10 research prompts

Competitor analysis, market framing, extracting the real question and checking primary sources. Written for ChatGPT, Claude, Perplexity. Square brackets mark what you fill in; everything else can be pasted as it stands.

I want to find out about [topic].

Before researching anything, help me state the question properly. Give me:
- Three sharper versions of my question, each answerable with evidence
- What kind of source would actually answer each one
- Which of the three I probably mean, and why

Do not start answering the question. I want the question first.

I have thirty minutes with [person and their role] about [topic]. I already know: [what you know].

Write ten questions, ordered so that the most valuable come first in case we run out of time.

Rules: no question answerable by a web search, no question with a yes/no answer, and at least two that invite them to disagree with me.

Here is what I know about [market]: [paste what you have].

Organise it by the axis that actually divides the players — not by company size or founding date. Say what that axis is and why it is the one that matters.

Then show where each player sits, and name the gap nobody occupies. If the gap is empty for a good reason, say what that reason is.

Here is a source I am thinking of citing: [paste or describe it].

Tell me:
- Who produced it and what they gain if I believe it
- When the underlying data was collected, as distinct from when it was published
- What it measured, exactly, and what it did not
- Which claim in it is weakest

If any of these cannot be determined from the material, say so explicitly rather than guessing.

I am deciding whether to [decision]. Here is a long document: [paste it].

Pull out only what bears on that decision. For each item, give the fact, where in the document it appears, and which way it points.

Ignore everything interesting but irrelevant. End by stating what the document does not tell me that I would still need.

I believe that [claim], and I would like it to be true.

Argue the other side as strongly as you can. Specifically:
- The strongest evidence against it
- The most likely way I am fooling myself
- What would have to be observed for me to be wrong

Do not reassure me. If the claim survives, say what it survived.

Summarise what is known about [market] based only on the material below.

Mark every number with its source and date. Where I have not given you a figure, write [no figure in sources] rather than estimating.

End with the three things I would need to find out before making a decision.

<sources>
[paste the sources]
</sources>

I am researching [question] to decide [decision]. So far I have found: [summarise].

Tell me whether more research would change the decision. Specifically:
- Which open question, if answered either way, would flip the choice
- Which open questions would not change anything
- What I should stop looking into

If nothing would flip the decision, say that I should decide now.

Here are several sources on the same question: [paste them].

Identify every point where they disagree, including quiet disagreements where they define a term differently or measure a different thing.

For each, say which source is better positioned to be right and why. Do not average them into a consensus.

Here is vendor material for [product]: [paste the page or document].

Split every statement into three lists: contractual commitments, measurable claims, and marketing language that cannot be tested.

For the second list, say how I would verify each during a trial. For the third, say what question would force a concrete answer.