Review

Rayyan: A focused system for evidence-synthesis screening

Rayyan is strongest when systematic-review screening and collaboration matter more than open-ended discovery.

Last updated April 2026 · Pricing and features verified against official documentation

Rayyan sits in a narrow but important part of research software: the ugly middle of a systematic review, where hundreds or thousands of citations have to be screened, disputed, deduplicated, and documented. That job is too structured for a general AI chatbot and too repetitive for a purely manual workflow. Rayyan works because it stays close to the task instead of trying to become a universal research assistant.

The product has also moved beyond its original screening-only identity. It now includes data extraction, a mobile app, an API, and an institutional AI suite layered on top of the core review workspace. That expansion matters because the real problem in evidence synthesis is not just finding papers; it is keeping a team aligned from the first abstract pass through the final extraction table.

The best case for Rayyan is straightforward. If your team needs a practical review workspace, collaborative screening, and a licensing model that does not require enterprise procurement for every small project, it is a strong fit. The free tier is real, the paid individual plans are inexpensive, and the institutional plans are built around the actual shape of research teams rather than around generic seat bundles.

The case against it is just as clear. Rayyan is specialized, its AI is helpful rather than magical, and its privacy story is acceptable without being especially modern or explicit. If your main need is broad literature discovery, drafting, or an open-ended research copilot, there are better tools. Rayyan is the right choice when screening discipline is the bottleneck.

What the Product Actually Is Now

Rayyan is no longer just a title-and-abstract screener. It is now a workflow platform for systematic reviews, with separate individual and institutional subscription tracks, role-based collaboration, duplicate resolution, PRISMA flow support, data extraction, and review chat. The public help center now documents the newer subscription structure, seat handling, and the current data-extraction stage.

That shift gives the product a clearer center of gravity. The core job is still evidence synthesis, but Rayyan now tries to cover more of the path between imported references and a finished review. ResearchPilot, the AI reviewer/analyzer/auto-extract package on institutional plans, is the clearest sign that the company wants to sell a review workflow rather than a single screening feature.

Strengths

It keeps the main job narrow. Rayyan is built for screening, deduplication, filtering, and review management, so it does not waste time pretending to be a general-purpose research browser. That focus makes the interface easier to reason about when the real constraint is reviewer throughput, not ideation.

It makes collaboration practical instead of ceremonial. Individual plans include free reviewers and viewers alongside paid seats, which is the right model for review work where one or two people own the subscription and the rest of the team only needs occasional access. Seats can also be reassigned, which matters when projects change hands.

It now covers more of the evidence-synthesis pipeline. PICO extraction, data extraction, duplicate auto-resolution, and mobile access turn Rayyan into more than a screening queue. Those features do not replace careful work, but they do reduce the number of times a team has to export and re-import the same project state just to keep moving.

The product has a real institutional path. Academic, Business, and Enterprise plans add organization-wide reviews, license management, SSO, API access, and dedicated support. That makes Rayyan more credible for libraries, methods teams, and research offices that need a repeatable operating model rather than a one-off subscription.

Weaknesses

The AI can speed work up, but it does not remove judgment. A 2022 BMC study found that Rayyan could be very sensitive for excluding ineligible records at certain thresholds, but that same evaluation showed poor specificity in other settings. That is useful if your job is to clear a large pile faster; it is less impressive if you want the tool to make final inclusion decisions for you.

The pricing is deliberately segmented. Rayyan’s main individual tiers are Essential at $4.99 per seat per month billed annually and Advanced at $8.33 per seat per month billed annually, with quarterly billing available at higher effective rates. Institutional plans push further into sales-led licensing, which is fine for research departments but makes the product less elegant for mixed-size teams.

The privacy policy is serviceable, not especially reassuring. Rayyan says it collects profile, account, payment, usage, and user-content data; it also says user content can be stored, reproduced, published, or made public inside the service. That is normal for collaboration software, but the public policy is old enough that I would want a clearer, current statement on how review data is handled in AI features before using it for sensitive work.

