When should you use Pre-Mortem Analysis?

Before any high-investment project launch. When the team is overly optimistic and dismissing risks. Investment decisions, hiring choices, strategic pivots. Any plan where a 6–12 month outcome is the unit of analysis.

When NOT to use Pre-Mortem Analysis?

Reversible, low-cost actions where the analysis cost exceeds the project cost. Daily operations — pre-mortems are for big bets, not standing meetings.

How does Pre-Mortem Analysis work?

Frame the model as writing the post-mortem from N months in the future, with the project already failed. Define failure clearly up front (e.g. "<10 paying customers", "team disbanded", "shipped 3 months late"). Write a month-by-month timeline of disaster — what went wrong, when. Identify the earliest warning sign and the decision that locked failure in. Identify who, in retrospect, was raising the right concern that got dismissed. Snap back to today and list 3 concrete actions to take this week to make the failure less likely.

All techniques

Glossary · Technique

Pre-Mortem Analysis

Also known as: Future-perfect failure analysis

Imagine the project failed 6 months from now. Work backwards from the failure to find the cause. Used at Amazon and NASA.

Try the interactive template

When to use it

Before any high-investment project launch.
When the team is overly optimistic and dismissing risks.
Investment decisions, hiring choices, strategic pivots.
Any plan where a 6–12 month outcome is the unit of analysis.

When not to use it

Reversible, low-cost actions where the analysis cost exceeds the project cost.
Daily operations — pre-mortems are for big bets, not standing meetings.

How it works

1Frame the model as writing the post-mortem from N months in the future, with the project already failed.
2Define failure clearly up front (e.g. "<10 paying customers", "team disbanded", "shipped 3 months late").
3Write a month-by-month timeline of disaster — what went wrong, when.
4Identify the earliest warning sign and the decision that locked failure in.
5Identify who, in retrospect, was raising the right concern that got dismissed.
6Snap back to today and list 3 concrete actions to take this week to make the failure less likely.

Example

Lazy prompt

Will my product launch succeed?

Using the technique

It's 6 months post-launch and you have <1,000 active users. Write the post-mortem: month-by-month timeline, earliest warning sign, decision that locked failure in. Then snap back to today and list 3 actions for this week.

Common pitfalls

Without naming failure concretely, the analysis is generic.
If month-by-month is skipped, you miss gradual-decay failures.
If the exercise ends in the imagined future, it's theater. The 'this week' actions are what convert it into an intervention.

Where this came from

Gary Klein, 2007. Adopted at Amazon (writing memos for failed launches before kickoff) and NASA.

Related techniques

Adversarial / Red-Team Prompting

Ask the model to attack your idea before defending it. Surfaces the failure modes before they ship.

Try it interactively

The interactive template lets you fill in your scenario and generates a copy-ready prompt that uses this technique.

Open the template