Most plans die from a risk someone quietly saw coming but never said out loud. A pre-mortem fixes that. It is a short exercise where the team assumes the plan has already failed, then explains why. This article shows you how to run one, when it helps, and how to turn the output into real changes before you commit.
What a pre-mortem is
A pre-mortem inverts the usual review. Instead of asking “what could go wrong,” you say: it is a year from now and this plan failed badly. Now write down the reasons. The framing sounds small, but it changes who speaks and what they admit.
The technique was described by psychologist Gary Klein and popularized through Harvard Business Review. Daniel Kahneman has praised it as a cheap counter to overconfidence. The core insight is real and well documented in decision research: prospective hindsight, imagining an outcome as already true, makes people generate more specific and more honest reasons.
Why it works when normal risk reviews don’t
Standard risk reviews fail for social reasons, not analytical ones. Once a team commits to a plan, doubt feels disloyal. Junior people stay quiet. The optimist looks like a leader and the skeptic looks like a blocker. A pre-mortem removes that penalty. Failure is now the stated premise, so naming a weakness is playing along, not attacking the plan or its owner.
It also beats vague worry. “I’m a bit nervous about timing” gets ignored. “We missed the deadline because the data migration took three times longer than promised” is concrete enough to act on.
How to run one in 45 minutes
Step 1: Set the scene
Gather the people who will execute the plan, not just approve it. State the premise clearly: the plan launched, and it failed. Give it a specific horizon, such as twelve months.
Step 2: Write silently
Everyone spends five to seven minutes writing failure reasons alone. Silent writing prevents the loudest voice from anchoring the room.
Step 3: Round-robin
Each person reads one reason at a time until the list is exhausted. No debate yet, just capture.
Step 4: Cluster and rank
Group similar reasons. Rank by two questions: how likely, and how damaging. Focus on the top few.
Step 5: Assign owners
For each top risk, decide one action: mitigate it, add an early warning signal, or accept it consciously. Give every action a name and a date.
A real scenario
A company plans to launch a paid tier. The room is excited. In the pre-mortem, one engineer writes: “We failed because billing edge cases created refunds we couldn’t process, and support drowned.” A marketer writes: “Our free users felt betrayed and churned.” Neither would have said this in a normal review, where enthusiasm was the price of admission. The team adds a billing test window and a grandfathering policy for existing users. The launch still happens, but two likely failures are defused first.
When to use it, and when not to
Use a pre-mortem before any decision that is expensive to reverse: a launch, a large hire, a market entry, a reorganization. It is most valuable when the team feels confident, because that is exactly when blind spots hide. Skip it for small, reversible choices, where the exercise costs more than the risk. And do not run it so late that the answer cannot change the plan. A pre-mortem after commitment is theater.
Common mistakes and how to fix them
- Turning it into a debate. Fix: forbid rebuttals during collection. Judge risks only after all are on the table.
- Inviting only leaders. Fix: include the people doing the work, who see the operational cracks.
- Producing a list with no owners. Fix: every top risk gets one action and one name, or the session was wasted.
- Running it too late. Fix: schedule it while the plan can still change, not after budget is locked.
- Letting it become doom. Fix: cap it at the top risks. The goal is a stronger plan, not a longer worry list.
Action checklist
- Invite executors, not just approvers.
- State the premise: the plan already failed.
- Write reasons silently before any discussion.
- Collect every reason with no debate.
- Rank by likelihood and damage.
- Assign one action and owner per top risk.
- Revisit those risks at the next milestone.
Conclusion
Your next step: before your next big commitment, block 45 minutes and ask the team to explain a failure that hasn’t happened yet. It is the cheapest insurance a strategy can buy.
FAQ
How is a pre-mortem different from a risk assessment?
A risk assessment asks what might go wrong in the abstract. A pre-mortem assumes failure has already happened, which pulls out more specific and more honest reasons.
Won’t imagining failure kill the team’s momentum?
Handled well, it does the opposite. Naming risks early builds confidence, because the team commits knowing what it is guarding against.
Who should run the session?
Someone neutral enough to stay quiet. The facilitator’s job is to protect the silent-writing rule and stop early debate, not to defend the plan.
How often should we do this?
Before every hard-to-reverse decision. For ongoing projects, repeat at major milestones, since new risks appear as reality shifts.
References
- Gary Klein, “Performing a Project Premortem,” Harvard Business Review.
- Daniel Kahneman, Thinking, Fast and Slow.