# Pre-Mortem: Imagine Failure Before It Happens

> A pre mortem asks your team to imagine the project has already failed and explain why. Learn Gary Klein's method, step by step, with examples and pitfalls.

Canonical: https://mentarium.io/mental-models/pre-mortem.md

Mental model

6 min read By [Nemo](https://mentarium.io/about) Updated October 30, 2026

> A **pre-mortem** is a planning exercise in which a team imagines that a project has already failed, then each person writes down the reasons it failed. Developed by psychologist Gary Klein, it surfaces risks people would otherwise keep to themselves and tempers overconfidence before a plan is locked in.

- **Category**: Mental model (risk and planning)
- **Origin**: Gary Klein, Harvard Business Review, 2007
- **Best used for**: Projects, launches and big decisions before commitment
- **Related**: Inversion, second-order thinking, margin of safety

## Ask yourself

- It is a year from now and this failed. What is the most likely reason?
- What risk have I noticed but not said out loud?
- Which assumption, if wrong, would sink the whole plan?
- What would an outsider see that we are too invested to see?

## What is a pre-mortem?

A pre-mortem is a meeting (or a solo exercise) where you assume your plan has already failed and work backward to explain why. It takes the post-mortem, which is normally done after a project dies, and moves it to the start, when the plan can still change.

The trick is in the framing. Asking “does anyone see any risks?” in a planning meeting rarely produces honest answers, because the room is invested in the plan and nobody wants to look negative. Saying “it is a year from now, this project failed badly; write down why” flips the social pressure. Naming problems becomes the task, not a betrayal.

A pre-mortem is useful before:

- Launching a product or a campaign.
- Signing a large contract, hire or partnership.
- Committing to a personal decision that is hard to undo.

## Where it comes from

The pre-mortem was developed by Gary Klein, a research psychologist known for studying how experienced people such as firefighters and military commanders make decisions under pressure. He described the method in a short 2007 _Harvard Business Review_ article, “Performing a Project Premortem.”

Klein’s article drew on earlier research on what was called “prospective hindsight”: a 1989 study by Deborah Mitchell, J. Edward Russo and Nancy Pennington suggested that imagining an event has already happened helps people come up with more specific reasons for it than simply asking whether it might happen. The pre-mortem turns that finding into a team routine.

The technique got a wider audience when Daniel Kahneman discussed it favorably in _Thinking, Fast and Slow_ (2011) as a practical way to reduce overconfidence in organizations. It is a close cousin of [inversion](https://mentarium.io/mental-models/inversion): instead of asking how to succeed, you ask how you would fail, then avoid that.

## How to run a pre-mortem

1. **Brief the plan.** Make sure everyone understands the decision or project as it currently stands.
2. **Set the scene.** Say it plainly: “Imagine it is [a date in the future]. We went ahead with this plan, and it was a clear failure.”
3. **Write silently.** Give everyone a few minutes to list every reason they can think of for the failure, on their own. Silent writing keeps the most senior or loudest voice from steering the list.
4. **Share one reason at a time.** Go around the room, each person giving one new reason per turn, until the lists are exhausted. Record everything without debate.
5. **Rank the risks.** Group similar reasons, then pick the few that are both likely and damaging.
6. **Change the plan.** For each top risk, decide what to do: add a safeguard, a check-in date, a [margin of safety](https://mentarium.io/mental-models/margin-of-safety), a fallback option, or a change to the plan itself.
7. **Assign owners and revisit.** Someone watches each top risk. Put a review date on the calendar.

For a solo pre-mortem, do steps 2, 3, 5 and 6 on paper in about fifteen minutes.

 Mentarium Decision Coach · free

Use this with your AI

Connect the free Decision Coach to Claude or ChatGPT and it walks you through the Decision Loop, picks the right Model Stack and runs the Bias Radar on your reasoning.

 [Connect it free](https://mentarium.io/coach)

## 3 real-world examples

### Business

Say a company plans to launch a subscription product in three months. In the pre-mortem, the engineering lead writes that the billing integration slipped and the launch was delayed. Marketing writes that the price was set without testing and conversion was poor. Support writes that nobody planned for the volume of setup questions. None of these risks were secret, but none had been said in the planning meetings. The team adds a billing milestone, a small pricing test and a support plan before committing to the date.

### Personal life

Say you are about to start a big home renovation. You imagine it is six months later and the project went badly. Your reasons: the budget had no buffer, the contractor’s schedule slipped and you were living in a half-finished kitchen, and you made design choices in a rush. You respond by adding a contingency fund, agreeing on a schedule with penalties, and finalizing the design before work starts.

### Tech

In a hypothetical case, a team is migrating a database. In the pre-mortem, an engineer imagines the failure: the migration corrupted a subset of records and nobody noticed for a week. That leads to a concrete change: a verification script that compares record counts and checksums before and after, plus a tested rollback plan. Thinking about what happens next, a habit from [second-order thinking](https://mentarium.io/mental-models/second-order-thinking), also flags that a week of bad data would propagate into reports, so the team plans to pause downstream jobs during the switch.

## Common mistakes

- **Running it after the decision is locked.** If nothing can change, a pre-mortem is theater.
- **Letting it become a debate.** The sharing round is for listing reasons, not defending the plan. Discussion comes after.
- **Skipping the silent writing.** Group brainstorming lets the first few voices set the agenda.
- **Stopping at the list.** The value is in the changes to the plan. A list of risks nobody acts on is just anxiety on paper.
- **Only listing external causes.** “The market changed” is easy to say. “We misjudged what customers wanted” is more useful.
- **Using it to kill every idea.** The goal is a stronger plan, not no plan. Every project has failure modes.

