The Scientific Answer Engine
Consensus is the AI research tool built for a specific question: “What does the science say?” Instead of searching the web, it searches a database of 200+ million peer-reviewed papers, finds the relevant studies, and synthesizes what the evidence collectively shows. For evidence-based questions — health, psychology, economics, environmental science — it is the fastest way to get a research-backed answer.
Where Perplexity searches the open web (including low-quality sources), Consensus searches only peer-reviewed literature. This makes its answers more trustworthy for scientific questions but useless for non-scientific ones. It is a specialist, not a generalist.
The Consensus Meter
The Consensus Meter is the feature that makes Consensus unique. Ask “Does creatine improve athletic performance?” and the meter shows: 85% Yes, 10% Possibly, 5% No — based on the aggregated findings of relevant studies. This gives you an instant read on scientific consensus without reading 50 papers.
The meter is not a vote count. It weighs studies by sample size, study design, and recency. A meta-analysis of 10,000 participants counts more than a pilot study of 20. This weighting is what separates Consensus from a simple literature search — it is doing the evidence evaluation work that a systematic review would do, but in seconds instead of months.
Evidence Synthesis
Consensus doesn’t just show you papers — it reads them and synthesizes their findings. For a question like “What is the effect of sleep deprivation on cognitive performance?”, it will:
- Find relevant studies across its database
- Extract the key findings from each
- Group findings by outcome (reaction time, memory, decision-making)
- Note where studies agree and disagree
- Present a synthesized answer with citations to the underlying papers
| Feature | Consensus | Perplexity (Academic) | Elicit |
|---|---|---|---|
| Source base | 200M+ peer-reviewed | Web + some papers | Semantic Scholar |
| Consensus indicator | Yes (Consensus Meter) | No | No |
| Study quality indicators | Yes | No | Yes (some) |
| Structured data extraction | No | No | Yes |
| Best for | ”What does science say?” | General research | Paper-level analysis |
Study Quality Indicators
Each study in Consensus’s results includes quality indicators: study design (RCT, observational, meta-analysis), sample size, and publication date. This helps you evaluate the strength of the evidence. A claim supported by a meta-analysis of 50 RCTs is stronger than one supported by a single observational study — and Consensus makes that distinction visible.
Where It Falls Short
Consensus is only as good as its database. It covers major scientific disciplines well — medicine, psychology, biology, economics, environmental science — but is thinner in humanities, computer science, and emerging fields where research is published in conference proceedings rather than journals. If your field publishes primarily in conferences (CS, AI), Consensus will miss much of the relevant work.
The synthesis quality depends on research volume. For well-studied topics (creatine, sleep, mindfulness), Consensus has hundreds of studies to draw from and its syntheses are robust. For emerging topics with few studies, the synthesis may be thin or based on a handful of papers. Consensus is honest about this — it shows you exactly how many studies inform each answer — but it means the tool is more useful for established research areas than cutting-edge ones.
Pricing
| Tier | Price | Searches | Fit |
|---|---|---|---|
| Free | $0 | Limited | Testing, casual use |
| Premium | $9/mo | Unlimited | Regular research |
| Teams | $15/user/mo | Unlimited + sharing | Research teams |
At $9/mo, Consensus Premium is one of the cheapest AI research tools. For academics, healthcare professionals, and evidence-based decision makers, it pays for itself in time saved on literature review. Against Perplexity Pro ($20/mo) and Elicit ($10/mo), it is competitively priced for a specialist tool.
Verdict
Consensus is the right tool for anyone who needs to know what the scientific evidence says on a topic — quickly and with citations. It is not a general-purpose search engine; for non-scientific questions, use Perplexity. For deep, structured analysis of individual papers, use Elicit. Consensus occupies a specific niche — evidence synthesis from peer-reviewed literature — and it is the best tool in that niche. For researchers, healthcare professionals, policy analysts, and science journalists, it is an essential tool at a price that is hard to beat.