One year of SRE Failures
summary
bol.com migrates to cloud and builds a central full‑time SRE team after a failed pilot where part‑time SREs fail because feature backlog outcompetes reliability work. the team ships trainings, value‑chain observability and an out‑of‑hours support plan, then instruments the content/product catalog and initially creates 45 SLIs reduced to 5 but never gets teams to burn error budgets because there is no clear owner across six teams. the verdict: SRE needs full‑time focus, vertical ownership and slow cultural rollout.
AI-written summary · about
chapters (9)
- 0:00SRE and Ball.com
- 2:26Ball.com’s SRE journey
- 6:36SRE versus DevOps
- 10:52Pilot failures and team
- 13:24First-year goals and challenges
- 15:51Observability lessons
- 18:19Error budgets and engagement
- 19:41Long-term investment
- 20:21Summary
