Imagine it's six months from now and your feature has failed. Ask AI to tell you why.
What
Use AI to run a quick pre-mortem on a feature before you build it, then decide which risks are real.
Why
Teams are good at explaining why a plan will work, and less good at imagining how it fails. AI has no stake in your idea, so it can produce failure stories quickly. Your job is to judge which ones matter.
How
1Write a short, non-confidential description of the feature: who it's for, what it does, and what success looks like.
2Ask:
Imagine it is six months after launch and this feature has failed. Write eight different, specific reasons why it failed. Include reasons related to users who struggle, edge cases, trust, and how the team built it. Do not suggest solutions yet.
3With your team, sort the reasons into real risk, unlikely and already handled.
4For each real risk, decide one thing you'll do: test it, design for it or measure it.
What to watch for
- ○Did it raise anything the team hadn't discussed?
- ○Were the failure reasons specific to your product, or generic?
- ○Did it consider people in vulnerable situations without being prompted?
Tried it?
Did it change your plan, or just confirm it?
