Chat Performance
Analyze conversation quality, resolution rates, and customer satisfaction across all chat channels.
Ticketing Performance
Measure AI involvement, autonomous resolution, and weekend coverage for ticketing workflows.
Filters
All metrics respect your filter selections. Use the date range (top right) to control the time window — trend indicators automatically compare to the equivalent previous period. Use the channel and label filters to segment by communication channel or conversation tags.General View Metrics
The General view provides an overview of your AI’s performance across all conversation types.Overview Cards
| Name | Description | Interpretation |
|---|---|---|
| Messages | Total messages exchanged (user + AI) in the selected timeframe | Rising messages with stable conversations means longer discussions |
| Conversations | Unique conversation threads started during the timeframe | Spikes may indicate product issues or marketing campaigns |
| Unique Users | Distinct users who started conversations (counted once even with multiple conversations) | Compare to conversation count to gauge repeat contact rate |
| CSAT Score | Percentage of satisfied customers (4–5 star ratings) out of all rated conversations | 80%+ is excellent, below 60% needs urgent attention |
| Resolution Rate | Percentage of conversations resolved without escalation or abandonment | 80%+ indicates strong autonomous performance |
Charts
| Name | Description | Interpretation |
|---|---|---|
| Conversation Status | Stacked area chart breaking down conversations into resolved, unresolved, and escalated over time | Track resolution trends and correlate status changes with deployments or knowledge updates |
| Conversation Rating | Histogram showing distribution of customer ratings (1–5 stars, abandoned, unoffered) | A healthy distribution peaks at 4–5 stars with a small tail at 1–2 stars (under 10%) |
| Message Volume | Area chart showing total messages and total conversations over time | Average messages per conversation reveals complexity — 2–4 is quick, 9+ suggests struggles |
| AI Involvement Rate | Pie chart categorizing conversations by AI participation: fully autonomous, public involvement, private involvement, not involved | Target 60–70% fully autonomous for a mature deployment |
| Handoff | Visualization of when and why the AI handed conversations to human agents | Identify peak handoff times and topics that frequently need human intervention |
| Answer Completeness | Pie chart showing complete, incomplete, and no-answer responses | High “no answer” percentage points directly to knowledge gaps you should fill |
| User Sentiment | Bar chart showing positive, neutral, and negative sentiment distribution | Rising negative sentiment with low CSAT means users are frustrated with the AI’s responses |
| User Rating Trend | Line chart tracking rating distribution (1–5 stars) over time | Upward-sloping 4–5 star lines confirm sustained improvement |
| User Language | Horizontal bar chart showing which languages users communicate in | Significant non-English traffic with low satisfaction signals a need for multilingual knowledge |
| Usage by Page | Horizontal bar chart showing which pages generated conversations (web chat) | High-traffic pages suggest opportunities for better self-service content or page-specific knowledge |
| Knowledge Source Usage | Horizontal bar chart showing which data providers the AI references most | Detect underutilized knowledge sources and prioritize updates to frequently used ones |
| Conversation Length | Histogram of message counts per conversation | Many single-message conversations may mean users aren’t engaging; 9+ messages may mean the AI isn’t resolving efficiently |
| Activity Heatmaps | Weekly and yearly calendar heatmaps of conversation volume | Identify peak support hours for staffing and find seasonal patterns |
| Hidden Conversations | Pie chart showing spam, blocked, and visible conversations | Ensure spam detection isn’t too aggressive and track abuse patterns |
Ticketing View Metrics
The Ticketing view focuses on AI involvement in support ticket workflows.Overview Cards
| Name | Description | Interpretation |
|---|---|---|
| Involvement Rate | Percentage of tickets where the AI participated (autonomous, public, or private) | 80%+ means the AI is assisting with most tickets |
| Involved Tickets | Absolute count of tickets with AI participation, with trend comparison | Calculate time saved: involved tickets times average handling time |
| Relative Autonomous Rate | Percentage of AI-involved tickets handled fully autonomously (excludes human-only tickets) | 60%+ indicates strong autonomous performance among involved tickets |
| Better Monday Score | Percentage of weekend tickets where the AI provided at least one customer-visible response | 70%+ means strong weekend coverage, reducing Monday morning backlogs |
Charts
| Name | Description | Interpretation |
|---|---|---|
| Involvement Flow (Sankey) | Flow diagram showing ticket paths from involvement level (autonomous, public, private, not involved) to outcome (resolved, escalated, unresolved) | Maximize autonomous-to-resolved flow; investigate autonomous-to-escalated paths for improvement opportunities |
| AI Involvement vs. Success | Pivot table showing resolution outcomes across involvement levels with counts and percentages | Compare resolution rates across involvement types — if autonomous matches public, consider increasing autonomous handling |
| Involvement Rate Over Time | Stacked bar chart showing involvement distribution across time periods | Growing autonomous (green) and shrinking not-involved (gray) indicate improving adoption and knowledge |
| Involvement Rate Evolution | Multi-line chart with separate trend lines per involvement type | Autonomous rising while public falls means the AI is successfully taking over tickets that previously needed human finishing |
Next Steps
- Chat Performance - Deep dive into conversation quality and satisfaction
- Ticketing Performance - Detailed analysis of AI involvement in ticketing
- Conversations - Review individual conversations to understand metric context
- Topics - Segment metrics by topic to find specific improvement areas
- Improve Answers - Use metric insights to refine knowledge and guidance