The pre-mortem: find out why your plan failed before it starts
Run this method in IdeaHone
Creates a space with the entities and scoring criteria of this method, ready for your next session.
Every plan gets a review once it is over. The project failed, the team sits together and everyone knows exactly what went wrong: nobody had time, the customers never came, the supplier was late. The pre-mortem moves that conversation to the start, while you can still do something about it. Gary Klein described it in the Harvard Business Review in 2007, and the idea is simple: assume the plan has already failed and explain why.
Why imagining the failure works
Asking "what could go wrong?" gets polite answers. People do not want to look negative about a plan their manager just presented, and the risks they name are the safe ones. Saying "it failed, why?" changes the task. The failure is a fact in the exercise, so naming a reason is no longer disloyal, it is the job. Research on this kind of prospective hindsight, imagining that an event has already happened, found that people come up with more and more specific reasons than when they only ask what might happen.
The second effect is quieter. A pre-mortem gives the sceptic in the room a legitimate place to speak. The person who has doubted the plan for weeks finally writes it down, and the doubt becomes a risk with a name instead of a mood in the corridor.
When to run one
Run a pre-mortem when a decision is made but nothing is spent yet: after the go for a pilot, before the budget is released, before a launch date is announced. Earlier, there is no plan to fail. Later, people defend what they have already built. Use it for decisions that are costly to reverse; a two-week experiment does not need one.
Running a pre-mortem in 45 minutes
- Name the plan in one line. "Refill station in all 12 stores by June." If you cannot write it in one line, the pre-mortem will find that out the hard way.
- Set the scene. "It is a year from now. The plan failed. Not a little, completely." Say it out loud and let it land for a moment.
- Write alone for five minutes. Every person writes the reasons why it failed, one per line. Alone matters: once the first reason is said aloud, the room starts to agree with it.
- Read out and merge. Everyone reads their reasons. Merge the ones that say the same thing, keep the wording of the sharpest one.
- Mark the most likely. Each person gets three dots for the reasons they find most likely. Not the worst, the most likely.
- For each of the top reasons, name the warning sign. What would we see early if this were happening? "The refill queue is longer than five minutes in week one" is a warning sign. "Low adoption" is not, because nobody will notice it until it is too late.
- Decide what to do now. One action per top reason, with an owner. Some reasons need a change to the plan, some only need someone to watch the warning sign.
- Turn the best warning sign into a kill criterion. The sign that would tell you most clearly to stop becomes the kill criterion of the decision, with the date it is checked.
Write the reasons as past events, not as fears: "Staff had no time to refill" instead of "staff might be too busy". The past tense keeps the exercise honest and makes the reasons concrete enough to act on.
Future backwards: the same picture, the other way
The variant asks the opposite question: "It is a year from now and this succeeded beyond expectations. What made it happen?" The reasons become ideas and actions instead of risks. Use it when the team is already anxious, or after a pre-mortem, to balance the list of dangers with a list of things to do more of. Do not run it instead of a pre-mortem on a risky decision: success stories are pleasant and rarely uncomfortable enough.
Where pre-mortems go wrong
- The sponsor writes first. Once the most senior person has said what they think, the rest write variations of it. The sponsor writes alone like everyone else, and reads last.
- A list without owners. Twenty risks on a wall feel thorough and change nothing. Only the top three or four get a warning sign and an owner; the rest stay recorded as risks of the concept.
- Warning signs nobody measures. "Customers are unhappy" will never trigger anything. A warning sign has a number or an observable event, and someone who looks at it.
- Running it after the money is spent. A pre-mortem after the launch is a post-mortem with extra steps. Run it while changing the plan is still cheap.
For a second round of doubt, let AI play the sceptic before the concept reaches the CFO.
Score and prioritise ideas with your team
IdeaHone scores ideas against your criteria, shows them on a matrix and keeps the reasons, so the decision is clear and traceable.
What is a pre-mortem?
A short workshop in which the team assumes a plan has already failed and writes down why. The likeliest reasons get a warning sign and an action, so the risks are dealt with before the plan starts. The method was described by Gary Klein in 2007.
How is a pre-mortem different from a risk assessment?
A risk assessment asks what might go wrong and usually collects the risks people are comfortable naming. A pre-mortem treats the failure as a fact, which makes it easier to name uncomfortable reasons, and every person writes alone before anything is discussed.
How long does a pre-mortem take?
About 45 minutes for one plan: five minutes of setting the scene, five of silent writing, fifteen of reading and merging, and the rest for the warning signs and actions of the top reasons.
What is future backwards?
The opposite variant: the team imagines the plan succeeded beyond expectations and writes down what made it happen. It produces ideas and actions instead of risks.
