returns conversations based on the term you type. After that, you can apply tags and use the CRM filters to segment contacts even further. Details about advanced filters are available on the platform. How can I tell whether the customer really intends to buy? When you open the conversation, you see the full context. Bússola shows the excerpt of the mention, but it's up to the agent to assess whether it was a question, a compliment or a buying intent. AI helps prioritize, but the final decision is human. Can I save a search to repeat later? The platform does not save searches automatically, but you can create tags based on the results and monitor those contacts periodically, generating always up-to-date lists. Does Bússola only show recent conversations? No. It searches your account's entire available conversation history, as long as the messages have been indexed. That lets you recover old opportunities that can still be converted. Read also - How to use Bússola to follow up with customers who request a quote but don't close? - What is Papo's artificial intelligence and how does it work? Start for free now → The first user is free and no credit card is required — try Bússola's intelligent search and turn product mentions into sales.