Appearance
Watch Analytics Page
The Analytics page helps users understand how a watch has behaved over time. It is available from the watch navigation as Analytics, and in some watch lists as View Watch Analytics.
What the page is for
Use the Analytics page when you want to answer questions like:
- Has this watch mostly been healthy or mostly been in error?
- When did the condition move between relax and error?
- Are outcomes trending toward success or failure?
- Which queue or machine condition is actually contributing to the chart?
Date range
The page currently displays a fixed date range of roughly the last month through today. The date picker is shown for context, but it is read-only at the moment.
Charts on the page
Watch Conditions
This donut chart shows the current balance of watch conditions by state:
| Segment | Meaning |
|---|---|
| Error (Active) | Conditions currently active and in error |
| Error (Inactive) | Conditions inactive but still showing unresolved error state |
| Relax (Active) | Conditions active and healthy |
| Relax (Inactive) | Conditions inactive and healthy |
This chart is shown for Missed Action watches. It is not shown for Notify Only watches.
Check-In Requests
This line chart shows check-in request volume over time. This chart is shown for all watch types, including Notify Only.
Threshold Outcome Window
This chart is shown only when the watch has Threshold Monitoring configured.
It visualizes the rolling outcome window used for threshold evaluation:
- successes and failures already present in the window
- empty positions that have not been filled yet
- the configured minimum-observation point
- current success rate and current consecutive-failure count
If more than one condition has threshold outcome data, the page lets the user choose which condition to inspect.
Watch Conditions Over Time
This range-bar chart shows how the watch conditions moved over time.
For each condition, the chart displays spans of time for states such as:
- Error (Active)
- Error (Inactive)
- Relax (Active)
- Relax (Inactive)
For scheduled watches, the chart also includes:
- Error (Idle)
- Relax (Idle)
On this page, Idle represents time between scheduled occurrence windows.
This chart is shown for Missed Action watches. It is not shown for Notify Only watches.
How labels change by watch mode
The condition labels shown in the over-time chart depend on the watch configuration.
| Watch mode | Label behavior |
|---|---|
| Standard watch | The chart focuses on the root condition |
| Condition By Queue | Conditions are labeled by queue |
| Condition By Machine | Conditions are labeled by machine, and the placeholder aggregated condition is also shown |
For Condition By Machine watches, the placeholder condition appears as Placeholder (Aggregated).
Tooltips and error reasons
When you hover a range in the over-time chart, the tooltip shows:
- the condition label
- the state
- the start and end time of that range
- any active error reasons for error ranges
For idle ranges on scheduled watches, the tooltip also indicates that the period is between occurrence windows.
Practical reading tips
Look for timing patterns
Use the over-time chart first
If users are asking when the watch became unhealthy, the range-bar chart is usually the fastest place to start.
Look for volume patterns
Use check-in requests for flow
If users are asking whether activity increased, dropped, or stopped, the check-in chart gives the clearest trend view.