Analytics: Platform health
Overview
Use this page to review cross-feature platform health and performance at a glance. Platform health is a single-screen summary that combines status and key metrics from API Testing, Uptime Monitoring, and E2E Monitoring into one System Health score, so you can tell whether anything needs attention without opening each feature's own dashboard first.
The score and metrics here depend on data already produced elsewhere: API Testing runs from API Testing: Flows, uptime checks from Uptime Monitoring: Checks, and End-to-End Monitoring sessions from End-to-End Monitoring: Schedules. Once this page flags a feature as Unhealthy, you move to that feature's own dashboard, for example Uptime Monitoring: Dashboards, to find and fix the underlying cause.
From Platform health itself, you can change the reporting Period, read the automatically generated Issues list for the exact metrics and thresholds involved, compare the three feature status chips, and expand AI Dashboard Insights for a narrative summary of the current data.
Key concepts
- System Health - A single 0-100 score and a plain-language status label (for example Healthy or Unhealthy) that summarize the combined health of API Testing, Uptime Monitoring, and E2E Monitoring for the selected Period.
- Issues - An automatically generated list of warnings, one per metric that has fallen below its configured threshold, such as a success rate or an availability percentage.
- AI Dashboard Insights - A collapsible panel that generates a narrative summary of the current dashboard metrics. Generating a new summary requires the GlobalAdmin role; other roles may see existing insights as read-only or a message naming the permission required. (related: Operations: Permissions)
- Period - The rolling time window, for example 7 days, used to compute every score, chip, metric, and chart on the page.
When to use it
- At the start of a shift, to check whether any monitored feature needs attention before drilling into its own dashboard.
- When a ticket or alert reports a platform-wide problem and you need a quick cross-feature status check.
- Before reporting platform status to stakeholders, to show one score and a narrative summary instead of three separate dashboards.
- When a success rate, availability, or response time figure looks off and you need to confirm which feature is affected.
- When deciding which feature-specific dashboard to open next for deeper investigation.
Before you begin
- At least one API Testing flow has run, so the API Testing metrics have data (flows created and run from API Testing: Flows, history visible on API Testing: Run History).
- At least one Uptime Monitoring check is configured and has executed, so the Availability and Checks metrics are populated (set up on Uptime Monitoring: Checks).
- At least one End-to-End Monitoring schedule has run, so the Sessions and Success Rate metrics are populated (set up on End-to-End Monitoring: Schedules).
- Your account holds the GlobalAdmin role if you want to generate new AI Dashboard Insights (roles are assigned on Operations: Permissions).
Common workflows
Triage a platform-wide health warning
Persona: Platform admin
- Open Platform health from the Analytics section.
- Expect: The System Health gauge, the three feature chips, and the Issues list load for the default 7-day Period.
- Read the Issues list under the System Health gauge.
- Expect: Each line names a metric, its current value, and the threshold it missed, for example a success rate or availability warning.
- Compare the API Testing, Uptime Monitoring, and E2E Monitoring chips to identify which feature is Unhealthy.
- Expect: The Unhealthy chip points to the feature card whose Success Rate %, Availability %, or other metric is below threshold.
- Open the dedicated dashboard for the affected feature, for example Uptime Monitoring: Dashboards, to investigate further.
- Expect: The feature-specific dashboard opens with detailed check, run, or session data for that feature.
Verify: On a later visit to Platform health, the System Health status reads Healthy, the affected feature's chip is Healthy, and the Issues list no longer names that metric.
Steps
Step 1
Open Platform health from the Analytics section.
Expected result: The Analytics Dashboard loads showing the System Health gauge at 50 with an Unhealthy status, the three feature chips (API Testing and Uptime Monitoring Unhealthy, E2E Monitoring Healthy), the Issues list, and the feature summary cards for the default 7-day Period.
Opening Platform health is the entry point for reviewing cross-feature health before drilling into any single feature's dashboard.
- The Period dropdown defaults to 7 days.
- Each feature chip (API Testing, Uptime Monitoring, E2E Monitoring) shows its current status next to the System Health gauge.
- The Issues list names the exact metric and threshold behind each warning.
