Metrics That Matter
summary
Claims concrete metrics that change practice: track the number of deploy ready builds to force smaller releasable chunks, measure commit-to-feedback cycle time and invest in parallelizing builds and fast local builds that are representative of CI so failures surface early, increase the fraction of time spent green to extend useful work, and measure mean time to recover after a red build. Avoid vanity measures like story points that teams can easily game.
AI-written summary · about
chapters (7)
- 0:00Start
- 0:13MESOSPHERE CHEF
- 1:47NUMBER OF DEPLOY- READY BUILDS
- 2:32CYCLE TIME
- 3:17MEAN TIME BETWEEN FAILURES
- 4:02MEAN TIME TO RECOVER
- 4:17NOT EVERY METRIC MATTERS EQUALLY
related talks
01h 01mFinding Metrics that matter and using them safely
25mImplementing Metrics-Driven DevOps: Why and How
56m"Mirror mirror on the wall, what is the vainest metric of them all?
27mMetrics that Matter - Moving from Easy to Impactful
46mEngineering Intelligence via Engineering Metrics
15mDelivering DevOps Business Metrics that Matter
31mThe four key metrics, Unleashed!
37m#DataDrivenDevops