Run a pre-mortem from the outside-view perspective

Imagine the project has failed — then ask which base-rate failure type caused it.

Why it works

A standard pre-mortem asks "why did this fail?" Gary Klein showed this loosens the group’s commitment to the plan and surfaces risk. Combining it with outside-view thinking sharpens it: rather than imagining novel failure modes, ask which known categories of failure (the ones that appear in the reference class) caused the imagined failure. This imports the distribution into the pre-mortem rather than relying on creative speculation.

How to do it

1. Gather the team before committing to the plan.
2. State: "Imagine it is one year from now and this project failed badly. What did the failure look like?"
3. Then ask: "Which of the common failure types for projects like this — late, over-budget, low adoption — is this?"
4. Document the top two or three failure types and build explicit mitigation measures for each.

Evidence

Klein’s pre-mortem technique is well described in the practitioner literature; its effectiveness in reducing overconfidence is plausible and anecdotally reported, though direct RCT evidence is limited. The outside-view extension is a conceptual combination rather than a separately studied technique. (anecdotal)

Pre-mortems work best with psychological safety — teams under hierarchical pressure tend to produce sanitized failure scenarios that don’t threaten the plan.

Sources

• Klein (2007), "Performing a project premortem," Harvard Business Review

Common mistake

Generating only creative or catastrophic failure scenarios ("what if the market collapses?") rather than the mundane base-rate failures that actually happen most often ("what if this takes three times longer than estimated?").