IX Coach

The Pre-Mortem: Imagine It Already Failed

Klein’s prospective-hindsight technique for surfacing hidden risks

What is a pre-mortem, and how does imagining failure help?

A pre-mortem, developed by psychologist Gary Klein, has a team imagine that a plan has already failed and then work backward to explain why — before any work begins. The technique exploits "prospective hindsight": research found that imagining an outcome as a certainty makes people generate more, and more specific, reasons for it, surfacing risks that ordinary risk assessment misses.

A normal plan review asks "what could go wrong?" — a question groups answer cautiously, because naming risks can feel like disloyalty to the plan. Klein’s pre-mortem changes the frame: assume the plan has already failed, then explain the failure. That shift turns risk-finding from pessimism into a storytelling task the mind is good at, and it draws on real research showing imagined certainty improves the quality of reasons people generate. Each practice below carries its mechanism and an honest read on the evidence.

Practices

How these practices connect

Every one of these practices works by forcing a mental operation the brain does not do on its own: treating an uncertain future outcome as an already-settled fact so it can be explained rather than merely forecast. That single shift from predicting to explaining is what unlocks more, and more specific, reasons for failure, and the remaining practices exist to protect that shift from being wasted, misdirected, or emotionally unsustainable.

What is a pre-mortem and why does it work better than just asking what could go wrong?

A pre-mortem has everyone assume the plan already failed and write down why, rather than asking what might go wrong. Stating the failure as a certainty switches the brain from forecasting an uncertain future to explaining a known outcome, and explaining is easier and produces more specific reasons than predicting does. This is called prospective hindsight, and it has been shown to increase both the number and the specificity of risks people generate. If you soften the framing back to a normal 'what could go wrong' question, you lose the certainty framing that makes the technique work and you are effectively back to an ordinary risk review.

Should the team write down failure reasons together or separately?

Separately, and first. If the group discusses out loud before anyone writes anything down, the first or most senior voice tends to anchor everyone else, and quieter disagreement gets suppressed, which shrinks the range of risks that ever gets named. Having each person write their reasons alone before any discussion preserves the diversity of perspectives in the room, which is what makes the pooled list actually comprehensive. Skipping this step is the single most common way a pre-mortem quietly collapses into two or three risks the loudest person already had in mind.

We have a long list of ways the project could fail. Now what?

Rate each listed failure reason on likelihood and impact, then work down the ranked list instead of trying to address everything at once. A long unranked failure list is overwhelming, and without prioritization teams tend to either freeze or cherry-pick whichever risk feels most vivid rather than most consequential. Scoring converts the raw list into a queue so mitigation effort goes to the risks that would actually move the outcome. Treat the ratings as a sorting aid rather than a precise measurement, since likelihood and impact scores are estimates and can be miscalibrated.

I identified the risk. Why didn't that stop it from happening anyway?

Because naming a risk in advance rarely changes behavior in the moment the risk actually starts to unfold. The slide into failure is usually gradual, and each small step is easy to rationalize away as not yet a problem. A tripwire closes that gap by defining a specific, measurable threshold in advance that triggers a pre-decided response, so you do not have to notice and judge the situation while it is already happening and harder to see clearly. This applies implementation-intention research, which has solid experimental support, though using it specifically as a pre-mortem tripwire is an applied extension of that evidence rather than something tested as its own package.

Can I run a pre-mortem by myself on a personal goal, or does it need a team?

Yes, a pre-mortem works solo. The same imagine-it-already-failed technique applies to a personal habit, project, or decision by asking 'a year from now this goal failed, why?' Personal plans suffer the same planning fallacy that team plans do, where people systematically underestimate what will derail them, and prospective hindsight surfaces those obstacles because it is easier to explain a settled failure than to predict one. The most common mistake solo is listing only external obstacles and skipping your own predictable patterns, like avoidance or over-commitment, which are usually the real cause.

Doesn't spending a whole session imagining failure just make everyone discouraged?

