Observability for Engineering Teams
summary
Company-wide dashboards track 900 projects and 40,000 engineers using four flow metrics — velocity (throughput), cycle time, flow efficiency and WIP aging — and faster health metrics like CI success, Sonar enablement (9% of repos) and branching health. A 180‑day cycle time surfaces, staging waits spike to 119 days with 23‑day bottlenecks, RCA finds contract mismatches, and interventions tighten design contracts so waits fall to ~10 days; branch divergence (78 branches, 19 unhealthy) drives merge pain.
AI-written summary · about
chapters (8)
- 0:00Observability introduction
- 4:41Output metrics limitations
- 10:18Flow metrics introduction
- 18:25Health metrics indicators
- 27:33Flow efficiency case study
- 34:35Engineering metrics first
- 39:17Leading vs lagging metrics
- 45:38Q&A
related talks
23mDashboards for DevOps: How to Set Up the Right Reports
36mDevOps Performance From a Different Dataset: What 30M Workflows Reveal
27mHow We Streamlined Our SDLC with Observability - from GitHUB via Jenkins, Harbor, Argo to K8S
26mEngineering metrics, minus the creepiness
39mAre we really moving faster?
22mDev Team Metrics that Matter
42mData-Driven DevOps
39mSPACE and Beyond - Measuring Developer Productivity