The Business Premortem: Name the Failure While It Is Still Cheap
A premortem is a planning exercise that assumes your plan has already failed, then asks everyone involved to write down why. Gary Klein described the method in Harvard Business Review in September 2007. Treating the failure as a fact, not a possibility, surfaces causes that a normal risk review leaves unsaid.
In a business, that means running it before the decision is locked: before the hire, the price change, the new offer, the client you are about to bet the quarter on.
What is a premortem?
Klein defines a premortem as “the hypothetical opposite of a postmortem.” A postmortem asks what killed the patient. A premortem declares the patient dead at the start of the project and asks the same question while the answer can still change the plan.
The difference is one word. A typical critique session asks what might go wrong. A premortem asks what did go wrong. Klein grounds the method in 1989 research by Deborah J. Mitchell (Wharton), Jay Russo (Cornell) and Nancy Pennington (University of Colorado), who found that prospective hindsight, imagining that an event has already occurred, increases the ability to correctly identify reasons for future outcomes by 30%.
It also changes who speaks. Klein’s diagnosis is that projects fail partly because people with reservations stay quiet during planning. Once the room has agreed the project is dead, naming the cause is the assignment, not an act of disloyalty.
The other meaning. In medicine and forensics, premortem (also written pre-mortem or pre mortem) is an adjective meaning before death, the same sense as antemortem. If you searched “pre mortem meaning” for a pathology context, that is your answer. The rest of this page is about the business method.
Premortem vs postmortem
| Risk review | Premortem | Postmortem | |
|---|---|---|---|
| When it runs | During planning | At the start, after the team is briefed on the plan | After the project ends or fails |
| The question | What might go wrong? | It failed. What went wrong? | What went wrong? |
| Starting assumption | The plan works | The plan failed | The plan failed |
| What it can still change | The plan, if anyone speaks up | The plan | The next project |
Klein on medical postmortems: everyone benefits except, of course, the patient. A premortem is the version where the patient can still be saved.
How to run a business premortem
Klein’s sequence: brief the team on the plan, declare it failed, have everyone write alone, read the reasons out in turns, then the project manager reviews the list for ways to strengthen the plan. In an owner-led business that last step needs more structure, so it gets its own rule below.
- State the decision in one sentence, with a date. “We move every retainer client to the new pricing in January.” A premortem on a vague intention produces vague causes.
- Declare it dead. Pick a failure date far enough out that failure is plausible. Klein’s leader opens by telling the room the project has failed spectacularly.
- Everyone writes alone. A few minutes, every reason each person can think of, especially the ones they would not normally say out loud. Nobody reads anything yet.
- Read one reason each, in turns, until the lists run dry. Every reason gets recorded. No debate during the round. Klein starts with the project manager; in an owner-led business, consider having the owner read last so the room does not anchor on the boss’s list.
- Cluster, rank, and change the plan. Group the reasons into causes. Rank them by how likely they are and how much damage they would do. For the top three, write the change to the plan, who owns it, and the early signal that says the cause is starting.
Project premortem vs structural premortem
A project premortem tests a plan. A structural premortem tests the business underneath the plan. On this site it is defined as a read of “a business for the load-bearing assumptions that will break it, while they are still cheap to change.”
The difference matters when the cause of failure was in the business before the plan existed. A team premortem on a new offer will find launch risks. It is less likely to find that one client is carrying the year, that nothing ships without the owner, or that the business has to replace a large share of its revenue every year just to stay flat. Those are structural causes, and each one has a free diagnostic here:
- Revenue concentrated in a few clients: the Revenue Concentration Risk Radar
- A business that stops when the owner stops: the Owner Dependency Score
- Revenue that has to be rebuilt every year: the Revenue Fragility Index
One read on the methodology page shows the gap. A skincare DTC brand at $1.8M GMV planned to move 30% of paid budget into brand. The structural finding sat underneath it: CAC had climbed from $24 to $51 while lifetime gross profit per customer was in the mid-thirties. “You’re paying $51 to acquire customers worth $36.” The brand campaign was not the failure point. The unit economics under it already were.
Where team premortems break
Klein’s method solves a real problem: it makes dissent safe. It cannot add a perspective the room does not have. Every reason on the list is one somebody in the room was able to imagine, and a team that built the business together tends to share the frame that built it. The lens that got the business here is the same lens that hides what is now wearing it down.
In a small company the team often reports to the person whose plan is on the table. Klein’s format lowers that pressure without removing it.
The fix is lenses that are built to disagree with each other and with you. Pharallax runs twelve adversarial personas, four archetypes (Operator, Architect, Critic, Customer) crossed with three lenses (Structural, Forensic, Behavioral), across six rounds of dialectic. A finding that survives disagreement across personas is treated as structural. A finding only one persona raises is dropped before it reaches you. A fact-check pass audits the claims in the report against your input and a knowledge graph of 320,000+ claims across 160+ domains. The full mechanics are on the methodology page, and three anonymized reads show what comes out the other end.
Run a premortem on your business
Describe the decision you are about to make on the free analysis. It reads the problem through one lens and returns results instantly, enough to see whether the read lands.
The Structural Read is the full premortem: twelve cognitive lenses argue your situation across six rounds. $97 founding rate, full report in 4 hours, money back if it doesn’t surprise you.