📚 Research & Learning

How to Use Consensus for Evidence-Based Research Answers

A practical guide to phrasing a question the Consensus Meter can handle, reading study snapshots instead of full papers, and knowing when you need a broader tool.

What you'll be able to do by the end

  • Understand that Consensus is built around a single research question answered by the Consensus Meter, which shows the share of studies agreeing, disagreeing, or remaining inconclusive
  • Phrase your question in the clear 'does X affect Y' form so the meter works accurately
  • Know the meter is a fast automated starting point, not a final verdict — for any consequential decision you still read the papers themselves
  • Distinguish unlimited free search from the monthly Pro Analysis and Study Snapshot caps on the free plan
  • Know its limits: an overwhelmingly English database, and it's neither a general research tool nor a substitute for a full literature review

Before you start

  • A specific, empirically testable question (does X affect Y) — not a broad topic or a complete literature review
  • A free account is enough to start; you only need Premium or Pro if you exceed the monthly deep-analysis limits

Consensus is an evidence-based search engine built around one specific problem: you want an answer backed by scientific evidence to a research question, not just a list of links. You write your question in plain language; it searches a database of more than 200 million peer-reviewed papers and returns real ones, each tied to its direct source.

Its signature feature is the Consensus Meter: for a question like “does X cause Y”, it shows visually how many studies agree, disagree, or are inconclusive, based on analysis of the relevant papers. That’s an excellent fast starting point for understanding where science stands — but it’s an automated summary, not a final ruling that replaces reading the papers when the stakes are real.

This is also the clearest difference from Elicit: Consensus is built around one research question at a time and answers it through the meter, while Elicit is broader — full systematic reviews and extracting data from dozens of papers into one table. Specific question, quick answer? Consensus. Full literature review? Elicit.

The steps

The steps below run from phrasing the question to exporting references. If your question is simple, you can stop after the meter and Study Snapshot steps.

Steps

  1. Step 1: Create a free account and write your question in evidence-answerable form

    Sign up at consensus.app with a free account. Instead of typing a general topic, phrase your question in plain language, ideally in the form 'does X affect Y' — that exact form is what the Consensus Meter works with correctly, because the tool is built around answering one research question with evidence, not returning a list of links.

    Note: The free plan includes unlimited search, so start there before considering any upgrade.

  2. Step 2: Read the Consensus Meter as a quick starting point, not a verdict

    After you ask, the Consensus Meter appears visually: the share of studies that agree, disagree, or are inconclusive on your question, based on analysis of the relevant papers. It's very useful as a fast read of scientific consensus on something, but it's an automated summary — not a replacement for reading the papers when the decision matters.

    Note: If the agree and disagree shares are close, that's itself a signal the topic is scientifically contested and worth deeper reading.

  3. Step 3: Open any paper and read its quick snapshot instead of the whole thing

    Every result links to a real paper from a database of more than 200 million papers, with a direct link to the source. Open any paper of interest and check its Study Snapshot — a quick summary giving you the methodology, sample size, and main finding without reading the full paper.

    Note: Use Study Snapshot for rapid triage between papers, and return to the full paper only when you genuinely need fine detail.

  4. Step 4: For complex questions needing multi-source evidence, use Deep Search

    If your question goes deeper than a quick answer and needs broader analysis across more papers, use Deep Search (Pro plan): it delivers deeper analysis drawing on more sources for the same question, useful for complex questions needing evidence from multiple angles.

    Note: The Premium plan unlocks unlimited analysis and summaries, but Deep Search specifically requires Pro.

  5. Step 5: Export the references you need straight to your citation manager

    Once you've picked the papers you'll cite, export them directly to Zotero, Mendeley, or EndNote instead of copying details by hand, so they slot straight into your research workflow or manuscript.

    Note: Export references as soon as you've confirmed them, rather than batching them at the end and forgetting where each came from.

Common mistakes — and how to avoid them

MistakeRelying on the Consensus Meter percentage alone as the final answer without opening a single paper.

Do this insteadOpen the Study Snapshot or original paper for any result before citing it in serious work — the meter is a fast automated summary, not a substitute for reading.

MistakeWriting your question as a vague, untestable prompt — an opinion question or a broad topic.

Do this insteadPhrase it in a specific 'does X affect Y' form so the Consensus Meter can analyse the studies and work accurately.

MistakeUsing it for a full literature review that needs extracting data from dozens of papers into one table.

Do this insteadUse Elicit for that purpose. Consensus is built around one research question at a time, not comprehensive systematic review.

❓ Frequently asked questions

What's the difference between Consensus and Elicit?

Consensus is built around a single research question answered through the Consensus Meter, which shows the agreeing share of studies visually — faster when your question is specific and you want a quick evidence-backed answer. Elicit is broader: full systematic reviews and data extraction from dozens of papers into one organised table. For a complete literature review, choose Elicit.

Is the Consensus Meter a final verdict on the question?

No. It's an automated summary based on analysis of the relevant papers, useful as a quick starting point for understanding scientific agreement — but not a replacement for reading the papers themselves when the decision matters.

How much does Consensus cost?

There's a free plan: unlimited search, capped at 10 Pro Analysis runs and 10 Study Snapshots per month, plus basic browsing and reference export. Premium at $8.99/month (or an annual discount) unlocks unlimited Pro Analysis and Study Snapshots plus advanced result filtering. Pro at $15/month adds 15 Deep Searches per month on top of everything in Premium.