Prophylactic autopsy

The premortem is a decision-making procedure devised by Gary Klein to challenge overconfident optimism before an organization commits itself to an important plan. Its defining move is to treat the proposed future as if it had already failed, then ask informed participants to explain the disaster. By creating a legitimate space for doubt while a decision can still be changed, the method turns imagined hindsight into a tool of foresight.

Definition and Procedure

The procedure begins when an organization has nearly reached an important decision but has not formally committed itself. A knowledgeable group is assembled and given a deliberately unsettling premise: imagine that a year has passed, the plan was implemented exactly as it exists, and the outcome was a disaster. Each participant then spends five to ten minutes writing a brief history of that disaster. The exercise does not ask for a general expression of unease or a prediction that the plan might fail; it asks people to construct concrete explanations for failure from the standpoint of an imagined future.

The method is a partial remedy rather than a cure for overconfidence. Subjective confidence is often produced by the coherence of the story a person has built, not by the quality or quantity of the evidence supporting it. Because the premortem changes the story that the group is invited to tell—from a success narrative to a disaster narrative—it can expose weaknesses that ordinary confidence assessments leave hidden. It does not eliminate the automatic tendencies associated with System 1, but it gives an organization a structured occasion to bring System 2 effort and informed criticism to bear.

What the Method Counters

The premortem is designed especially to resist groupthink. As a team converges on a decision, and particularly once a leader has revealed her preference, public doubts tend to disappear. Reservations may gradually be treated not as useful information but as signs of disloyalty to the team and its leaders. The result is a group in which supporters of the decision have a voice while objections are suppressed, producing confidence that reflects social alignment as much as independent judgment.

By asking everyone to write about a disaster, the premortem temporarily reverses that social pressure. The exercise grants dissent an authorized form: participants are not required to accuse colleagues of making a bad decision, because the decision is being examined through a stipulated future failure. This makes it easier to mention obstacles, overlooked dependencies, and unfavorable possibilities before they become real consequences. The procedure therefore reduces, though does not completely remove, the damage caused by uncritical optimism and by What You See is All There is, the tendency to reason from the information currently in view while neglecting what is absent.

Relation to Forecasting and Organizational Control

The premortem belongs to a broader approach that treats decision making as a process requiring quality control. Organizations can impose orderly procedures more reliably than individuals can monitor their own judgment in the pressure of a major choice. Useful practices include collecting independent judgments before discussion, separating sources of information so that errors do not become correlated, and using reference-class forecasting to compare a proposed project with the outcomes of similar ventures.

This organizational perspective also addresses the planning fallacy, the tendency to make forecasts that lie unrealistically close to best-case scenarios. The outside view and reference-class forecasting correct that tendency by using distributional information from comparable projects, including likely cost overruns, delays, and underperformance. The premortem performs a complementary task. Reference classes supply evidence about what has happened across similar cases; the premortem asks knowledgeable insiders to imagine how this particular plan could fail. One method disciplines optimism with historical distributions, while the other mobilizes local knowledge and imagination against complacency.

The larger lesson is that a decision should be inspected at several stages: how the problem is framed, how relevant information is collected, how judgments are formed and discussed, and how the final choice is reviewed. In this model, the premortem is not a dramatic emergency ritual reserved for disasters already underway. It is a routine inspection before commitment, comparable to checking the quality of a product during design and fabrication rather than waiting for a failure to reveal defects.

Reception and Narrative Significance

The premortem’s immediate appeal lies in the relief it gives to suppressed criticism. When Gary Klein’s idea was described casually at a session in Davos, an attendee remarked that it had been worth coming to Davos just for that idea; the speaker was later recognized as the chief executive of a major international corporation. The response suggests why the method travels readily across organizational settings: it addresses a familiar discomfort, the sense that a group has begun defending its decision before it has adequately tested it.

The method also illustrates a central principle of behavioral decision research: good intentions and intelligence do not protect people from systematic error. Individuals may be informed, experienced, and sincere while still mistaking a coherent story for a reliable forecast. Institutions can do better when they build independence, dissent, checklists, reference classes, and structured reflection into ordinary practice. The premortem’s significance therefore extends beyond one meeting technique. It makes uncertainty speakable at the precise moment when organizational loyalty and optimism would otherwise make it quiet.