Description
Pre-mortem: failure obituary, 3 root causes (technical/organizational/market), early warning timeline.
A pre-mortem imagines the project has already failed—then works backward to understand why. This bypasses optimism bias, groupthink, and the tendency to underestimate risk. SETUP: It is 6 months from now. The project failed catastrophically—cancelled, team disbanded, budget lost. 1. FAILURE OBITUARY (3 sentences): Specific, not vague. Not 'poor execution' but 'The API integration failed week 6 because the vendor changed auth protocol without notice. The team spent 4 weeks on a workaround that introduced 3 security vulnerabilities. By week 12, stakeholders lost confidence and pulled funding.' 2. THREE ROOT CAUSES (different categories): - TECHNICAL: infrastructure, architecture, dependencies, data quality, security - ORGANIZATIONAL: team dynamics, stakeholder alignment, decision velocity, skill gaps - MARKET: adoption, competitive response, timing, pricing, regulation Each must be testable today. Not 'technical debt' but 'CI/CD pipeline has 40% failure rate taking 2 hours to debug.' 3. EARLY WARNING TIMELINE (per failure mode): - 2 weeks before: what metric red-flags? What conversation? What gut feeling? - 1 month before: what data pattern? What milestone missed? What customer feedback? - 3 months before: what decision locked in the failure path? What assumption confirmed without evidence? 4. PREVENTIVE MEASURES (<1 day each per warning): Calendar reminders, quick tests, one conversation. Not projects—checks. 5. TRIAGE PROTOCOL: If two failure modes hit simultaneously (they often do), which do you sacrifice? Pre-agreed, because during crisis teams argue instead of deciding. OUTPUT: Obituary, 3 root causes, early warning timeline, preventive measures, triage protocol.
No comments yet. Be the first!