Premortem
The Premortem, developed by psychologist Gary Klein and popularised by Daniel Kahneman in 'Thinking, Fast and Slow', is a structured technique for identifying failure risks before a project or decision begins — not after. It uses prospective hindsight: participants are asked to imagine that the project has already failed spectacularly and to work backwards to diagnose why. This mental time-travel trick is powerful because it bypasses the optimism bias and groupthink that typically suppress risk discussion during planning. When people are asked 'what could go wrong?' they often give polite, vague answers. When asked 'the project failed — what happened?', they access a completely different cognitive mode and produce specific, actionable failure diagnoses. Klein's research showed that prospective hindsight increases the ability to identify correct reasons for future outcomes by 30%. The method is fast, energising, and requires no special materials. It works equally well for product launches, strategic decisions, go/no-go decisions, and organisational changes. Unlike a risk register, it produces vivid narratives that stick — making it far more likely teams will take preventive action.
Copione di facilitazione
- 1
Present the plan in five minutes or less — scope, timeline, success criteria — and check for shared understanding. Do not defend the plan; the group is about to break it, and that is the point.
5 min - 2
Deliver the time-travel prompt: 'It is 18 months from now. The project has failed — not underperformed, failed completely. Picture that.' Then ask everyone to silently write specific reasons for the failure, one per sticky note, and explicitly welcome unlikely or 'paranoid' causes.
8 min - 3
Collect reasons round-robin, one per person per pass, posting each on the board without discussion or rebuttal. Keep circulating until reasons run dry — the second and third passes often carry the most honest material.
10 min - 4
Cluster similar failure modes together, then have the group vote on the failures they consider most likely and most damaging. Circle the top five to seven.
8 min - 5
For each top risk, ask three questions in order: Can we prevent it? Can we detect it early? Can we limit the damage if it happens? Push for concrete countermeasures — a named change to the plan, not an intention to 'be careful'.
12 min - 6
Close by assigning an owner and a milestone to each countermeasure, and agree where the premortem output will live so it stays visible for the life of the project.
5 min
Suggerimenti
The facilitator's job is to create psychological safety — explicitly normalise 'paranoid' or 'unlikely' failure modes.
Some of the most valuable insights come from people who privately worried but never voiced their concern.
Keep the premortem output visible throughout the project, not just at kickoff.
Run a postmortem against the premortem predictions at the end to sharpen future risk identification skills.
Errori comuni
Softening the prompt to 'what might go wrong?' — the method depends on stating failure as an accomplished fact, which unlocks a different level of specificity than polite risk speculation
Letting the project sponsor rebut failure reasons as they are shared, which shuts down exactly the candour the exercise exists to create
Skipping the silent individual writing and going straight to open discussion — loud voices set the frame and the quieter, more uncomfortable risks never surface
Ending with a ranked list of risks but no owners or plan changes; a premortem without countermeasures is just organised pessimism
Variazioni
Run a 'Pre-success' variant alongside the premortem: imagine the project succeeded beyond all expectations — what made it work? This balanced lens reduces excessive negativity. Combine with Scenario Planning: run a premortem for each of the four scenario quadrants. Run a 'Pre-Parade' alongside it: 'Imagine the project succeeded beyond all expectations. What made it work?' This gives you both the risks and the success conditions.
Casi d'uso
Quando usarlo
A project kickoff where the plan looks solid on paper and nobody has voiced a serious concern yet
A go/no-go or investment decision where you suspect optimism bias or sunk-cost momentum is suppressing honest risk talk
A team with quieter members who privately hold doubts they have not raised in planning meetings
The weeks before a launch, migration, or organisational change, while there is still time to build countermeasures into the plan
Quando non usarlo
The project is already in trouble — run an after-action review or a structured problem-solving session on what is actually happening instead of imagining hypothetical failure
The plan is not yet concrete; a premortem needs a specific plan to attack, otherwise it produces generic worries nobody can act on
The decision is irreversible and no scope, budget, or timeline changes are possible — surfacing risks nobody can act on breeds cynicism
You need a quantified risk register for compliance or governance; a premortem surfaces risks vividly but does not assess probability or cost, so feed its output into formal risk analysis rather than replacing it
Metodi correlati
Approfondimenti
Domande frequenti
How long does a premortem take?▾
Between 30 and 60 minutes end to end. A group of six to ten people typically needs about 45 minutes: a short plan briefing, silent writing, round-robin collection, prioritisation, and countermeasure planning. Larger groups need more time for the round-robin.
How many people should take part in a premortem?▾
It works with 3 to 20 participants; five to ten is comfortable. Include the people who will actually do the work, not just the planners — implementers usually know failure modes the plan's authors cannot see.
When in a project should you run a premortem?▾
After the plan is concrete but before the team is fully committed — typically at kickoff or just before a go/no-go decision. Many teams also rerun a short premortem at major phase gates, and compare the original predictions in the final postmortem.
What is the difference between a premortem and a risk assessment?▾
A risk assessment lists and scores risks analytically; a premortem asks the team to imagine the failure has already happened and diagnose it, which bypasses optimism bias and produces more specific, memorable failure stories. They complement each other: use the premortem to surface risks, then a register to track and quantify them.
Can you run a premortem remotely?▾
Yes, and it works well. Use a shared board with private or hidden-note mode for the silent writing so nobody anchors on others' answers, then reveal all notes at once and run the round-robin verbally. Dot-vote digitally to prioritise the top failure modes.
Pianifica il tuo prossimo workshop con l'IA
Workshop Weaver ti aiuta a combinare metodi come Premortem in un'agenda completa e temporizzata in pochi minuti.
Prova gratisMethod descriptions on Workshop Weaver are original content written by our team, based on established facilitation practices. This method was inspired by work from Gary Klein, 'Performing a Project Premortem', Harvard Business Review (2007).