Run reviews like releases.
Roles, comments, two-person sign-off, and an audit trail—all built into the same end-to-end review platform. Designed for HEOR consultancies, HTA agencies, guideline committees, and Cochrane-style review teams.
Six capabilities. One coherent team product.
Roles & permissions, enforced server-side.
Four roles — admin, lead reviewer, reviewer, viewer — are enforced server-side on every action. Private projects are invisible to anyone outside the team — including search engines and people who guess at URLs.
- Admin can invite, promote, delete, and approve sign-offs.
- Lead reviewer edits protocol, assigns workload, adjudicates conflicts.
- Reviewer submits screening decisions and extraction reviews — read-only on protocol.
- Viewer is read-only across the team — useful for stakeholders, methodologists, and auditors.
Comments & @mentions on every object.
Threads anchor on analyses, studies, individual screening decisions, extracted outcomes, and extraction reviews. You can only @mention people on the same review, so a stray name from another team never sends a notification by accident.
- One level of replies per thread, collapsed for readability.
- Authors edit their own comments; admins and lead reviewers can soft-delete inappropriate content while preserving audit history.
- Resolve threads when the discussion is finished; re-open any time.
- Read and write access gated by the same team-role permissions as the parent analysis.
Unified notifications & email digest.
Every team event lands in one bell-icon inbox: @mentions in comments, invitations, adjudication pings, sign-off requests, and the living-review digest. One pipeline, one set of preferences.
- Per-user preferences route each kind to immediate email, digest email, or in-app only.
- Digest cadence is daily, weekly, or off — picked per notification kind.
- Safe defaults: invitations and adjudication requests send immediately; informational events are opt-in.
- Notification access is strictly per-user — you only ever see and act on your own.
Two-person protocol sign-off.
Five strict milestones — protocol locked → screening complete → extraction complete → analysis frozen → manuscript locked — let Cochrane- and HTA-style reviews enforce each stage before the next opens. The two-person rule is built in: whoever requests sign-off can never be the same person who approves it.
- A lead reviewer requests sign-off with a milestone and optional notes.
- The system freezes the protocol, eligibility, and extraction form as an immutable snapshot.
- Every admin on the team is notified; one admin approves or rejects.
- Once locked, edits to that stage are blocked until an admin reverts the milestone.
Workload assignments & conflict adjudication.
Split studies across reviewers per phase (abstract / full text) and slot (A / B). Auto-balance round-robins deterministically — refresh the page and you see the same preview before you commit.
- Per-reviewer “my queue” view stays fast even on 50,000-study reviews.
- Pick the review mode per analysis: single, dual AI + me, or dual two humans.
- Per-study peer-status badges — Agreed, Awaiting peer, or Conflict — visible without opening the study.
- Conflict adjudication shows a field-by-field diff between reviewer submissions.
Versioned protocols & audit-ready exports.
Every sign-off snapshots the analysis state into an immutable version. Every signed-off version is listed in chronological order, and you can diff any two consecutive versions side-by-side to see exactly what changed in the protocol, eligibility, or extraction form.
- PRISMA flow updates automatically as studies move through the pipeline.
- Source-linked audit trail: every effect estimate traces back to the originating PDF snippet.
- Sign-off events, role changes, and adjudication decisions are all logged.
- Sign in with Google or Microsoft — no extra account or password to manage.
From kickoff to manuscript, on a single timeline.
The same five milestones every Cochrane- and HTA-style review already uses, but enforced by the platform instead of a shared spreadsheet. Every milestone snapshots the protocol so a regulator can replay the review months later from the same inputs.
Protocol locked
Lead reviewer drafts PICO/PEO, eligibility, and extraction form. An admin signs off and the protocol becomes immutable.
Screening complete
Workload assignments split records across reviewers. Conflicts route to the adjudication queue with a field-by-field diff.
Extraction complete
Per-outcome accept / reject decisions instead of batch approval. Reviewers leave threaded comments anchored to each outcome.
Analysis frozen
Forest, funnel, and sensitivity plots are saved with the protocol version that produced them. The deterministic engine reruns same-input → same-output.
Manuscript locked
PRISMA flow, study tables, and pinned narrative are exported to Word. RIS/BibTeX/RevMan 5 round-trips preserve every identifier.
Bring your team. Ship your next review.
Free tier runs the full pipeline on one real review. Team plans start at 3 seats with 7 pooled reviews per seat per month.