The Pre-mortem
The pre-mortem (Gary Klein) inverts the post-mortem: before work begins, the team assumes the plan has already failed, then writes down why. Research on prospective hindsight found that imagining an event as having already happened materially improves failure-cause detection, roughly a thirty percent gain.
Why the ritual works
Two things make it more than a checklist. First, certainty of failure changes the social contract: in normal planning, raising doubts reads as disloyalty to the plan; in a pre-mortem, finding failure causes is the assignment, so the most sceptical person in the room becomes the most valuable. That is psychological safety by design rather than exhortation. Second, it catches the failure modes the planners are blind to precisely because they built the plan: de-biasing built into the cadence rather than left to individual virtue, the same territory as avoiding biases and blindspots.
Where it lives in the system
Inside the OKR creation process, run once per new OKR, after drafting and before commitment. Its output does not evaporate into a workshop feeling; it seeds the Issues and Obstacles Log, giving every named failure cause an owner and a date on day one. It complements the OKR risk assessment: the assessment supplies the categories to sweep, the pre-mortem supplies the imagination that fills them. And the confidence set at launch should be readable against the pre-mortem's output: high confidence with a long, live obstacle list is a contradiction someone should have to explain.
The shelf agrees from different directions
Mental Contrasting and WOOP supplies the individual-psychology evidence that obstacle-naming fuels rather than dampens pursuit, and V2MOM hard-wired an Obstacles field into its goal document. The pre-mortem is the team-scale ritual that does what both point at. The full lineage is mapped at Goal Setting.
Credited to Gary Klein; prospective-hindsight research by Mitchell, Russo and Pennington. Machinery connections are ZOKRI methodology.
How is a pre-mortem different from a risk assessment? +
The risk assessment supplies categories to sweep; the pre-mortem supplies imagination, by assuming failure has already happened. Run both at OKR creation: the assessment for coverage, the pre-mortem for the failure modes planners are blind to.
How do we stop pre-mortem findings evaporating? +
Give every named failure cause an owner and a date on day one, in a live obstacle log that check-ins keep current and escalation drains. That lifecycle is part of every ZOKRI implementation.
A ritual is only as good as the machinery that catches its output. We train teams to run pre-mortems properly and install the log that keeps the findings alive all quarter.