CORE GUIDE
Research question decomposition
A strong research workflow converts a broad question into smaller claims and evidence needs before searching, so synthesis is driven by what must be established rather than by whichever sources appear first.
Mental model
Research decomposition creates an evidence map: define the decision, split it into answerable subquestions, state what evidence could support or challenge each one, then search and synthesize against that structure.
Why it matters
AI-assisted research can become search-result summarization. The first persuasive sources shape the narrative, missing counterevidence stays invisible and the final answer sounds more complete than the investigation actually was. Decomposing the question before collection reduces this anchoring and makes unresolved gaps explicit.
01
Move from decision to subquestions to evidence criteria
Clarify what decision or understanding the research must support. Break it into factual, causal, comparative or forecasting subquestions and define the type of evidence each requires. Identify key terms, time boundaries and likely disagreements before searching. During collection, attach sources to the relevant subquestion and mark areas where evidence remains missing or contradictory.
02
Example: should a solo founder enter a market?
Instead of searching for 'is this a good market', the research plan separates customer pain, existing alternatives, willingness to pay, acquisition channels, market constraints and founder advantage. Each subquestion has different evidence needs. The final recommendation can then show which assumptions are well supported and which remain speculative.
Common failure modes
- Starting broad search before defining what claims need evidence.
- Treating a long list of sources as proof that all parts of the question were investigated.
- Forcing every subquestion into a confident answer even when evidence is sparse.
Engineering heuristics
- Write the decision and key subquestions before collecting sources.
- Define evidence criteria that could disconfirm attractive hypotheses.
- Carry unresolved gaps and disagreements into the final synthesis.
Takeaways
- 01Decomposition protects research from search-result anchoring.
- 02Different subquestions require different evidence.
- 03A good synthesis shows both supported conclusions and remaining uncertainty.
Used in
This Concept is reused across these canonical learning paths.
Related concepts from the Knowledge Graph
These relationships come from the canonical graph, not a separate Guide taxonomy.