Skip to content
Research & AnalysisAdvanced

Turn a Big Question Into a Research Plan

Break a fuzzy research question into sub-questions, search strategies, source types, and what evidence would change your mind.

Share on X (opens in a new tab)Share on LinkedIn (opens in a new tab)

Make it yours

Fill in each variable and watch the prompt update. Anything still in {{braces}} needs your input before you use it.

Fill in the blanks

0 of 4 filled

Your research question, even if it's rough.

What the research is for and who will read it.

How much time you can spend.

Your current hunch or what you've already read.

Live previewUnfilled
You are a research methods advisor helping me plan an investigation before I start searching.

Research question: {{question}}
Purpose: {{purpose}}
Time available: {{time_budget}}
What I already know or suspect: {{prior_knowledge}}

Produce a research plan with these sections:
1. **Sharper question** — rewrite my question so it is specific and answerable; note any ambiguous terms I need to define.
2. **Sub-questions** — 4–6 sub-questions, ordered so earlier answers inform later ones.
3. **Where to look** — for each sub-question, the types of sources most likely to help (e.g. systematic reviews, government statistics, industry reports, interviews) and 2–3 example search queries I could type.
4. **Evidence that would change my mind** — what findings would weaken my current suspicion.
5. **Pitfalls** — common biases or data problems in this topic.
6. **Time plan** — how to split my available time across the steps.

Important:
- Do NOT name specific papers, authors, datasets, or URLs unless you are highly confident they exist; describe the kind of source instead. If you do name one, mark it "(verify)".
- Flag any part of the plan that depends on information you may not have, such as recent developments after your training data.
Open in Playground

Example

Example input

Filled prompt
You are a research methods advisor helping me plan an investigation before I start searching.

Research question: Does remote work hurt junior employees' career growth?
Purpose: A 2-page briefing for our HR leadership team deciding on a hybrid policy
Time available: about 6 hours over one week
What I already know or suspect: I suspect juniors get less mentoring remotely, but I've only seen opinion pieces.

Produce a research plan with these sections:
1. **Sharper question** — rewrite my question so it is specific and answerable; note any ambiguous terms I need to define.
2. **Sub-questions** — 4–6 sub-questions, ordered so earlier answers inform later ones.
3. **Where to look** — for each sub-question, the types of sources most likely to help (e.g. systematic reviews, government statistics, industry reports, interviews) and 2–3 example search queries I could type.
4. **Evidence that would change my mind** — what findings would weaken my current suspicion.
5. **Pitfalls** — common biases or data problems in this topic.
6. **Time plan** — how to split my available time across the steps.

Important:
- Do NOT name specific papers, authors, datasets, or URLs unless you are highly confident they exist; describe the kind of source instead. If you do name one, mark it "(verify)".
- Flag any part of the plan that depends on information you may not have, such as recent developments after your training data.

Example output

Illustrative example response — written by Promptly, not generated live

1. Sharper question "Compared with in-office peers, do early-career employees (0–3 years) in hybrid or fully remote knowledge-work roles receive less mentoring, feedback, and promotion?" Define: junior, remote vs. hybrid, career growth (promotion rate? skills? pay?).

2. Sub-questions

  1. How do researchers measure career growth for early-career staff?
  2. Is there evidence on feedback/mentoring volume by work arrangement?
  3. Are promotion or pay outcomes different?
  4. Which factors (structured mentoring, office days) change the effect?
  5. What do our own employees report?

3. Where to look

  • Q2–3: peer-reviewed labor/management studies, working papers from economics research institutes. Queries: "remote work" junior feedback field experiment; hybrid work promotion early-career
  • Q4: HR practitioner surveys (treat as lower-quality evidence)
  • Q5: an anonymous internal pulse survey

4. Evidence that would change my mind Studies finding no mentoring gap when structured onboarding exists, or equal promotion rates after controlling for role.

5. Pitfalls Self-selection into remote roles; pandemic-era data that may not generalize; opinion pieces presented as research.

6. Time plan 1h scoping, 3h sources, 1h internal data, 1h writing.

Note: research in this area is evolving; check for studies published after my training data.

Why it works

Planning before searching is a form of prompt chaining: you get a structure you can then work through one sub-question at a time. Explicitly forbidding invented papers (and requiring "(verify)" on any named source) addresses one of the most common AI research failures — plausible-looking but fake citations. The "evidence that would change my mind" section builds in protection against confirmation bias.

Techniques used