In most high-stakes strategic sessions, the risks that matter most are already known. The people in the room have privately assessed the plan, identified the points where it is fragile, and formed views about what is likely to go wrong. Most of those views never surface. Not because the participants are withholding — but because speaking them out loud requires breaking social norms that a standard meeting format never gives anyone permission to break.
The pre-mortem is the simplest structural intervention that changes this. It does not require different participants, a longer session, or a different analytical framework. It requires one design choice: before the group commits to a strategy, it imagines that twelve months have passed and the initiative has failed. Then it generates the reasons why.
Why it works: the mechanism of prospective hindsight
The pre-mortem was developed and studied by research psychologist Gary Klein, and its effectiveness is grounded in a well-documented cognitive phenomenon called prospective hindsight. When people are asked to assess a plan abstractly — to consider what might go wrong — they produce thin, generic risk lists. When they are anchored to an imagined failure as a concrete reality and asked to explain it, the mind generates far richer causal explanations. The failure feels real, which unlocks analytical resources that abstract risk assessment does not reach.
The secondary mechanism is social. In most group deliberations, the risks that are most sensitive — the things everyone suspects but no one has named — carry the highest social cost to surface. Raising them requires implying that the plan has a serious flaw, which can read as disloyalty, pessimism, or a challenge to the authority of whoever proposed the direction. The pre-mortem removes this cost entirely. It is not dissent. It is a process requirement. Every participant is doing the same exercise. The risk that would have been held privately is now a contribution.
What the pre-mortem actually surfaces
The risks generated in a well-run pre-mortem tend to cluster into categories that standard risk assessment misses. Resource risks: the assumption that the programme will be resourced at the level required, when organizational history suggests it will not be. Coordination risks: the commitment from another function or external partner that everyone is counting on and no one has explicitly secured. Assumption risks: the external conditions — market, policy, clinical — that the strategy requires to be true and that no one in the room has actually tested.
And then there is the category the second round is specifically designed to surface: the political risks. The implementation actor whose private objection has never been raised in a room where their sponsor is present. The cross-functional dependency that two teams have been silently negotiating around for months. These are not risks that any formal risk register has ever captured. They are the risks that produce execution failure.
The protocol
The most effective pre-mortem design begins with a written individual exercise before any group discussion. Each participant writes their list of failure causes independently, without seeing anyone else's. This prevents anchoring — the tendency for all subsequent contributions to cluster around whatever the first voice in the room articulates. When the lists are shared, the distribution of concerns becomes visible in a way that group discussion never produces.
A second round, explicitly framed around political and social risks, extracts the category of concern that participants are most likely to self-censor in the first pass. Naming the round explicitly — "now we are looking for the things everyone suspects and no one has said" — gives participants the permission structure they need.
The session closes by identifying the top three failure causes from the exercise and building explicit mitigation commitments into the plan before anyone leaves the room. Not action items for follow-up. Commitments with named owners that become part of the strategy itself. This is the difference between a pre-mortem that informs a plan and one that changes it.
The one hour that changes the rest of the engagement
Leaders who have run pre-mortems consistently report the same experience: the most valuable information in the entire strategic process surfaced in the first hour of a structured failure exercise that cost nothing to design and required no additional analytical work. The knowledge was already in the room. The pre-mortem is simply the mechanism that gets it out before the plan is committed rather than after it has failed.