It can, if the failure analysis is left standing on its own. Pairing the pre-mortem with an equally concrete picture of what success requires mirrors mental-contrasting research, which finds that holding the desired outcome and the obstacles to it together, rather than dwelling on either alone, is what reliably energizes action. The contrast between the two is the motivating structure, not the failure list by itself. Mental contrasting itself has solid experimental backing, though formally combining it with a pre-mortem is a sensible integration rather than something tested as a single package.

How do you choose between these?

Start with the basic pre-mortem and the independent-writing step together, since one is the mechanism and the other is what keeps the mechanism from collapsing into groupthink the moment more than one person is in the room. If you are doing this alone rather than with a team, the personal-goals version applies the identical certainty-framing lever without needing anyone else present. Once you have a list of failure reasons, whether from a team or solo, rate them by likelihood and impact before doing anything else with them, otherwise the list just sits there as anxiety rather than becoming a plan. For whichever risks land at the top of that ranked list, convert them into tripwires so you have an automatic signal instead of relying on noticing the problem while it is already underway. Pair the whole exercise with a success vision whenever the failure analysis is being run close to when action needs to start, since that is when losing motivation to pessimism is most costly.

What these approaches do not settle

The core prospective-hindsight effect, that framing failure as certain rather than possible surfaces more and better reasons, is a real, observed finding, but controlled studies showing that a full pre-mortem improves the actual end results of a project are limited. The likelihood-and-impact ratings that prioritize the list are a practical convention, not a validated measurement, so treat rankings as directional rather than precise. Several of these practices, including personal pre-mortems, tripwires, and pairing with a success vision, apply well-supported findings from adjacent research (the planning fallacy, implementation intentions, mental contrasting) to the pre-mortem context specifically, which is a reasonable extension but not the same as each combination having been separately tested.

Practice this with IX Coach

Practice this with IX Coach

Frequently asked questions

What is a pre-mortem, and how does imagining failure help?

A pre-mortem, developed by psychologist Gary Klein, has a team imagine that a plan has already failed and then work backward to explain why — before any work begins. The technique exploits "prospective hindsight": research found that imagining an outcome as a certainty makes people generate more, and more specific, reasons for it, surfacing risks that ordinary risk assessment misses.

How do you run the basic pre-mortem?

Stating the failure as a fact, rather than a possibility, switches the brain from forecasting to explaining — and explaining a known outcome is easier and richer than predicting an uncertain one. This "prospective hindsight" was shown to increase both the number and specificity of reasons people generate, so more real risks come to light.

How do you write reasons independently before discussing?

In open discussion, the first or most senior voice anchors the group and quieter dissent gets suppressed, shrinking the range of risks named. Independent writing before sharing preserves the diversity of perspectives, which is what makes the pooled list comprehensive. It defeats anchoring and conformity at the exact moment they would narrow the search.

How do you rate each failure by likelihood and impact?

A long failure list is overwhelming and invites either paralysis or cherry-picking. Scoring each reason on likelihood and impact converts the list into a ranked queue, focusing scarce mitigation effort on the few risks that actually move the outcome. Prioritization is what turns insight into action.

How do you turn top risks into tripwires?

Knowing a risk exists rarely changes behavior in the moment, because the slide into failure is gradual and easy to rationalize. Defining a concrete tripwire — a measurable threshold that triggers a pre-decided response — converts a vague worry into an automatic alarm, bypassing the in-the-moment motivated reasoning that lets risks build unchecked.

How do you pre-mortem your personal goals?

Personal plans suffer the same optimism that team plans do — the planning fallacy makes us underestimate what will derail us. Running a solo pre-mortem ("a year from now this goal failed — why?") applies the same prospective-hindsight lever to your own life, surfacing the personal obstacles your motivation wants to ignore.

How do you pair the pre-mortem with a clear success vision?

A pre-mortem alone can tip into pessimism or analysis paralysis. Pairing it with a vivid success vision mirrors mental-contrasting research: holding both the desired outcome and the obstacles together — rather than either alone — is what reliably energizes effective action. The contrast, not the failure list by itself, is the motivating structure.

