| name | itk-premortem |
| description | Frame and explore your problem by imagining a future scenario in which the proposed activity fails to achieve its objective. |
| intent | Clarify what success looks like. Uncover hidden assumptions about activities, outcomes, success, and failure. Mitigate risks by identifying potential causes of failure. Give your team insight into priorities and success criteria |
| type | component |
| phase | define |
| outcome | define |
| difficulty | intermediate |
| group_size | 6+ people |
| time_required | 60+ minutes |
| best_for | ["Kicking off a major initiative or epic where the team is overconfident and hasn't stress-tested its assumptions","Aligning a cross-functional team on success criteria before locking OKRs or a quarterly roadmap commitment","De-risking a high-stakes launch or platform migration where failure would be costly and politically visible","Surfacing unspoken doubts from stakeholders who privately disagree with a roadmap direction but won't say so openly","Validating a PRD or product strategy when scope is set but risks and dependencies remain unexamined"] |
| sources | ["MITRE Innovation Toolkit (ITK)","https://itk.mitre.org/toolkit-tools/premortem/"] |
PREMORTEM
What Is It
Frame and explore your problem by imagining a future scenario in which the proposed activity fails to achieve its objective.
Why Use It
Clarify what success looks like. Uncover hidden assumptions about activities, outcomes, success, and failure. Mitigate risks by identifying potential causes of failure. Give your team insight into priorities and success criteria
When to Use It
When the definition of success is being determined or assessed.
How to Do It
- Assemble the project team, and give each member a copy of the Premortem Canvas. Invite everyone to imagine it is two years in the future. Your current project, activity, or initiative has utterly failed. Spend a few minutes writing down answers to the questions in the canvas, then discuss answers with the larger group. Describe the failure. Using vivid detail, describe what it would look like if your project failed completely. Focus on outcomes and attributes, not behaviors or causes. Don’t hold back or self-edit. Use extreme language and explore the worst-case scenarios. It is important for the facilitator to adopt a curious, open posture and create a “high-trust” environment where each participant is free to share and contribute.
- List the causes. Assemble a list of potential causes for the failure. Consider a wide range, including action, inaction, assumptions, priorities, distractions, people, organizations, and processes. Include anything that might be a contributing factor.
- Update goals and risks. Compare the failure description with the team’s current definition of success (goals, objectives, etc.) and the team’s current action plans. Discuss whether accomplishing the team’s goals and activities would prevent the future failure scenarios. Adjust goal statements as needed.
- Examine the list of causes and identify the most likely failure drivers. Discuss which causes are avoidable and whether the team is taking appropriate steps to mitigate or prevent them from occurring.
Key Concepts
Prospective Hindsight — The cognitive technique of imagining an outcome has already occurred and reasoning backward to explain it. Research shows this framing increases the team's ability to identify reasons for a future outcome by roughly 30% versus standard forecasting.
Psychological Safety — A high-trust environment where participants can voice doubts and worst-case scenarios without fear of being seen as disloyal. Without it, the premortem collapses into polite, low-signal answers that miss the real risks.
Failure Imagination — Deliberately describing vivid, extreme failure outcomes rather than hedged 'risks.' The emotional concreteness bypasses optimism bias and groupthink that normally suppress dissent in roadmap and planning sessions.
Outcomes vs. Causes — Separating the description of what failure looks like (outcomes, attributes) from why it happened (causes, drivers). Conflating the two too early narrows the search and lets the team skip past uncomfortable root causes.
