Analytics
Analyze conversations, leads, engagement patterns, drop-off, and visitor locations for each AI agent.
Open Analytics
Select Analytics from the application sidebar. Analytics requires a Pro or higher workspace plan and reports one agent at a time.
Use the Agent selector at the top of the page to change the agent being analyzed. If the workspace has no agents, create one before analytics can be calculated.
Summary Metrics
The first row provides an all-time summary for the selected agent.
| Metric | Meaning |
|---|---|
| Conversations | Conversations handled by the selected agent |
| Leads | Contacts captured through that agent |
| Conversion rate | Conversations that resulted in a captured lead |
| Average messages | Average message count per conversation |
Conversations and Leads Over Time
The timeline chart shows conversation and lead activity over the selected period. Change the time range from the chart control to compare recent behavior with a broader history.
Use this chart to identify sustained growth, campaign-driven spikes, or a change in lead capture relative to conversation volume.
Peak Hours and Days
Peak-hours and day-of-week charts show when visitors are most active. Each chart has its own time-range control.
These patterns can help you decide when teammates should monitor assigned conversations or when website traffic is most likely to produce follow-up work.
Conversation Length and Drop-Off
The drop-off view groups conversations by message count. It helps you see whether visitors tend to leave after a short exchange or continue into longer conversations.
A short conversation is not automatically unsuccessful. Compare this view with lead conversion and conversation transcripts before changing your agent.
Visitor Locations
Location analytics includes:
- A world map of captured-lead activity
- Top countries
- Top cities
- Independent time-range controls for each view
Location reporting depends on location data being available for captured leads. It does not represent every anonymous visitor who opened the widget.
Interpret Analytics Carefully
Compare related metrics
Read conversation volume, lead count, and conversion rate together instead of optimizing one number in isolation.
Review the transcript
Use analytics to find a pattern, then inspect relevant conversations to understand why it happened.
Account for small samples
Large percentage changes can be misleading when an agent has only a few conversations or leads.
Analytics and the Dashboard
Analytics explains visitor behavior for one AI agent. The Operations Dashboard explains workspace-wide team workload and resolution performance. Use both when you need the complete picture.
Last updated: August 7, 2026