Consensus

Consensus is built on a simple and powerful idea: when you ask a research question, you should get an answer that reflects the state of scientific evidence, not just a list of papers. The platform searches over 200 million peer-reviewed papers, extracts key findings from each relevant result, and synthesizes them into a Consensus Meter — showing what percentage of papers agree, disagree, or are inconclusive on the question asked.

The Consensus Meter

The signature feature is the Consensus Meter, which appears at the top of results for empirical research questions. If you search "Does intermittent fasting reduce blood pressure?", the meter shows a breakdown across relevant studies — say 67% support, 18% inconclusive, 15% mixed — along with the individual study findings below. This is fundamentally different from Google Scholar, which returns a ranked list with no synthesis. The meter gives you the shape of the evidence before you read a single paper.

• Consensus Meter — synthesized percentage breakdown of what peer-reviewed literature says about your question.
• Study snapshots — one-sentence findings extracted from each paper, plus population, methods, and sample size tags.
• GPT-4 Synthesis — Premium users get a paragraph-length evidence summary written by GPT-4 with citations.
• Filters — filter by study type (RCT, systematic review, meta-analysis), journal quality, sample size, and publication year.
• Citation export — export references in APA, MLA, or BibTeX from results directly.

Who it's built for

Consensus is most useful for evidence-based professionals: clinicians doing rapid evidence checks, health journalists verifying medical claims, nutrition researchers, and science communicators who need to accurately represent what the literature says. It also serves students learning to evaluate evidence quality rather than just finding papers. The Consensus Meter makes it excellent for fact-checking health and science claims — particularly useful in contexts where misinformation is common.

Coverage and limitations

The 200M+ paper database skews toward biomedical, psychology, and social sciences — STEM coverage is strong while humanities and law are limited. The Consensus Meter is computed probabilistically and can be misleading if the included studies vary dramatically in quality; a meta-analysis carries more weight than five small observational studies but the meter doesn't always reflect that weighting. Premium filters for study type and journal quality help mitigate this.

Pricing

The free tier allows 20 searches per day with study snapshots and basic filters — enough for occasional research. Premium at $8.99/month adds unlimited searches, GPT-4 synthesis, advanced filters, and citation export. Annual pricing at $74.99 ($6.25/month effective) is the best value for regular users.

Try it at consensus.app — no sign-up needed for free searches.