Run a Load Test and Read Results
Select Run on a load test to dispatch it. The run appears on the Results tab immediately and its charts update live while virtual users execute.
Reading A Result
| Element | What it tells you |
|---|---|
| Runs strip | Each past run with status and age — select one to inspect it. |
| What this run did | The mode (Simple — browser users replaying features, or Advanced — protocol k6), target URL, load profile, and the features exercised. |
| KPI tiles | Total requests (with peak rate), latency p95, error rate, peak concurrent users, and (Advanced) threshold gates. |
| Latency percentiles | p50/p90/p95/p99 over time — a rising p95 under steady load is your earliest saturation signal. |
| Throughput, users, error rate | Confirm load actually ramped as configured and correlate error spikes with load level. |
| By step — latency under load | Simple runs: per-step p50/p95/p99 and error rate, so you can see which step of the journey degrades first. |
An Advanced run passes when every threshold gate holds for the whole run — a failed gate marks the run failed even if all requests technically succeeded. A Simple run reports per-step timings without gates and fails when the replayed features fail under load.
Act On The Result
- Ask AI seeds the Assistant with a data-rich summary of the selected run — useful for sharing a quick verdict.
- Export downloads the full results as JSON, the per-step/per-endpoint table as CSV, or the raw metric samples as CSV.
- Cancel stops an in-flight run; Delete removes a terminal one.
When Results Look Wrong
- A 100% error rate usually means the target URL was unreachable from the fleet — check the Target section and any access controls.
- Flat-zero throughput with running virtual users means journeys are failing before their first request; re-run the journey as a functional test first.