WhatsApp chats are recorded along with your other conversations. This allows you to view the message history, monitor AI performance, and manage customer interactions.
Viewing WhatsApp Chats
WhatsApp chats appear in your Chats section (Conversations) with a “WhatsApp” badge.
Each conversation includes:
- The complete message history
- The customer’s phone number
- The connected AI assistant
- Timestamps for all messages
- Cost and usage metrics
Chat Types
| Type | Description |
|---|
| WhatsApp | Chats conducted via WhatsApp |
| Web Widget | Chats through the website widget |
| Test | Test chats from the dashboard |
How WhatsApp Chats Work
- The customer sends a message to your WhatsApp number
- The system creates a new conversation or finds an existing one
- The message is forwarded to your AI assistant
- The AI generates a response and sends it
- The 24-hour messaging window opens
Within the 24-Hour Window
- The AI automatically replies to every message
- Free-form responses are allowed
- The conversation continues naturally
After 24 Hours
- The customer must send a message again to reopen the window
- You can send template messages to re-engage the customer
- The AI remains ready to respond once the customer replies
Chat Features
Message History
View the entire exchange between customer and AI:
- Incoming customer messages
- Outgoing AI responses
- Timestamps for each message
- Function calls made during the conversation
Conversation Variables
Variables collected or set throughout the conversation:
- Customer identifier
- Data extracted by the AI
- Context passed from automation flows
Cost Tracking
Each conversation records:
- Total number of messages
- Cost per message
- Total cost of the conversation
Managing Chats
Chat Filters
Filter your chat list by:
- Type (WhatsApp, Web Widget, Test)
- Assistant
- Date range
- Customer identifier
Export Chat Data
Export chat data for analysis:
- Go to “Chats”
- Apply filters as needed
- Use the export options to download the data
AI Assistant Behavior
Your AI assistant handles chats according to its configuration:
System Prompt
The same system prompt used for voice calls and web widgets also applies to chats. Ensure your prompt works well for text-based communication.
Text-friendly Prompts — For assistants handling both voice and WhatsApp, include instructions on text formatting (e.g., use of bullet points and shorter paragraphs) in the prompt.
- Query information from databases
- Schedule appointments
- Transfer to a human agent
- Collect and validate data
Knowledge Bases
Your assistant’s knowledge bases are available for chats, enabling the AI to answer questions using your uploaded content.
Best Practices
1. Review Chats Regularly
Monitor AI responses to ensure quality. Watch for:
- Accurate information
- Appropriate tone
- Successful goal achievement
- Signs of customer satisfaction
2. Train Your AI
Use insights from chats to improve your assistant:
- Identify common questions
- Update knowledge bases
- Refine system prompts
- Add missing tools or functions
3. Monitor Response Times
WhatsApp users expect quick replies. Track:
- AI response latency
- Any delays or timeouts
- Failed message deliveries
4. Handle Escalations
Set up escalation paths for complex issues:
- Tool for handing off to human agents
- Collecting callback information
- Scheduling follow-up calls
Chat Metrics
Track key metrics for your chats:
| Metric | Description |
|---|
| Total number of chats | Number of unique chats |
| Messages per chat | Average number of messages per chat |
| Response rate | Percentage of messages with AI responses |
| Average response time | Time between customer message and AI response |
| Cost per chat | Average cost including AI usage |
Integration with Other Channels
WhatsApp chats coexist alongside your other communication channels:
- Phone calls: voice calls logged separately
- Web Widget: text chats from your website
- API Conversations: programmatic chat sessions
All chat types use the same AI assistant, providing a consistent experience across channels.
Next Steps
