the pre-mortem: assume it already failed
asking "what could go wrong" gets you a shrug. asking "it went wrong, explain how" gets you the real list. the difference is not psychological trickery, it is grammar.
gary klein's pre-mortem is the highest return per minute of any decision tool i know of, and it takes about fifteen.
you run it after you have decided but before you commit. the instruction is specific: imagine it is a year from now and this decision failed badly. do not imagine it might have failed. it failed. now write the story of how.
why the wording matters that much
"what could go wrong here" is a question you can answer with "not much, i think it is fine". it invites reassurance, and if you are the person who made the call you are motivated to give it.
"it failed, explain why" removes the option. the failure is a given, so your mind stops defending the decision and starts doing what it is good at, which is constructing a plausible story from an assumed ending.
klein's research found this framing surfaced substantially more concerns than a standard risk review, from the same people, about the same plan.
running one on your own
most write-ups assume a team. solo works, it just needs discipline about the writing.
- state the decision and the date. "we moved pricing to 4,500 on 1 september."
- set the clock forward. pick a horizon where the outcome would be obvious. usually three to twelve months.
- write the failure as fact. "by march it was clearly a mistake." past tense, no hedging.
- write for ten minutes without stopping. the fourth and fifth reasons are the useful ones. the first three are the ones you already knew.
- underline anything you can act on now. most of the list will be noise. one or two items will be things you can cheaply de-risk this week.
what usually comes out
in practice the pre-mortem tends to surface one of three things.
a dependency you had not noticed. a person whose reaction you assumed and never checked. or a timing assumption that only works if nothing else slips.
none of those are exotic. they are just invisible from inside a decision you have already made, which is exactly the position you are in when you run it.
when not to bother
skip it for anything reversible. a pre-mortem on a two-way door is procedure for its own sake, and the cost of just trying the thing is lower than the cost of the exercise.
use it on the calls where being wrong is expensive and slow to undo. knowing which kind of door you are holding is the prerequisite.
then close it
a pre-mortem is not a reason to reopen the decision. it is a way to commit with your eyes open. write the call down with your confidence score, note the two risks you actually acted on, and set the review date.
if it has been sitting a while already, the decision fatigue calculator will tell you roughly what the delay has cost, on your own figures.