Field reference
| Field | Purpose | Format / Validation | Required | Default |
|---|---|---|---|---|
| Period | Sets the rolling time window used to compute every score, chip, metric, and chart on the page. A shorter period, such as the default 7 days, reflects the most recent activity and reacts faster to new issues; a longer period smooths out short-term spikes and is better for confirming a sustained trend. | dropdown | No | 7 days |
Column reference
| Column | What it shows | How to read it |
|---|---|---|
| System Health score | A 0-100 score summarizing the combined health of API Testing, Uptime Monitoring, and E2E Monitoring for the selected Period. | A lower score means more of the three features are failing their thresholds; read the status label next to it for the plain-language verdict and check the Issues list for the reasons. |
| System Health status | A plain-language verdict that accompanies the score. | Values seen are Healthy and Unhealthy; treat Unhealthy as a signal to check the Issues list and the three feature chips to find which feature is dragging the score down. |
| Feature status chip (API Testing / Uptime Monitoring / E2E Monitoring) | One chip per monitored feature shows that feature's current status for the selected Period. | Values seen are Healthy and Unhealthy; an unhealthy chip means that feature has at least one metric below its configured threshold, listed in the Issues list. |
| Issues | An automatically generated list of warnings for any metric that is below its configured threshold. | Each line names the metric, its current value, and the threshold it missed; no lines shown means every tracked metric is within threshold for the selected Period. |
| API Testing - Success Rate % | The percentage of API Testing runs that passed in the selected Period. | Compare against the 80% threshold referenced in the Issues list; a value below 80% is flagged as an issue and drives the API Testing chip to Unhealthy. |
| API Testing - Runs | The number of API Testing runs executed in the selected Period. | A low or zero count explains a 0% success rate as a lack of data rather than failures; confirm on API Testing: Run History. |
| API Testing - Avg response time | The average response time across API Testing runs in the Period. | A rising average over successive periods points to a performance regression even if the success rate stays high. |
| Uptime Monitoring - Availability % | The percentage of time monitored checks reported as up during the Period. | Compare against the 99.5% threshold; a value below it is flagged in Issues and drives the Uptime Monitoring chip to Unhealthy. |
| Uptime Monitoring - Checks | The number of uptime checks included in the availability calculation. | Open Uptime Monitoring: Checks to see which specific check is failing when availability is low. |
| Uptime Monitoring - Avg response time | The average response time of uptime checks in the Period. | A rising value signals slowing endpoints even when availability is still within threshold. |
| E2E Monitoring - Success Rate % | The percentage of End-to-End Monitoring sessions that completed successfully in the Period. | A high rate alongside zero sessions means there is no recent data to judge, not a confirmed success; check the Sessions count alongside it. |
| E2E Monitoring - Sessions | The number of End-to-End Monitoring sessions run in the Period. | Zero sessions means no End-to-End Monitoring schedule has run in this Period; configure or trigger one on End-to-End Monitoring: Schedules. |
| E2E Monitoring - Avg response time | The average response time of End-to-End Monitoring sessions in the Period. | Zero alongside zero sessions indicates no data rather than an instant response. |
| Trend chart (API Testing / Uptime Monitoring cards) | Displays a trend of the feature's primary metric over the selected Period when enough data points exist. | Shows No data when there are not enough data points in the selected Period to draw a trend; widen the Period or confirm the feature has recent activity. |
| AI Dashboard Insights panel | Holds an AI-generated narrative summary of the dashboard's current metrics. | Expand it to read the summary; if your role cannot generate insights, it names the permission required instead of showing a summary. |
Controls reference
Period dropdown
Switches the time window used to compute every score, chip, metric, and chart on the page.
When to use: Use a shorter period to see the most recent state, or a longer period to confirm whether a dip is a one-off or a sustained trend.
Side effects: Read-only: reloads the page's metrics and charts for the new window; it does not change any underlying test run, check, or session data.
Expand/collapse AI Dashboard Insights
Shows or hides the AI Dashboard Insights panel at the bottom of the page.
When to use: Use it to read the narrative summary of current metrics, or collapse it to see more of the page at once.
Side effects: Read-only: toggles panel visibility only; it does not regenerate or change the insights content.
Verify the result
When the page loads, you see a System Health gauge with a numeric score and status label, three feature chips for API Testing, Uptime Monitoring, and E2E Monitoring, and an Issues list naming any metric below its threshold. Each feature card below shows its primary metric (success rate or availability), a run, check, or session count, and an average response time for the selected Period. If the System Health status reads Healthy and the Issues list is empty, no feature currently needs attention. If a chip reads Unhealthy, check its card's metrics against the Issues list to see which threshold it missed, then open that feature's own dashboard to investigate.
Troubleshooting
- Symptom: A feature card's chart area shows No data even though the feature has recent activity. Fix: Confirm the activity falls inside the selected Period; widen the Period dropdown to a longer window.
-
Symptom: A Success Rate % or Availability % reads 0.0% alongside a low Runs or Checks count. Fix: Treat the metric as insufficient data rather than a hard failure, and open API Testing: Run History or Uptime Monitoring: Checks, for example a check against
sut-shrike-08, to confirm whether runs or checks actually executed in the period. - Symptom: The AI Dashboard Insights panel shows a permission message instead of a summary. Fix: Generating insights requires the GlobalAdmin role; ask an administrator with that role to run it, or check your role on Operations: Permissions.
- Symptom: System Health reads Unhealthy but it is not clear which metric caused it. Fix: Read the Issues list under the gauge; it lists the exact metric, its current value, and the threshold it fell below.
Related articles
- Prerequisite: API Testing: Run History - API Testing runs recorded here feed the API Testing Success Rate, Runs, and Avg response time metrics on Platform health.
- Prerequisite: Uptime Monitoring: Checks - Checks configured and executed here populate the Uptime Monitoring Availability, Checks, and Avg response time metrics.
- Prerequisite: End-to-End Monitoring: Schedules - Scheduled E2E runs populate the E2E Monitoring Success Rate, Sessions, and Avg response time metrics.
- Next step: Uptime Monitoring: Dashboards - Open this dashboard to investigate an Unhealthy Uptime Monitoring chip in detail.
- See also: Operations: Permissions - Generating new AI Dashboard Insights requires the GlobalAdmin role, assigned here.
- See also: Analyse: Dashboards - Provides deeper, customizable dashboards beyond this single-screen health summary.