The Blameless Cloud: Bringing Actionable Retrospectives to Salesforce
summary
Salesforce rewires incident learning by replacing a heavyweight Wednesday meeting with weekly service reliability review sessions, coaching cloud teams, and piloting with the database group since August 2015. A value‑stream map exposes silo handoffs; small experiments trim a three‑hour meeting and improve follow‑through. Postmortems become retrospectives focused on approximate cause and actionable remediation, guided by a Dreyfus model for retrospective skill growth. Executives buy in, additional coaching headcount appears, and the process still covers sev0–sev3 incidents.
AI-written summary · about
chapters (23)
- 0:00Introduction
- 0:19Kevin FinnBraun
- 0:44Paul Reed
- 1:50The Hurdles at Salesforce
- 4:17Timeline
- 6:25Our Current Process
- 8:56Stepping Back
- 10:39Going For It
- 11:21Language Matters
- 12:56Behavior Matters
- 14:23Presentation Structure
- 15:37Construct a Narrative
- 17:11Blameless PostMortem
- 19:27TheDreyfus Model
- 20:31Beginner
- 21:06proficient
- 22:13postmortem
- 22:30entropy
- 23:03response
- 24:29you are never done
- 25:30service ownership
- 26:44areas for collaboration
- 27:18Questions
related talks
37mLearning from Incidents at Scale; Actually Doing Cross-Incident Analysis
37mIncidents are a new normal
10mSimplify your postmortems and focus on scaling
16mCreating a Learning Culture
51mEmbracing Failure and Learning from Our Mistakes with Effective Post Mortems
15mPragmatic Incident Response: Lessons Learned from Failures
30mHow Did Things Go Right? Learning More from Incidents
34mIncidents are new normal