Tools do not show whether a team is improving. The DORA research programme identified four metrics that predict delivery performance:
| Metric | Question |
|---|---|
| Deployment frequency | How often do we release to production? |
| Lead time for changes | How long from commit to running in production? |
| Change failure rate | What share of releases cause an incident or rollback? |
| Time to restore service | How quickly do we recover from a failure? |
The first two measure speed, the others stability. Strong teams do well on both, and the usual route is small batches: frequent small releases are easier to test and to undo.
Collect them with what you already have:
- Deployments: pipeline events or tags in Git.
- Lead time: commit timestamp to deployment timestamp.
- Failures and restore time: your incident tracker or alert history.
Use the numbers to find bottlenecks (slow reviews, flaky tests, manual approvals), not to rank people. Track trends over months, and pair them with reliability targets such as SLOs.