READING GUIDE · By ResearchWeave · September 14, 2026

How to Turn a Broad Topic into a Focused Research Question

“Tell me about cancer research” names a topic. It does not yet tell a reader which evidence would answer your question. Start by deciding what you need to do with the answer.

1. Name the task behind the search

Finish this sentence: “I need this answer so that I can …”

Choose a few original papers to read before a journal club.
Understand a proposed mechanism and the evidence behind it.
Compare how studies measured a particular outcome.
Check whether a specific published claim is supported.

These are different reading tasks. A broad overview may help you get oriented, but it cannot automatically substitute for a focused comparison or claim check.

2. Choose the boundaries that matter

Specify the elements relevant to your task; you do not need to fill every possible field.

Population or system: which condition, group, cell type or experimental setting?
Main subject: which intervention, exposure, mechanism or method?
Outcome: what do you want to explain or compare?
Comparator: compared with what, if a comparison is needed?
Time range: recent developments, a defined period, or foundational work as well?

A question about a mechanism may not need a comparator. A question about measurement methods may need an experimental setting more than a population. Avoid forcing every question into the same template.

A recent-publication limit is a choice about scope, not proof that older work is irrelevant. Record it so you can revisit it if important evidence is missing.

3. Rewrite the question around one answerable task

Illustrative question-writing example—not a research result

Broad topic: “CRISPR safety.”

More focused: “Which methods have been used to detect CRISPR-Cas9 off-target editing in human cells, and what limitations do the studies report for each method?”

Intended use: “Choose methods papers to read before a journal club.”

The rewritten question identifies a system, a technology and the information you want from the papers. It does not assume that one method is universally best.

If you want to ask “which is best?”, first define best: sensitivity, specificity, cost, experimental effort, or suitability for a particular setting. If the question contains several separate decisions, split it into smaller questions.

4. Check the question before searching

Ask yourself:

Would another reader know which studies belong in scope?
Would I recognize evidence that disagrees with my expectation?
Can I describe what a useful answer would contain?
Have I removed confidential details and unpublished proprietary information?

A copyable starting template

I want to understand [mechanism, intervention or method] in [population or system], focusing on [outcome or uncertainty].
I will use the answer to [reading task].
Include [relevant scope] and distinguish [important differences].

Treat this as a first draft. Too few relevant papers may mean the question is too narrow, the terminology differs, or evidence is scarce. Many loosely related papers may mean the scope needs tightening. Neither situation, by itself, establishes scientific novelty.

This guide helps frame an exploratory search. It is not a systematic-review protocol, a promise of comprehensive retrieval, or advice for an individual patient.

Bring a question you already need to answer

ResearchWeave’s early pilot focuses on English-language life-science questions. Explore the sample first, then ask whether your non-confidential question fits the current scope.

Read the sample excerpt Ask about a pilot trial

Access and trial availability are confirmed individually. Clicking a link does not start report generation.