Most risk registers are written by someone who doesn’t believe the project will fail. They list category risks (“vendor delay,” “scope creep,” “key person dependency”) with politely round probability scores and inoffensive mitigations. They get filed. They get presented monthly. They almost never name the actual thing that ends up sinking the project.
The pre-mortem fixes that.
What it is
The pre-mortem is a 30-minute exercise we run during Discover, before any plan has crystallised. The setup is one sentence:
Imagine it’s twelve months from now and this project has been a disaster. Why?
Everyone in the room — sponsor, owner, technical lead, MHA — writes silently for ten minutes. No discussion, no review of risk frameworks, no consulting a register. Just write the obituary. What happened? Who was furious? What were we doing in month three that we should have stopped doing in month two?
Then we go around the room. Each person reads their list. We capture every distinct failure mode. No debate, no dismissal, no “well, technically…” We just record.
That’s it. Half an hour.
Why it surfaces what frameworks don’t
Conventional risk identification asks “what could go wrong?” and gets back the categories you’re already prepared for. Pre-mortems ask “what did go wrong?” and get back the specific, embarrassing, named failures that nobody wants to write down in a formal document. The mental shift from prospective to retrospective unlocks the things people privately suspect but won’t say in a status meeting.
We’ve had pre-mortems surface:
- “The owner stopped responding to permit questions in month two because they were dealing with a family emergency and nobody noticed.”
- “We picked the cheapest contractor and they ghosted us in week six.”
- “Karim flew to Tunisia for two weeks and the project sat idle because no one else had decision authority.”
- “We were so focused on the demolition timeline that nobody verified the buyer’s financing.”
None of these are on any standard risk template. All of them are real failure modes that experienced people already knew about but wouldn’t have written in a “Project Risk Identification” Word doc.
What we do with the output
The pre-mortem output flows directly into the Plan phase, but with two filters:
-
De-duplicate by failure mode, not wording. “Vendor ghosts us” and “contractor stops returning calls” are the same risk; capture it once. Around 30–40% of pre-mortem output is duplicates rewritten — that’s a signal of which failure modes are most salient, not a problem.
-
Triage by what we’d do differently right now. A risk you can act on this week beats a risk you’ll monitor for nine months. We sort the de-duplicated list into three buckets:
- Design out — change the plan so this failure can’t happen (e.g., “pick a contractor with bonded references and a 50% retainer structure” eliminates the ghosting risk).
- Instrument — accept the risk but add a tripwire that catches it early (e.g., “weekly check-in with the owner specifically about permits” catches the family-emergency case).
- Accept — it’s a real risk, we can’t design it out, we don’t have a useful tripwire, we’ll handle it if it happens.
Bucket three is the smallest, by design. Most risks are designable-out or instrumentable; the work is identifying which.
When to skip it
Two situations where we don’t bother:
- Engagements under two weeks. The overhead doesn’t pay for itself; we still ask the four questions informally during the kickoff call.
- Repeat engagements with the same client and same problem shape. The pre-mortem from the first one is still relevant; we re-read it and update rather than redoing the exercise.
The pre-mortem template we use
It fits on one page. Four prompts, ten minutes silent, twenty minutes capture.
- Twelve months from now, this project has failed. What happened?
- Which decision, made in the first month, set the failure in motion?
- Who saw it coming but didn’t speak up — and why didn’t they?
- What’s the cheapest thing we could do in the next two weeks to prevent this?
The fourth question is the one that makes the whole exercise useful. The first three surface the risk. The fourth converts it into an action that lands in Plan.
If you want a copy of the template we use, email contact@mha-management.com.