It is still a niche tool, not a broad discovery engine. If your workflow starts with searching for papers, comparing claims, or mapping a field, Rayyan is not the first place to begin. It becomes valuable after the search stage, when the work turns into organized screening and extraction.

Pricing

Rayyan’s pricing makes sense if you think of the product as workflow infrastructure rather than software for casual use. The free tier is not a fake teaser: it gives you 3 active reviews, up to 2 free reviewers or viewers, duplicate detection, and basic workspace tools, which is enough for small projects and for testing whether the workflow fits your team.

Essential is the default paid tier for solo researchers and small groups. At $4.99 per seat per month billed annually, it adds the practical features that matter most in everyday use: more collaboration, mobile access, PRISMA flow support, and duplicate automation. The quarterly option exists, but the annual rate is the one Rayyan is clearly trying to steer people toward.

Advanced is the right tier only if you actually need the extra capacity and AI features. At $8.33 per seat per month billed annually, it adds PICO extraction, a higher active-review limit, and a better fit for busier review workloads. If you will not use those features, it is mostly a more expensive way to get the same core workflow.

The institutional plans tell you who Rayyan is really selling to. Academic and Business are license-based packages for organizations, while Enterprise adds SSO, API access, admin controls, and dedicated support. That structure is sensible for research offices, but it also means the product becomes materially more expensive once you move beyond the individual tier.

Privacy

Rayyan’s privacy policy is conventional for a collaboration platform, but it is not minimalist. The policy says it collects profile, account, payment, usage, and user-content data, and it explicitly covers references, comments, decision labels, messages, and forum activity inside the service. It also says publicly marked content can be accessible by anyone, which is the biggest practical privacy risk for professional users.

The policy also says personal data may be transferred outside the EEA and UK and that standard contractual clauses are used where required. That is helpful, and the GDPR language is a sign that Rayyan understands institutional expectations, but the policy is still written around service operation rather than around a clear modern promise about AI training boundaries. If you are working with sensitive review material, the public-sharing controls deserve careful attention.

Who It’s Best For

The systematic-review lead coordinating a small team. Rayyan is a good fit when one person owns the workspace and several collaborators only need occasional reviewer or viewer access. The seat model, shared reviews, and review chat make that structure workable without forcing everyone into paid seats.

The research office or library team running repeatable evidence synthesis. Academic and Business plans are built for organizations that need central license management, SSO, and a stable operating process across multiple reviews. Rayyan wins here because it is designed around governance, not just individual productivity.

The solo PhD student or postdoc who only runs a few reviews a year. The free tier is genuinely usable, and Essential is cheap enough to justify if you want mobile access, duplicate automation, and a cleaner workspace than ad hoc spreadsheets and shared inboxes.

The methods-heavy team that wants structure more than discovery. If your work is already past the literature search stage and you need screening, extraction, and auditability, Rayyan is a better fit than a broader research assistant.

Who Should Look Elsewhere

People who want a literature discovery engine first should evaluate Elicit or Consensus. Those tools are more natural when the main task is finding and comparing claims, not managing a screening queue.

Researchers building maps of a field should look at ResearchRabbit. Rayyan is built for review operations; ResearchRabbit is built for exploration and networked discovery.

Teams that care about citation checking and evidence tracing may prefer Scite. It is the better fit when the question is how a paper is cited and supported across the literature rather than how to move hundreds of records through a review pipeline.

Bottom Line

Rayyan is one of the few research tools that knows what problem it is solving. It is not trying to be the best place to brainstorm a topic, write a paragraph, or wander through the literature. It is trying to make evidence-synthesis work faster, more collaborative, and more auditable.

That gives it a real place in the market. The product is strongest when screening volume, reviewer coordination, and workflow discipline matter more than open-ended intelligence. If that is your problem, Rayyan is a credible answer. If it is not, the product will feel narrower than its feature list suggests.