A pre-mortem is a structured technique in which a team assumes, before executing a decision, that it has already failed catastrophically. Members then write independently for five minutes explaining why the failure happened. The exercise surfaces hidden risks, bypasses groupthink, and gives every participant explicit permission to voice pessimism. It was developed by psychologist Gary Klein and is used by organisations including NASA, Google, and the US Army.
Where this came from
Gary Klein, a research psychologist who spent decades studying how experts make decisions under pressure, developed the pre-mortem technique and first described it formally in a 1989 paper. His research on naturalistic decision-making, which examined firefighters, military commanders, and intensive care nurses, showed that experienced practitioners rarely used formal decision trees. Instead, they recognised patterns and simulated outcomes mentally before acting.
Klein observed that one of the most consistent failure modes in group decisions was what he called "premature closure": a team converges on a plan and then systematically discounts evidence that contradicts it. Standard risk analysis did not solve this because it asked people to imagine risks while still mentally committed to the plan's success. The pre-mortem changes the starting assumption. Failure is not a possibility to be quantified; it is the declared outcome. The only task is to explain it.
Daniel Kahneman, in his book "Thinking, Fast and Slow," called the pre-mortem "a powerful antidote to overconfidence" and recommended it as a standard practice for any high-stakes planning process. Its adoption across NASA, the US Army, and multiple technology companies reflects the fact that it requires no special training, adds minimal process overhead, and consistently produces failure modes that conventional planning misses.
How it works
The process has four steps. First, the decision or plan is announced to the group. Everyone understands what has been decided and is about to be executed.
Second, the facilitator says: "Imagine it is one year from today. We proceeded with this plan and it has failed completely. Not a partial setback, a full failure. Now write down, independently, every reason you can think of that explains why it failed." Each person writes alone, without discussion, for five to ten minutes. Independence is critical. If people discuss first, the most senior voice anchors the room.
Third, participants share their lists one item at a time, going around the table until all unique failure modes are captured. The facilitator records them without judgment or debate during this phase.
Fourth, the team reviews the compiled list and identifies which failure modes are (a) high-probability, (b) high-impact, and (c) preventable. Items that are all three become action items: plan modifications, additional safeguards, or explicit go/no-go criteria to monitor during execution.
A practical note on facilitation: the technique works best when the most senior decision-maker in the room explicitly endorses the pessimistic framing. If the CEO calls for a pre-mortem and then visibly dismisses critical input, the exercise becomes theatre. When senior leaders model genuine receptivity to bad news, the quality of input rises significantly.
When to use it and when not to
Use the pre-mortem for decisions with high stakes and limited reversibility: product launches, major capital allocations, significant hires, strategic partnerships, technology migrations. It is most valuable when a team is already committed to a direction and enthusiasm is high, because that is when critical thinking is most suppressed.
Do not use it for routine operational decisions, low-stakes choices, or situations where speed is the primary constraint. A pre-mortem for a weekly marketing email creates bureaucracy without proportionate benefit.
One common misuse: treating the pre-mortem as a veto mechanism. The goal is not to find reasons to abandon the decision. It is to identify which risks are real and addressable before they become costly. Teams that run the exercise correctly sometimes change the decision; more often, they change the execution plan. Both are valuable outcomes.
Optimism Bias
Optimism bias is the tendency to overestimate the probability of positive outcomes and underestimate the probability of negative ones. Studies consistently show that most people believe they are less likely than average to experience failure, illness, or financial loss. In group planning, optimism bias compounds: teams selectively weight evidence that supports the plan and discount or ignore signals of risk. The pre-mortem is a direct countermeasure. By making failure the stated starting assumption, it forces the brain to model the downside, which it will not do spontaneously when excitement about a plan is running high.
Run this framework on your actual decision.
DecisionsMatter.ai walks you through a structured 5-step analysis. Your first analysis is free.
How to apply it in practice
Before your next significant decision, block 30 minutes. Write the decision at the top of a page in one sentence: what is being decided, by when, and what the expected outcome is.
Then write the following: "It is [date one year from now]. This decision has failed. Here is why." Set a timer for 10 minutes and write every failure mode you can generate. Do not filter for likelihood. Write the implausible ones too. Volume matters more than precision at this stage.
When the timer stops, review the list. Mark each item with H (high probability), M (medium), or L (low). Then mark each item with a second dimension: P (preventable with a plan change) or N (not preventable). Any item marked H-P is a priority. Build a specific mitigation for it, or articulate an explicit trigger that would cause you to stop and reassess.
If you are working with a team, run the independent writing step first, before any discussion. Collect all lists before anyone speaks. This protects the diversity of the failure modes and prevents anchoring on the first voice in the room.
This is one model from the upcoming Decisions Matter book.
30 mental models. 40 cognitive biases. A 5-step decision system. Get the relevant chapter in your inbox when it publishes.
Frequently asked questions
What is a pre-mortem?
A pre-mortem is a risk-identification technique in which a team or individual assumes, before a decision is executed, that it has already failed completely. Participants then write down every plausible reason the failure occurred. The goal is to surface risks that optimism, social pressure, or groupthink would otherwise suppress. Psychologist Gary Klein developed the technique and first described it in a 1989 paper.
How is a pre-mortem different from a post-mortem?
A post-mortem is conducted after failure has occurred; it reconstructs what went wrong. A pre-mortem is conducted before execution; it imagines failure prospectively. The pre-mortem is more valuable because it can still change the decision or the execution plan. Post-mortems diagnose the past. Pre-mortems shape the future.
When should you run a pre-mortem?
Run a pre-mortem whenever the cost of failure is high and the decision is not yet fully reversible. Product launches, large capital investments, strategic pivots, hiring decisions for senior roles, and major project plans are all good candidates. The technique adds the most value when a team has already decided to proceed and enthusiasm is high, because that is precisely when critical voices are most likely to stay silent without a structured prompt to speak.
How do you run a pre-mortem without a team?
Working alone, set a timer for 10 minutes and write as if you are a hostile analyst explaining why your decision failed. Use a different physical environment from where you normally work, if possible, to shift your mental frame. You can also write from the perspective of your most sceptical critic. The goal is to generate a genuine list of failure modes, not to reassure yourself. After writing, review the list and ask which risks are preventable, which are acceptable, and which change your view of the decision itself.