What is a pre-mortem and why does it work better than just asking what could go wrong?

A pre-mortem has everyone assume the plan already failed and write down why, rather than asking what might go wrong. Stating the failure as a certainty switches the brain from forecasting an uncertain future to explaining a known outcome, and explaining is easier and produces more specific reasons than predicting does. This is called prospective hindsight, and it has been shown to increase both the number and the specificity of risks people generate. If you soften the framing back to a normal 'what could go wrong' question, you lose the certainty framing that makes the technique work and you are effectively back to an ordinary risk review.

Should the team write down failure reasons together or separately?

Separately, and first. If the group discusses out loud before anyone writes anything down, the first or most senior voice tends to anchor everyone else, and quieter disagreement gets suppressed, which shrinks the range of risks that ever gets named. Having each person write their reasons alone before any discussion preserves the diversity of perspectives in the room, which is what makes the pooled list actually comprehensive. Skipping this step is the single most common way a pre-mortem quietly collapses into two or three risks the loudest person already had in mind.

We have a long list of ways the project could fail. Now what?

Rate each listed failure reason on likelihood and impact, then work down the ranked list instead of trying to address everything at once. A long unranked failure list is overwhelming, and without prioritization teams tend to either freeze or cherry-pick whichever risk feels most vivid rather than most consequential. Scoring converts the raw list into a queue so mitigation effort goes to the risks that would actually move the outcome. Treat the ratings as a sorting aid rather than a precise measurement, since likelihood and impact scores are estimates and can be miscalibrated.

I identified the risk. Why didn't that stop it from happening anyway?

Because naming a risk in advance rarely changes behavior in the moment the risk actually starts to unfold. The slide into failure is usually gradual, and each small step is easy to rationalize away as not yet a problem. A tripwire closes that gap by defining a specific, measurable threshold in advance that triggers a pre-decided response, so you do not have to notice and judge the situation while it is already happening and harder to see clearly. This applies implementation-intention research, which has solid experimental support, though using it specifically as a pre-mortem tripwire is an applied extension of that evidence rather than something tested as its own package.

Can I run a pre-mortem by myself on a personal goal, or does it need a team?

Yes, a pre-mortem works solo. The same imagine-it-already-failed technique applies to a personal habit, project, or decision by asking 'a year from now this goal failed, why?' Personal plans suffer the same planning fallacy that team plans do, where people systematically underestimate what will derail them, and prospective hindsight surfaces those obstacles because it is easier to explain a settled failure than to predict one. The most common mistake solo is listing only external obstacles and skipping your own predictable patterns, like avoidance or over-commitment, which are usually the real cause.

Doesn't spending a whole session imagining failure just make everyone discouraged?

It can, if the failure analysis is left standing on its own. Pairing the pre-mortem with an equally concrete picture of what success requires mirrors mental-contrasting research, which finds that holding the desired outcome and the obstacles to it together, rather than dwelling on either alone, is what reliably energizes action. The contrast between the two is the motivating structure, not the failure list by itself. Mental contrasting itself has solid experimental backing, though formally combining it with a pre-mortem is a sensible integration rather than something tested as a single package.

Research

  • Mitchell, Russo & Pennington (1989), prospective hindsight increased reasons generated for an outcome, J. Behavioral Decision Making
  • Klein (2007), "Performing a Project Premortem", Harvard Business Review About this work
  • Group ideation research on independent vs interacting groups (e.g., nominal-group / brainstorming literature)
  • Gollwitzer & Sheeran (2006), meta-analysis of implementation intentions (if-then plans), Advances in Experimental Social Psychology About this work
  • Buehler, Griffin & Ross (1994), planning fallacy in personal task estimation, J. Personality & Social Psychology
  • Oettingen, mental contrasting / WOOP research on contrasting desired outcomes with obstacles

Articles that cover this