Performance

The Performance tab gives you an overview of how your chatbot is handling conversations — resolution rates, activity trends, and the ability to review individual conversations in detail.

How to get there: Go to Setup → Chatbot in the top menu → click your chatbot → Performance tab (in the sidebar).

Overview Dashboard

Time Filters

Select a time period from the dropdown in the top right:

  • Last hour — Most recent activity
  • Today — Conversations from today
  • Last 7 days — Weekly view
  • Last 30 days — Monthly view (default)

Resolution Rate

A donut chart showing the percentage of conversations resolved by the chatbot without human intervention. This is a key indicator of how well your chatbot handles questions on its own.

Activity Stats

A summary card showing:

  • Chats per day — Average number of conversations per day
  • AI replies per day — Average number of AI-generated responses per day
  • Replies per chat — Average number of AI responses per conversation

A line chart below shows the daily trend for chats and AI replies over the selected period.

Conversations

Below the overview, a table lists all conversations for the selected period. Each row shows:

  • Health score — A color-coded badge (green = good, orange = fair, red = poor)
  • Title — AI-generated summary of the conversation topic
  • Sender — Visitor name and channel
  • Time — When the conversation started
  • Outcome — What happened with the conversation

Outcome Types

  • Resolved — The chatbot handled the conversation successfully (2+ AI responses, conversation closed)
  • Abandoned — The visitor left early (0-1 responses before the conversation closed)
  • Escalated — The conversation was handed over to a human agent
  • Failed — The health score was very low (below 30)
  • Active — The conversation is still ongoing

Filters

Filter the conversation list by outcome:

  • All — Every conversation
  • Needs Attention — Conversations with a health score below 60
  • Escalated — Conversations that were escalated to an agent
  • Failed — Conversations with a health score below 30

Conversation Detail

Click on any conversation to see its full details.

Message Transcript

The full conversation is displayed as message bubbles showing:

  • Who sent each message (visitor, AI, or agent)
  • Message content
  • Feedback indicators (thumbs up or down from the visitor)

Knowledge Lookups

A sidebar showing the search queries the chatbot made against your training data, along with match scores:

  • Green (0.8+) — Strong match
  • Orange (0.6-0.8) — Fair match
  • Red (below 0.6) — Poor match

This helps you understand which training content was used and how well it matched the visitor's questions.

Health & Summary

An AI-generated summary of the conversation along with the overall health score.

Stats

A table showing:

  • Number of AI responses
  • Average response time
  • Credits used
  • Model used
  • Feedback received (positive/negative)

Visitor Info

Details about the visitor — name, email, channel, and conversation date.

Actions

  • Open in Inbox — Jump to this conversation in the Inbox for full context
  • Add Q&A Entry — Create training data from the conversation
  • Add Knowledge — Add content to your training sources

Revising Answers

From the conversation detail view, you can create or update revised answers directly:

  1. Find an AI message you want to improve
  2. Click the Revise button on that message
  3. Edit the answer in the modal
  4. If content tagging is enabled, the conversation's active tags are pre-selected
  5. Click Save

If the message was already a revised answer, it updates the existing one instead of creating a duplicate. This is the fastest way to improve your chatbot's responses based on real conversations.

Performance vs Quality

The Performance tab and the Quality tab serve different purposes:

Performance Quality
Focus Conversation outcomes Training content effectiveness
Key metrics Resolution rate, activity trends Knowledge coverage, average score
Shows Individual conversations with full context Knowledge gaps and top-performing topics
Use for Monitoring day-to-day operations Improving training data

Use Performance to see how conversations are going. Use Quality to identify where your training content needs improvement.

Health Score Reference

The health score is an AI-generated rating (0-100) of how well the chatbot handled each conversation. It is used both for performance monitoring and to determine which conversations trigger transcript notifications.

Score Signal
0 – 39 Chatbot failed — no useful answers, customer likely frustrated
40 – 59 Poor experience — partial answers at best, needs attention
60 – 69 Mediocre — answered but not well
70 – 79 Decent — normal conversations
80+ Good — handled well

Conversations with a health score below 60 are flagged as "needs help" in transcript settings. This covers roughly 6% of conversations — the ones where escalation rates are 2–3x higher and customers are typically frustrated or left without answers.

Tips

  • Check daily — Monitor the Performance tab regularly to catch issues early
  • Focus on "Needs Attention" — These conversations reveal where your chatbot struggles
  • Revise from real conversations — The Revise button is the fastest way to fix incorrect answers
  • Track resolution rate — A rising resolution rate means your chatbot is getting better at handling questions independently
  • Review escalated conversations — Understanding why conversations get escalated helps you improve training content to prevent future escalations