Premortem

Trevor L Taylor, a Miami private investor whose investing grew out of systematic and quantitative models, frames excellence as the product of method rather than luck, and the premortem is one of the simplest methods for testing a plan before it is carried out. The idea is to declare a project dead before it starts. A team imagines that its plan has already failed, writes down why, and then repairs the weak points while change is still cheap. The technique has spread from project offices to boardrooms because it asks for almost nothing except an hour of honest imagination.

Where the imagined failure exercise came from

The psychologist Gary Klein gave the method its name. In a short Harvard Business Review article¹, he described a session in which team members assume the project they are planning has just failed, as so many do, and then generate plausible reasons for the failure. The medical analogy is deliberate. A postmortem tells doctors why a patient died, which helps the next patient but not this one; a premortem runs the same inquiry while the patient can still be saved.

The method rests on older research into what psychologists call prospective hindsight. In a study published in the Journal of Behavioral Decision Making², Deborah Mitchell, J. Edward Russo and Nancy Pennington defined it as explaining a future event as if it had already happened: one goes forward in time and then looks back. Their experiments found that certainty about an outcome changed the kind of explanations people produced. Explanations for events treated as sure tended to be longer, richer in specific episodes and told in the past tense. A premortem borrows that effect by stating the failure as a fact rather than a possibility.

Klein has written that he developed the method for his own company³ to cut down on problems in client projects, with no grand ambitions for it. He reports that it has since been used in corporate boardrooms, wildland firefighting operations, Army programs and on Wall Street, and that the Nobel laureates Daniel Kahneman and Richard Thaler have encouraged people to try it. The columnist Oliver Burkeman, in a piece reprinted by the Gulf Times⁴, noted that Kahneman described the premortem as his favourite method for making better decisions.

How a premortem session runs

Klein’s own description is brief. The session works best at a kickoff meeting, once the team understands the plan. The leader then switches gears and announces that an infallible crystal ball has shown the plan to be a fiasco. Each member takes two minutes to write down the reasons it failed, and the group goes around the room, each person giving a top reason in turn, until the list is exhausted. Klein says the whole exercise can take as little as 20 to 30 minutes. The final step is the one that matters: the plan is revised to deal with the most serious reasons on the list, and some teams return to the list at later milestones.

The reason this framing works is social as much as cognitive. In a McKinsey Quarterly conversation with Kahneman⁵, Klein called the premortem a sneaky way to get people to do contrarian, devil’s advocate thinking without encountering resistance. Burkeman’s account adds that the question removes pressure from people worried about seeming disloyal and turns the meeting into a contest to find ever more convincing reasons for failure. Someone who has quietly doubted a schedule or a supplier now has permission, even an obligation, to say so. Klein’s sequence also protects the quieter members of a group: writing reasons privately before anyone speaks keeps the first or most senior voice from setting the agenda, and taking one reason per person per turn spreads the floor around the table instead of letting a single critic dominate.

The premortem also targets a specific bias. Reporting for The Atlantic⁶, Ben Yagoda listed premortems alongside checklists as procedures that slow thinking organizations can require, explaining that the exercise counters optimism bias by asking team members to picture a project that went very badly and to describe how it happened.

Evidence, uses and limits

Optimism is the problem the method is built to correct. Dan Lovallo and Kahneman argued in Harvard Business Review⁷ that the evidence on major business initiatives such as mergers, acquisitions and capital investments is disturbingly clear: most fall short of the expectations set for them. Public projects show the same pattern. NPR reported⁸ that NASA’s James Webb Space Telescope reached its launch site only after years of repeated delays and cost overruns, and that at one point the observatory was threatened with cancellation. No single meeting would have prevented that history, but such cases are why advocates want failure considered at the start rather than explained at the end.

Consultants have adopted the tool for the same reason. A McKinsey Quarterly article⁹ written by Klein with Tim Koller and Dan Lovallo opens with a parking app launch that went wrong in ways a later postmortem found obvious, from delayed financing to sensors that failed to talk to central servers, and notes that studies show project leaders tend to be overconfident. Schools have picked it up as well: a research summary from the Chartered College of Teaching¹⁰ recommends premortems for new classroom initiatives, observing that failures often look avoidable in hindsight.

The limits deserve equal weight. The research base is modest; the prospective hindsight study concerned how people explain events, not whether projects that use premortems succeed more often. A premortem can produce a long list of risks without saying which ones are likely, and a team that treats the list as a ritual gains little. It works best paired with other disciplines in this cluster, such as a checklist for known failure points, expected value estimates for weighing the risks it surfaces, and the habit of second order thinking about what a fix might set in motion. For an investor, the natural application is investment due diligence: before committing, assume the position has lost money and write the story of how. A postmortem can only explain a loss after the fact. A premortem, held early enough, can still prevent one.