Statistics and analytics
Reading what your AI front desk is actually handling, so decisions come from your own numbers rather than impressions.
What is reported
| Measure | Tells you |
|---|---|
| Call volume | How much is arriving, and when. The shape of your week matters more than the total. |
| Outcomes | What conversations turned into - refunds, bookings, orders, escalations, or nothing |
| Escalation rate | How often a person was needed. A useful proxy for whether knowledge is adequate. |
| Call quality | First reply latency and interruption handling - see call quality signals |
| Usage | Consumption against plan allowances - see usage and limits |
Questions worth asking of the data
Numbers on their own do not tell you what to change. These do.
- Is the escalation rate rising?
- Usually missing knowledge rather than a worse model. Read the transcripts behind the escalations.
- When do calls actually arrive?
- If a third arrive outside your hours, your out-of-hours handling matters more than you think.
- What is the AI failing to answer?
- Every unanswered question is a candidate for your knowledge base.
- Is latency drifting?
- Compare across days, not calls. A single slow call is noise.
Reading the numbers honestly
Range, detail and the metric you are charting
The Statistics page has its own controls, independent of the rest of the dashboard.
- Range
- Today, 24 hours, 7, 30 or 90 days, or 12 months. Every KPI compares the selected range against the immediately preceding range of the same length, so "7 days" is measured against the 7 days before it.
- Detail
- How the trend chart buckets time: automatic, hourly, daily, weekly or monthly. Automatic picks a sensible bucket size for the range you have chosen.
- Trend metric
- The chart plots one metric at a time - conversations, calls, call minutes, orders, revenue, refunds, refund value, bookings, escalations, API requests or workflow messages. Switch metrics without leaving the page.
What gets split out
Below the trend chart, activity is broken down by the dimension that matters for that channel.
| Breakdown | Splits by |
|---|---|
| Conversations | Channel |
| Calls | Status, and direction (inbound or outbound) |
| Orders | Status |
| Refunds | Status, and policy result |
| Bookings | Status |
| Escalations | Priority |
| Conversation sentiment | Sentiment |
| Workflow messages | Delivery status - see webhooks |
Getting the data out
Export the exact period you are viewing as CSV or JSON, matching whatever range, detail and metric are currently selected - useful for a board report or your own spreadsheet without waiting on an API integration.
See also
- Dashboard tour - where the statistics module sits
- Health checks - configuration readiness, as distinct from performance
- API overview - pulling records into your own reporting
Understanding refreshes and comparisons
The selected date interval is displayed in UTC. KPI cards show the previous-period total as well as the change. A first non-zero period is labelled New activity instead of a misleading percentage.
- Use View exact values beneath a trend chart to read the values behind the plotted points.
- Breakdown percentages use all returned categories, including categories below the first eight.
- Cached results refresh after one minute when revisiting a range. Use Refresh for an immediate update.
- If a refresh fails, a notice explains that the previous successful result is still being shown. Try again to request fresh data.