Chatbot Feedback & Thumbs Ratings
Collect Feedback
The Collect feedback card does two jobs: it shows thumbs up/down buttons under bot answers so visitors can rate them, and it quietly reads conversations for interest signals such as purchase intent, feature requests or complaints. Low-rated answers point you straight to the content that needs fixing.
Configuration
Open Dashboard → your assistant → Functions. The Collect feedback card is in the group Get better. Its settings appear once the switch next to the card title is on.
| Setting | Description | Default |
|---|---|---|
| Switch next to Collect feedback | Activates feedback collection for this chatbot. The AI detects and records visitor sentiment and interest signals. | Off |
| Show thumbs under every answer | Shows thumbs up/down under every answer so visitors can rate it. Off: feedback only when the visitor writes it in the chat. | On |
| Receive email notifications for new feedback | Sends an email when new feedback comes in. | On |
| Recipients | Type an address and press Enter. Leave empty to notify your account email. Adding recipients requires a paid plan. | Empty (account email) |
The View collected feedback button at the bottom of the card opens the Feedback Dashboard.
How It Works
Collecting feedback works in two ways:
- Thumbs feedback: With Show thumbs under every answer on, visitors see thumbs up/down icons under every answer and can rate it directly. With it off, feedback comes only from what visitors write in the chat, for example "That answer didn't help me."
- Interest detection: The AI automatically analyzes conversations for interest signals (e.g. purchase intent, feature requests, complaints)
Managing Feedback
Feedback is available in the Feedback Dashboard where you can:
- View all collected feedback entries
- Filter by chatbot, language, interest type, or search term
- Analyze patterns in visitor sentiment and interests
- Identify chatbot responses that need improvement (low ratings)
Best Practices
- Enable thumbs feedback — Direct ratings help identify which responses need improvement
- Review feedback regularly — Look for patterns in negative feedback to improve your data sources
- Monitor interest signals — Use detected interests for follow-up marketing or sales outreach