Risk-based monitoring, run continuously

Your sites don't wait for the next visit. Now your monitoring doesn't either.

Rigora runs source data verification and source data review continuously— reconciling each site's source against the eCRF and reviewing it for protocol deviations as the data lands, not in a scramble the week before the next visit. It flags the discrepancies, drafts the queries, and drafts the visit report. Rigora drafts; your CRA signs.

So your CRA walks in knowing where to look — and signs the report instead of writing it at 11pm.

Built on drug-trial monitoring; the same engine works for device programs.A sponsor? Bring risk-based monitoring to your CROs →

Risk-based monitoring, done — ICH E6(R3) asks CROs to focus monitoring on what carries risk; Rigora is how you actually do it
No EMR seat — works from certified copies, lab feeds, and sponsor-mediated exports; never a site login
A command center for leads— portfolio risk across every site in one view, so lead CRAs see where trouble is building and steer their team's time before the next visit
Weeks, not quarters — our deployment design target is four weeks, validated together on your trial; no integration project to start

Shows its work

Every draft shows its inputs. Every value carries its provenance.

This is the actual artifact — the same components our product renders, with synthetic data.

No competitor demos accountability; they demo magic. Rigora is built the other way around.

A certified copy is not the same claim as an EMR view — so every source value carries its provenance class, and an attestation, once made, is frozen. Corrections create new versions.

Reviewers see the exact structured data sent to the model beside every draft. Where the model lacks confirmation, it says so — drafting stops being “magic” the moment you see it.

And every state-changing action lands in an append-only audit trail attributed to the person who took it.

Read an annotated visit report →
Visit VST-014 · Site-014 · Interim monitoringPT-0142 · synthetic

Source data verification

ANC (×10⁹/L)Src1.8 eCRF1.8MatchLab rpt p.2
Creatinine (mg/dL)Src0.9 eCRF1.9FlagLab rpt p.1

Source provenance

Certified copyFHIR feedEMR view

Inputs to AI2 SDV findings · 1 protocol deviation · visit metadata. Structured data only — the CRA sees exactly what the model sees.
Drafted outputCreatinine in the eCRF (1.9 mg/dL) does not match the certified lab report (0.9 mg/dL); a query has been drafted for site resolution.AI-generated content. Review recommended.
2026-07-06T14:02:11Zcra.demosdv_finding.disposition_set → flag
2026-07-06T14:03:40Zsystemmvr.draft_generated (model recorded on draft)
2026-07-06T14:11:23Zcra.demomvr.signed (typed name)

Sits above your stack

Reads source and EDC. Writes to neither.

No EDC integration project, no site IT burden, no new system of record to validate.

Reads, never writes

Rigora reads source documents and your EDC. It writes to neither — read-only is an architectural commitment, enforced at the credential level.

Your CRA raises the query

Queries are drafted here and raised by your CRA in your EDC — Medidata Rave, Veeva Vault, Castor, OpenClinica. Your workflow, faster.

Your systems stay the record

The signed record of record stays in your validated eTMF/QMS. We export audit-grade evidence of the human review; we don't replace your systems.

The platform

Start with zero integration. Connect as your trials warrant it.

Tier 1 · Start here

Continuous monitoring, zero integration.

Works from what your CRAs already have. Between visits, Rigora runs source data verification and source data review continuously — reconciling source against the eCRF, reviewing it for protocol deviations, and triaging what deserves on-site time — all with full provenance. It drafts the queries and the visit report (MVR) your CRA signs. No site IT project, no EDC integration required; deployment design target measured in weeks, not quarters.

Tier 2 · Expand — as your trials warrant it

Connected source, one feed at a time.

Central-lab and ECG feeds reconciled automatically — “the eCRF says ‘ECG performed’ — but no result in the central feed. Flagged.” EHR data via FHIR where a site supports it, sponsor-mediated exports where it doesn't. Each connection deepens coverage; none is required to start.

The economics

Monitoring is the biggest line you can still move.

Every figure sourced — methodology and citations here.

$10–15M
typical SMB biotech Phase 2 CRO contract
Sertkaya et al., 2016
$2.4M
median monitoring line per Phase 2 trial
Industry benchmark · 15–22% of contract
~70%
of CRA monitoring labor is templated work - SDV, MVR drafting, queries, oversight
Rigora analysis · Sertkaya + Tufts CSDD
$700K
per Phase 2 trial replaceable by AI + human-in-the-loop
Andersen 2015 · Abbasi 2025 · ICH E6(R3)

The evidence

10,101 eCRFs re-verified. Zero conclusions changed.

“Zero errors that changed any result or conclusion.”

A 2025 retrospective study in Communications Medicine (Nature portfolio) performed source data verification across 10,101 eCRFs from the I-SPY COVID platform trial (NCT04488081) and found zero errors that changed any conclusion. The trial used centralized monitoring by design.

Abbasi et al., Communications Medicine 5, 444 (2025) · DOI: 10.1038/s43856-025-01126-9

If 100% SDV across 10,000+ fields finds zero consequential errors, the marginal value of verifying every field on site is near zero. Risk-based monitoring focuses that time instead — Rigora productizes exactly that.

Design partners

Monitoring that shows its work.

We are onboarding a small number of design-partner CROs. Tell us about your monitoring load; we will show you the product on synthetic data.

Synthetic demo data — no PHI on this site·Typed-name attestations, not cryptographic signatures·Append-only audit trail·Read-only toward your EDC and EHR