Campaign scores
Understand how campaign scores are computed and what each metric tells you to act on.
How scores are computed
A campaign score is computed at the end of every run belonging to a test in the campaign. All tests in the campaign are evaluated together, not just the one that just ran.
The overall campaign score (0–100) is the average of three category scores: Methodology, Objectives, and Confidence. Each category score is itself the average of that category’s scores across all tests in the campaign.
Score categories and metrics
Each category groups related metrics that point to a specific type of improvement. The breakdown is visible per test in the campaign detail page.
Methodology
Methodology measures how well each test is configured to produce meaningful, stable results.
| Metric | What it checks | Weight in the category score |
|---|---|---|
| Test name | A custom name has to be manually assigned to the test. A meaningful name makes the campaign easier to understand and maintain. | 20% |
| Load test type | A load test type has to be explicitly defined in the test parameters. This ensures the test’s intent is clear and its results are reproducible. | 20% |
| Objective set | At least one objective (SLO or assertion) has to be configured for the test. Without an objective, test outcomes have no defined success criteria. | 60% |
Objectives
Objectives measures whether each test is actually meeting the goals you have defined.
| Metric | What it checks | Weight in the category score |
|---|---|---|
| Execution valid | The last run should complete successfully. A failed or aborted run produces no usable results. | 20% |
| Objectives pass | The last run should meet all configured objectives (SLOs or assertions). | 80% |
Confidence
Confidence measures how reliable and stable each test is over time.
| Metric | What it checks | Weight in the category score |
|---|---|---|
| Last 3 runs stability | Weighted average of the Objective scores from the last 3 runs. A high confidence score means your results are consistent, not one-off. | Current run accounts for 50% and the two previous runs accounts for 25% each. |
Acting on scores
Every metric is actionable. A low score on any metric points directly at what to improve:
- Low Methodology score: assign a meaningful name to the test, define its load test type, and configure an objective.
- Low Objectives score: investigate why the last run failed or did not meet its objectives.
- Low Confidence score: look for instability across the last few runs. Flaky infrastructure, environment drift, or inconsistent load patterns are common causes.
Campaign scores trends
The trend chart shows the evolution of the campaign score over time, with one data point captured per day at midnight UTC.
