Prompt: Write a Market Research Brief

You are a market analyst who is careful to separate evidence from assumption.

Objective: Write a research brief on {{MARKET OR QUESTION — e.g. demand for X among Y}}.

Context: What's already known or supplied: {{EXISTING DATA OR CONTEXT}}. What you're trying to decide based on this: {{THE ACTUAL DECISION THIS INFORMS}}.

Constraints:
- Clearly mark any claim you're not certain about as an assumption, not a fact
- Do not cite specific statistics unless they were supplied in the context — describe general patterns instead
- Include at least one factor that could invalidate the brief's conclusion
- Keep the brief to 400–600 words

Output format: Return sections for Summary, What We Know, Assumptions, Risks to This Analysis, and Recommendation.

Before finishing, check that every statistic mentioned was actually supplied in the context, and that assumptions are labelled as such, not stated as fact.

MARKET OR QUESTION — e.g. demand for X among Y: demand for a subscription meal-kit service among students in university accommodation
EXISTING DATA OR CONTEXT: a signup page collected 240 email registrations in two weeks after being shared in three university Facebook groups, with no paid ads run
THE ACTUAL DECISION THIS INFORMS: whether to run a paid pilot in one city this term

Summary
Early, informal signal suggests some organic interest in a student-focused subscription meal-kit service, but the available data is too thin on its own to justify a full launch. A signup page shared organically in three university Facebook groups collected 240 email registrations in two weeks, with no paid promotion behind it. That is a genuine signal of curiosity, not evidence of willingness to pay, and the gap between the two is the central question this brief tries to frame rather than answer.

What We Know
- 240 email registrations were collected in two weeks from three university Facebook groups.
- No paid advertising or promotion was run to generate this interest; all registrations came from organic sharing.
- The registrations were email sign-ups; the supplied data doesn't say whether the page asked for anything beyond an email address.

Assumptions
- Assumption: registrations from a free, low-friction signup page likely overstate actual willingness to pay for an ongoing subscription, since committing an email address carries far less cost than committing a card.
- Assumption: interest is not evenly distributed across the three universities where the page was shared, though the supplied data doesn't break registrations down by group, so this cannot currently be confirmed either way.

… (excerpt — the full brief runs to about 424 words, continuing with a third assumption, three risks to the analysis, and the recommendation)

Why the basic version fails

A generic “research X for me” prompt often returns confident-sounding statistics the model has no actual source for, presented with the same certainty as things that are genuinely known. That’s the single most dangerous failure mode for a research brief someone will act on.

What this prompt fixes

Banning invented statistics and requiring assumptions to be explicitly labelled turns an overconfident guess into an honest, checkable brief. The “risks to this analysis” section does the job a good analyst would do anyway — stating what would change the recommendation.

Tips per model

• ChatGPT will still occasionally slip in a plausible-sounding statistic; the explicit “don’t cite unsupplied statistics” constraint and the finishing check both help catch this.
• Claude tends to flag its own uncertainty more readily, which pairs well with this prompt’s assumption-labelling requirement.
• Supply whatever real numbers you actually have, even informal ones like signup counts — the brief is only as evidence-based as the context you paste in.