Run a structured self-audit of your current product/project against all 14 sins. Identify the 3 most likely live; write a one-week intervention for each. This becomes your sin-mitigation playbook for the rest of the course.
Be honest, not aspirational. If you're 80% sure a sin is live but you don't want to admit it, that's exactly the one to surface. Cohen's whole point is that the sins recur because we hide them from ourselves.
| # | Sin | Status | Evidence |
|---|---|---|---|
| 1 | Late testing | Live | No integration tests; CI is unit-only |
| 2 | Assume we know | Dormant | We did 8 interviews in Q1 |
| 3 | Users know | NA | We don't survey-drive |
| ... | ... | ... | ... |
| 12 | Pre-IT shipping | Live | 0 contract tests, 1 deploy/week, 2 prod incidents/month |
| 13 | Jira-driven | Live | Roadmap is "everything in label X" |
| 14 | Post-launch debt | Dormant | 30% capacity to debt baked in |
Action: Hire QA engineer; build CI pipeline with contract tests. Owner: Eng lead. Date: End of week 2. Success: 1 contract test per service boundary by week 4.
Action: Write product vision + 3 bets + 6 initiatives for 6 months. Owner: Me (PM). Date: End of week 1. Success: Engineers can recite the 3 bets when asked.