Research brief writer

A Claude skill that turns a business question into a research brief, naming the method to run and the decision waiting on the result, with the limits written in.
Study design
June 25, 2026
Download the skill
$ claude skill install \
greatquestion/research-brief-writer
Free and open. No account needed. Read the install guide →

What it does

A research brief is the document fixing what a study is for, written before anyone books a session. Some teams call it a research plan or a project brief and the contents run much the same. Its real work is translation. A stakeholder arrives holding a business question, and the brief turns that into research questions a method can answer, along with the decision sitting downstream waiting on the result.

This skill takes the business question in whatever state it reaches you, a Slack message included, and builds the brief around it. It proposes a method and explains why that method suits the question. The alternatives it passed over are named alongside it, so you can argue with the choice in review. When the question isn't researchable it says so early. Some questions belong in your analytics, and some are decisions the team already made and would like cover for.

What you get back

  • The business question restated as research questions a study can actually answer.
  • The decision this research feeds, named alongside the person who owns it.
  • A method recommendation with the reasoning, plus the alternatives it considered.
  • A sample definition and the recruitment criteria following from it.
  • A timeline shape from kickoff to readout, marking the dependencies that usually slip.
  • Risks: what could leave the study inconclusive, and what to do about each one.

How to use it

Install it once, then ask in plain language. Claude picks the skill up on its own when the request matches.

Example Prompts:

"Write a research brief for this: we think the new pricing page is confusing people."

"Turn this Slack thread into a brief, and tell me whether it's even a research question."

"Give me that brief again with an unmoderated method in place of interviews."

What it won't do

It won't know your capacity. Method recommendations come from the question and say nothing about whether your panel can supply that sample, or whether anyone has two weeks free, so treat the timeline as a shape to argue with. It also won't judge whether the underlying decision deserves research. It can only tell you when the question as written has no study behind it.

Questions

What's the difference between a research brief and a research plan?

Most teams use the words interchangeably. Where they do differ, the brief is the shorter document that wins agreement on why the study exists, and the plan carries the operational detail once that agreement is in hand. This skill writes the brief and will extend it into a plan when you ask for the detail.

Will it recommend a method I have no way of running?

It can, unless you tell it what you're working with. Name your constraints in the request, say an unmoderated-only quarter or a panel with no enterprise admins in it, and the recommendation gets made inside those limits. Left unconstrained, it recommends what the question deserves and you negotiate from there.

Can it work from a stakeholder request that isn't really a research question?

That's one of the better uses for it. Paste the request in and it separates the parts a study can answer from the parts needing instrumentation or a product decision. When nothing is left for research it says so, with reasoning you can forward to the stakeholder.

Learn the method
Our full guide to running an affinity mapping session, including how to tell when the groups are wrong.
Read the guide →
Synthesis, without the sticky notes
Great Question stores every session, transcript and highlight in one place, so the mapping starts from evidence instead of memory.
Book a demo