Julia Moser·
Generated a project risk register from nothing but the kickoff brief — probability, impact, owners, and triggers. The steering committee thought I spent a day on it.
Derives a project risk register from a kickoff brief, with scored risks, response strategies, owners, and review triggers.
Kickoff Brief to Risk Register
You are a project risk manager. Derive a risk register from this project kickoff brief:
{{kickoff_brief}}
Additional context — team, budget band, deadline, and known organizational constraints: {{project_context}}
Build the register:
1. **Risk identification.** Extract 8-15 risks across these categories: scope (ambiguities and unstated assumptions in the brief — quote the vague sentence that spawned each), schedule (dependencies on parties outside the project's control), resource (single points of failure in the team, competing priorities), technical (unproven integrations or estimates stated with suspicious confidence), and stakeholder (approval bottlenecks, silent stakeholders who can veto late). Every risk must be phrased as cause → event → consequence ("Because the API vendor's timeline is unconfirmed, the integration may start 3 weeks late, delaying UAT past the freeze window").
2. **Scoring.** Probability (1-5) and impact (1-5) with one line of justification each — no bare numbers. Compute exposure (P×I). Present as a table sorted by exposure.
3. **Response per risk.** Choose avoid / mitigate / transfer / accept, then one concrete action for the chosen strategy, an owner role, and — critically — a trigger: the observable early-warning sign that this risk is materializing, plus the pre-agreed response when the trigger fires. A risk without a trigger is a worry, not a managed risk.
4. **Top 3 deep-dive.** For the three highest-exposure risks: what the brief would need to change to reduce inherent exposure (not just manage it), and what question to ask at the next steering meeting.
5. **Register hygiene.** Propose the review cadence, who updates it, and the rule for retiring risks. One short paragraph.
Tone: precise, no drama. The register should make the project look managed, not doomed.
Ergebnisse
## Risk Register: CRM Migration Project (v1.0, from kickoff brief)
### Register (sorted by exposure, excerpt)
| # | Risk (cause → event → consequence) | P | I | Exp. |
|---|---|---|---|---|
| R1 | Because "data cleanup will happen alongside migration" (brief §3, no owner named), cleanup may stall mid-project → migrated CRM inherits duplicate/dead records → adoption fails on day one | 4 — no owner is a proven stall pattern here | 5 — adoption is the project's stated success metric | **20** |
| R2 | Because the ERP connector is "expected to work with minor adjustments" (§5 — suspicious confidence, never tested), integration effort may triple → go-live slips past the November freeze | 3 | 5 | **15** |
| R3 | Because Sales leadership was consulted only at kickoff, requirements may surface during UAT → late scope additions or a veto in steering | 3 | 4 | **12** |
| R4 | Because both technical migrators also support the Q4 reporting release, capacity may drop 50% in October → schedule slip | 4 | 3 | **12** |
| R5 | Because trainer availability is unconfirmed for November, training may compress to one week → shallow adoption | 2 | 4 | 8 |
### Responses (top rows)
- **R1 — Mitigate.** Action: split cleanup into a named workstream with its own owner and a done-definition before migration wave 1. Owner: data workstream lead. **Trigger:** cleanup backlog burndown flat for 2 consecutive weeks → response: escalate to sponsor, descope wave 1 to clean segments only.
- **R2 — Mitigate.** Action: 3-day connector spike in the first two weeks, before the plan hardens. Owner: tech lead. **Trigger:** spike reveals >5 days of adaptation work → response: pre-approved decision point: reduce phase-1 integration scope OR move go-live, decided in steering within one week.
- **R3 — Mitigate.** Action: standing Sales rep in the biweekly review with sign-off duty per milestone. Owner: PM. **Trigger:** rep misses two consecutive reviews → response: PM escalates to Sales director before UAT, not during.
### Top 3 Deep-Dive (R1)
Brief change that reduces inherent exposure: replace "alongside migration" with a sequenced precondition — wave 1 migrates only segments passing a defined quality gate. Steering question: *"Who owns data cleanup, and what happens to the timeline if the answer is 'the team, collectively'?"*
### Register Hygiene
Review biweekly in the project review, 15 minutes, PM updates. Risks retire when their trigger window passes or exposure drops below 4 — retired, not deleted, with a one-line outcome note. New risks enter through anyone, scored by the PM within 48 hours.
Modell: Claude Sonnet 4.6
29 Likes18 SavesScore: 20
2 Kommentare
Anna Hofmann·
'A risk without a trigger is a worry, not a managed risk' — that sentence is worth the whole prompt.
Maya Patel·
Quoting the vague brief sentence that spawned each risk is elegant. The register becomes evidence for fixing the brief itself.