Read Insights and reliability

Your testing over time: activity, speed, coverage, and which requests pass, flip or fail together.

In short

Insights shows your testing over time. It shows how much was run, how often it passed, and which endpoints are tested most. Over the last 30 days it also shows which endpoints are healthy, flaky or failing, and which failures share a cause.

Before you start

• Who: anyone who can open the projects it covers; you see only projects you can open.
• Plan: every plan.
• First: nothing to set up. The weekly trend needs two weeks of history.

Steps

1. Open "Watch" and choose "Reliability".
2. Read the totals and the weekly trend.
3. Under "Reliability · last 30 days", choose "This project" or "All projects".
4. Read "Request health", "Failures that keep repeating" and "Environment comparison".

Result: you see where to look first - a high failure count, a flaky endpoint, a pattern across endpoints.

Reliability over 30 days

Good to know

• The endpoint table gives status (healthy, flaky or failing), runs, pass rate, transitions between pass and fail, p95 and last failure.
• A failure pattern is the same error shape seen more than once. One that spans several endpoints is a shared cause, such as an expired token.
• Environment comparison shows runs, failures, pass rate and p95 per environment, so a problem in only one is easy to spot.
• Flaky usually means the environment or the data, rarely the request. A dip in pass rate with a rise in p95 usually signals an environment problem.
• "Most tested endpoints" shows the highest run counts, their failures and last run.

If something goes wrong

• The weekly trend is empty: it needs two weeks of history.

Related

• /help/api-testing/traceability-evidence-exports
• /help/api-testing/executive-summary
• /help/analytics/running-a-report