How to Run a Pre-Mortem Before a Big Decision
A post-mortem examines a failure after it happens. A pre-mortem, a technique developed by the psychologist Gary Klein, does the same investigation before you commit — while there is still time to act on what you learn. The move is deceptively simple: you assume the decision has already gone badly wrong, and you work backward to explain why. That small shift in framing pulls out risks that a hopeful forward-looking discussion reliably misses.
Why imagining failure works better
The engine behind the pre-mortem is a finding called prospective hindsight. Research by Deborah Mitchell, Jay Russo, and Nancy Pennington showed that imagining an event has already occurred makes people markedly better at generating concrete reasons for it — in their work, by around 30% — compared with speculating about what merely might happen. “Why did this fail?” is a far more productive question than “What could go wrong?” because the certainty of failure frees the mind to search for specific, plausible causes rather than vague hypotheticals.
It also fixes a social problem. Once a team has rallied behind a plan, voicing doubts feels disloyal, and quiet skeptics stay quiet — the classic recipe for groupthink. A pre-mortem makes doubt the assignment. When everyone is instructed to explain a disaster, the person who spotted the fatal flaw finally has permission, even encouragement, to say it aloud. And by forcing the group to confront how the plan could collapse, it punctures the overconfidence that makes ambitious decisions feel safer than they are.
Running one, step by step
- Gather the people and present the plan. Bring together everyone close to the decision and lay out the chosen course as it currently stands, so everyone shares the same picture.
- Set the scene of failure. The facilitator says: “Imagine it is a year from now. We went ahead with this decision, and it has been a complete disaster. Take a few minutes and write down every reason you can think of for why it failed.”
- Write independently and in silence. Each person lists reasons on their own first, before anyone speaks. Independence is essential — it keeps the loudest or most senior voice from anchoring the room and preserves the diversity of concerns.
- Collect every reason. Go around the group and gather all the failure causes onto one shared list, without debating or dismissing any of them yet. The aim at this stage is coverage, not judgement.
- Strengthen the plan. Now review the list. Sort the risks by how likely and how damaging each would be, and for the serious ones, decide on a concrete mitigation, an early warning signal to watch, or — if the risks are grave enough — a change to the decision itself.
A worked example
Imagine a small team about to launch a new subscription product in six weeks. Everyone is excited, the demo works, and the meeting is drifting toward a rubber-stamp. Instead, the lead calls a pre-mortem: “It is launch day plus one year, and this product flopped. Why?” Working silently, the team writes, then pools a list: onboarding was too confusing and users churned in week one; the pricing sat awkwardly between competitors and appealed to no one; a key integration was not ready, so early reviews were harsh; support was swamped and response times cratered; and the launch collided with a rival’s bigger release and got buried.
None of these surfaced in the earlier optimistic planning. Acting on them, the team simplifies onboarding to three screens, runs a quick pricing test against two competitors, pulls the fragile integration out of the launch scope, staffs up support for the first month, and checks the competitor’s rumored timeline before locking a date. The decision to launch stands — but it is now a far sturdier version of that decision, hardened against its five most likely ways of dying.
When to use it
A pre-mortem earns its keep on decisions that are consequential and hard to reverse — the one-way doors where a hidden flaw is expensive to discover late. It takes twenty to thirty minutes and needs no special tools, only the discipline to imagine failure honestly. Make it a standing step before you commit to anything big, and you will catch problems while they are still cheap to fix rather than after they have already cost you.