Direct Gemini to map a market by segment, size each claim with a source and date, and flag every number it cannot support so you know where to dig.
The prompt
Produce a market scan of [MARKET] for [PURPOSE]. Break it into segments. For every quantitative claim, attach the source and its date, and mark estimates as ESTIMATE. End with the three biggest unknowns and what evidence would resolve them. Do not present unsupported numbers as facts.
Replace the variables
[MARKET]Category + geography + timeframe.
[PURPOSE]Investment, entry, positioning, etc.
Which model to use
- Gemini 3.1 Pro — Structured, multi-segment synthesis is more reliable on Pro.
Example input
- [MARKET] = "EU B2B e-signature tools, 2026". [PURPOSE] = "assess entry for a new startup".
Example output
We only publish example outputs from real model runs. This template has not been formally tested yet, so no output is shown. Run it yourself with the input above.
Why this structure works
- Segment-first structure prevents a vague, averaged answer.
- Tagging ESTIMATE vs sourced keeps you honest with stakeholders.
Common failures & fixes
Numbers look precise but are unsourced.
Fix: Reject any figure without a named source; ask the model to downgrade it to a range.
Known limitations
- Market sizes drift fast; re-run and re-verify before publishing.