Build a Research Plan
Turn a broad research topic into an ordered set of specific sub-questions and source types, before spending time answering anything.
Prompt template
Replace the parts in curly braces: research topic or question, how deep/broad the research needs to be, intended output (report, presentation, decision)
I need to research {research topic or question} in order to produce {intended output}. Propose a research plan: the specific sub-questions I should answer, in what order, and what kind of source would answer each one (e.g. official docs, academic paper, primary data). Don't answer the questions yet — just the plan.
Depth needed: {how deep/broad the research needs to be}Example
Input: A broad topic that needs to become a structured research plan before diving in.
Output: Illustrative only. A good result is an ordered list of sub-questions with a source type suggested for each — not the answers themselves, which is the point of asking for a plan first.
Quality checklist
- Check the proposed sub-questions actually cover the real question, not just adjacent topics that were easy to generate
- Verify at least one suggested source type actually exists and is accessible to you before committing to the plan
Where this tends to go wrong
- A generated plan can look thorough while missing the one sub-question that's actually hardest to answer
- Suggested source types are sometimes generic ("industry reports") without pointing to anything findable