## Pre-mortem vs post-mortem

Both ask why a project failed. The difference is timing, and timing changes everything about what you can do with the answer.

|  | Pre-mortem | Post-mortem |
|---|---|---|
| When | Before commitment or launch | After the project ends |
| Failure is | Imagined | Real |
| What it can change | The current plan | Future projects |
| Main bias it fights | [Overconfidence bias](https://mentarium.io/biases/overconfidence-bias), planning fallacy | Hindsight bias, blame |
| Typical output | Ranked risks and plan changes | Lessons learned and process changes |
| Emotional tone | Low stakes, a bit playful | Often tense, sometimes defensive |

Use both. The pre-mortem protects this project; the post-mortem (ideally blameless) improves the next one.

## The Mentarium take

The pre-mortem is the best return on time of any decision tool we know. An hour of imagined failure regularly saves weeks of real failure, and it costs nothing but a willingness to look at your own plan with suspicion. It is also one of the few techniques that fixes a social problem, not just a thinking problem: it gives people permission to say what they were already worried about.

Not every decision deserves one. That is what the Decision Weight Score (DWS) is for. Rate Reversibility, Impact and Uncertainty, and if the score lands at 7 to 10, the Mentarium Decision Loop calls for the full process: a pre-mortem, a written journal entry and a night’s sleep before deciding. Below that, a quick solo version is usually enough. You can get the score in a minute with the calculator on our [tools page](https://mentarium.io/tools).

Inside the Decision Loop (Frame → Map → Stress-test → Decide → Review), the pre-mortem lives in Stress-test, alongside the Bias Radar. For a business decision, the Bias Radar’s list (confirmation bias, planning fallacy, overconfidence bias, survivorship bias, sunk cost fallacy, anchoring bias, bandwagon effect) is a good prompt for the silent-writing step: for each bias, ask how it might have caused the failure you are imagining.

For launches specifically, our Model Stack is Pre-mortem + Pareto principle + Second-order thinking. The pre-mortem finds what could sink you, the [Pareto principle](https://mentarium.io/mental-models/pareto-principle) keeps the launch focused on the few things that matter, and second-order thinking checks the consequences of your fixes.

If you want a partner to run a pre-mortem with you, the free _Mentarium Decision Coach_ does it inside Claude or ChatGPT as part of the Decision Loop. See [the coach](https://mentarium.io/coach) to set it up, or the [decision-making process](https://mentarium.io/decision-making/decision-making-process) guide for the full picture.

**Counters** [Planning Fallacy](https://mentarium.io/biases/planning-fallacy)

## FAQ

### Who invented the pre-mortem?

The research psychologist Gary Klein developed the pre-mortem and described it in a short 2007 Harvard Business Review article titled 'Performing a Project Premortem'. Daniel Kahneman later praised the technique in Thinking, Fast and Slow.

### What is the difference between a pre-mortem and a post-mortem?

A post-mortem analyzes why a project failed after it is over, when it is too late to change it. A pre-mortem imagines the failure in advance, while the plan can still be changed.

### How long does a pre-mortem take?

A basic pre-mortem fits in 30 to 60 minutes: a few minutes of silent writing, a round of sharing reasons, and time to turn the biggest risks into changes in the plan. You can also run a solo version in about 15 minutes.

### Can I do a pre-mortem alone?

Yes. Imagine your decision has failed, write every reason you can think of for ten minutes, then pick the two or three most likely and decide what you would change. It is less powerful than a group session but still catches blind spots.

### Why does imagining failure work better than asking what could go wrong?

Asking 'what could go wrong?' invites polite hedging. Stating that the project has failed makes it safe, even expected, to name problems, and research on 'prospective hindsight' suggests that imagining an event as already happened helps people generate more specific explanations for it.

## Related models & biases

[Mental model Inversion Inversion is a mental model that approaches a problem from the opposite end: instead of asking how to succeed, you ask what would guarantee failure, then avoid those things. It also means testing a belief by asking what would have to be true for the opposite to be right. Charlie Munger made it famous in business.](https://mentarium.io/mental-models/inversion) [Mental model Second-Order Thinking Second-order thinking is the habit of looking past the immediate result of a decision to the effects that follow from it, and the effects that follow from those. Where first-order thinking asks "what happens next?", second-order thinking keeps asking "and then what?" until the picture includes how people and systems react.](https://mentarium.io/mental-models/second-order-thinking) [Mental model Margin of Safety A margin of safety is the buffer between what you expect to need and what you actually prepare for, so that a plan still works if your estimates are wrong. Investors buy well below estimated value; engineers build stronger than the expected load; anyone can apply it to time, money and health.](https://mentarium.io/mental-models/margin-of-safety) [Decision Making The Decision-Making Process The decision making process is the sequence of steps a person or group follows to make a choice: defining the decision, finding and comparing options, checking the favorite for errors, committing, and reviewing the result. A good process scales its effort to the stakes and ends with a review, so each decision improves the next one.](https://mentarium.io/decision-making/decision-making-process) [Cognitive bias Overconfidence Bias Overconfidence bias is the tendency to be more sure of your judgments, abilities or predictions than the evidence justifies. Researchers split it into three forms: overestimating your own performance, overplacing yourself relative to others, and overprecision, being too certain your beliefs are accurate. It is one of the most studied and most costly biases.](https://mentarium.io/biases/overconfidence-bias)

 The Mentarium Field Guide

60 mental models, organized by the decision in front of you

Each model on one or two pages: when to use it, the questions to ask, an example and the trap. Plus the 25 biases that break decisions and 12 ready Model Stacks.

 [See Field Guide · $14](https://mentarium.io/shop/field-guide)

 The Mentarium Field Guide

60 mental models, organized by the decision in front of you

 [See Field Guide · $14](https://mentarium.io/shop/field-guide)