How it works

Let the clear nos leave the queue.

Upload a finished search and your inclusion criteria. SyRev calibrates a conservative screen, keeps the uncertain middle with humans, and returns the evidence you need to explain what happened.

Five steps from search to proof

SyRev is designed to complement your existing review workflow, not hide decisions behind a score.

01 · IMPORT

Bring your search

Upload RIS, PubMed .nbib, EndNote XML or CSV, up to 10 files at once, or run a built-in search of PubMed, OpenAlex or Europe PMC. Clear duplicates merge automatically across files before screening; possible matches go to a check screen where you merge or keep both. Each imported file is logged in a PRISMA-S search log New.

02 · DEFINE

Set the criteria

Provide the inclusion and exclusion rules that make a record relevant to your review.

03 · SCREEN

Calibrate the clear nos

Each record is scored against your criteria. Only conservative, low-probability exclusions leave automatically.

04 · REVIEW

Own the middle

Humans review uncertain records and candidate includes with per-criterion confidence and keyboard-friendly decisions. Includes then move to full-text screening New, free on every plan, with a structured reason for every exclusion. At both stages you can undo or change a vote, add notes and labels, and highlight your include and exclude keywords.

05 · EXPORT

Keep the proof

Get counts, a validation pack, a PRISMA 2020 diagram (SVG/PNG/PDF), RIS export (included records, or all records with decisions), a methods paragraph, and an append-only audit export that makes the replay explainable.

Then extract the data New Manual extraction is free on every plan, including the free evaluations. With SyRev Plus, AI pre-fill and an AI second extractor cover up to 500 studies per review, with an exact quote and page number on every suggested answer for you to check. Export CSV, Excel, a blank form or a RevMan Web pack. See data extraction
Then assess risk of bias New Risk of bias and quality appraisal, free on every plan, including the free evaluations: RoB 1, AMSTAR 2, SYRCLE, the NHLBI study quality assessment tools and more, with traffic-light and summary plots, tables and a methods paragraph that names the tool and version. See risk of bias

Choose how deep

Start with two free evaluations on a review you have already completed.

Replay

Title/abstract screening at Careful. No PDFs required; ideal for a fast first replay.

SyRev Plus

Everything in Replay, plus AI full-text screening on the uncertain middle and candidate includes. Open-access PDFs are fetched when records have a DOI, PMID or PMCID; you upload the rest.

Careful by default

Careful protects recall: 96.5% of relevant studies kept across 310 public review datasets, with individual reviews sometimes lower. See full results

Honest by design: savings vary by review. Gold labels are often full-text decisions while Replay scores title/abstract only, so perfect recall is not guaranteed. SyRev shows the result on your review instead of promising a headline number.

Ready to see your numbers?

Create an account, upload a finished search and criteria, and use your two free evaluations to see your own numbers.

Create account · 2 free title/abstract evaluations See pricing