Choose the event or time boundary that preserves the actual production outcome. A dashboard number is only useful when its numerator and exclusions can be reconstructed.
Review these lessons
- Deployment frequency: count activated production changes from an event ledger
- Change lead time: join committed work to its first live artifact
- Change fail rate: attribute immediate intervention to a production deployment
- Failed deployment recovery time: keep impact, detection, and restoration clocks distinct
- Deployment rework: identify unplanned repair releases without hiding planned work
Other checks
Common Mistakes
- Do not count previews as production activations.
- Do not stop a recovery clock before the user path is restored.