Prompt library
Expose the assumptions inside a forecast
Sensitivity-ranked assumptions with plausible-pessimistic scenarios produce planning conversations instead of the dismissal that doom scenarios trigger. The correlation-trap section fixes the standard modeling sin of stress-testing inputs one at a time when reality moves them together.
Last reviewed July 17, 2026
The prompt
Expose every assumption inside the forecast below.
The forecast and its model or logic: {{forecast}}
Produce:
1. Assumption inventory: every input belief, explicit or buried (growth rates, conversion stability, seasonality repeating, no competitor response, price holding, capacity limits ignored). For each: its current value, where it came from (data, convention, hope), and its blast radius on the output (high/medium/low).
2. The load-bearing three: the assumptions the forecast is most sensitive to. For each, show the output under a plausible pessimistic value, not a catastrophic one; catastrophes get dismissed, plausible pain gets planned for.
3. Correlation traps: which assumptions fail together (a downturn hits acquisition AND churn AND payment failures), because independent-failure math understates real risk.
4. Expiry dates: which assumptions have a natural check-by date, and what signal would invalidate each early.
5. The rewrite: the forecast's headline number restated as a range with its real drivers ("4.2M if retention holds at 92 percent; each retention point is worth 180k").
A forecast is a stack of assumptions wearing a number; make the stack visible.Run in idaptOpens a new chat with the prompt prefilled. Nothing sends until you press send.
Fill in the variables
| Variable | What it is | Example |
|---|---|---|
| {{forecast}} | The forecast and how it was built | [paste the model logic or the plan's numbers section] |