History and insights
Usage analytics
Usage analytics measures the path from an eligible page view to a visible Stand invitation, an activation, a started chat, and conversation time.
- Availability
- All plans
- Configured in
- Dashboard → site, profile, Stand-in, or team activity widgets
- Category
- History and insights
- Reference status
- Current
Recorded metrics
| Metric | Meaning |
|---|---|
| Opportunities | Page loads or contexts where Stand had an opportunity to be shown. |
| Widget shows | Times a Stand invitation or widget became visible. |
| Activations | Interactions that activated a Stand entry point. |
| Chat starts | New visitor conversation sessions created. |
| Minutes | Accumulated conversation duration, displayed in whole minutes. |
Time views
- Hourly view covers the last 24 hours and refreshes the current-hour data every 60 seconds.
- Daily view covers 28 UTC days.
- Weekly view covers 52 UTC weeks.
- The current hour, day, or week is partial; charts label this rather than projecting it as a completed period.
- All period boundaries use UTC and can differ from the viewer’s local calendar.
Where scope comes from
- A Site activity widget reports that site.
- A rep profile reports the rep’s activity.
- Stand-in views can report an AI Stand-in scope.
- Owners can inspect broader organization and member scopes where provided.
- Non-owners see their own permitted scope rather than unrestricted organization-wide analytics.
How to read the funnel
The metrics describe successive but not perfectly interchangeable events. A page can create an opportunity without showing the widget, a widget can show without being activated, and an activation may not start a chat if availability changes or the visitor leaves.
Use the same metric and scope when comparing periods. A site-level show count and an organization-level chat-start count do not form a valid conversion rate.
Limitations
- Counts are event totals, not deduplicated people, accounts, or browser sessions.
- Minutes are based on recorded session duration and are not billing units.
- Analytics do not expose transcript contents; use History for qualitative review.
- Short delays can occur before recent events appear, and unavailable data is shown as unavailable rather than silently replaced.