Site selection decisions made on outdated feasibility data leave trials carrying sites that enroll slowly or not at all.
A recurring pattern in trial operations is sites that looked strong during feasibility but enroll far below plan — or never randomize a single patient — while the trial keeps paying for monitoring, startup, and management overhead. The root cause is usually site selection based on questionnaires and stale historical impressions rather than current patient-population data. AI-driven site selection uses real-world data on where eligible patients are actually treated, plus site-level performance history, to rank candidate sites and flag enrollment risk before contracts are signed.
How AI improves clinical trial site selection
Data-driven site selection replaces the two weakest inputs in traditional feasibility — self-reported questionnaires and dated impressions — with observable evidence: real-world data showing where patients matching your eligibility criteria are actually treated today, and site-level history showing how candidate sites enrolled in comparable past trials. Ranking candidate sites against both dimensions flags the classic failure mode early: an enthusiastic site whose current patient population cannot support its enrollment promise. The same analysis also surfaces non-obvious candidates — sites with strong eligible-patient volume that were not on the usual list — which is where the approach tends to add the most value over experience alone.
What to evaluate before buying site selection AI
Coverage is the first question: the platform's data must actually reach your target geographies, and real-world data density varies sharply by country. Probe recency (patient populations shift; last year's density map may mislead) and whether site performance metrics account for trial complexity — a site that enrolled well in a simple study may struggle with a demanding protocol. Insist on transparent rankings: you need to see why a site scored as it did to defend selections to study teams, CRO partners, and — when enrollment is later questioned — to your own leadership.
How teams typically get started
The cleanest evaluation is a backtest: give the platform the protocol and candidate site list from a completed trial and compare its rankings against actual site-by-site enrollment. Every sponsor has at least one study whose site-level results they know painfully well — that study is your ground truth.
AI Use Cases That Address This Problem
Patient Recruitment & Enrollment
Site Selection & Performance
Frequently asked questions
Why do sites that pass feasibility still underperform?
Feasibility questionnaires capture intention, not patient flow: sites answer optimistically, competing trials draw from the same population, and eligible-patient estimates are often based on impressions rather than data. Selection informed by current real-world treatment data addresses the estimate; competing-trial dynamics and startup execution still need operational management.
Can AI predict how many patients a site will enroll?
Platforms produce enrollment estimates with meaningful uncertainty, grounded in patient-population data and site history. They are consistently more informative than questionnaire self-reports, but they are estimates — plan contingency around them rather than treating any per-site number as a commitment.
What data goes into AI site rankings?
Typically de-identified claims and EHR data locating patients who match the protocol's eligibility criteria, combined with site-level indicators such as past enrollment performance, startup timelines, and data quality history. Ask each vendor exactly which sources they use and how current each one is in your geographies.
Does this replace CRO feasibility work?
It sharpens rather than replaces it. Data-driven rankings narrow and reorder the candidate list, but qualification visits, contract negotiation, and relationship management remain human work — and many CROs now use these same platforms internally, which is worth asking about to avoid paying twice.
Launched the Verana Research Network on the AAO IRIS Registry to accelerate clinical research
Verana Health and the American Academy of Ophthalmology launched the Verana Research Network, an IRIS Registry initiative in which academic medical centers and ophthalmology practices use IRIS Registry data and Verana's platform to increase clinical trial access, accelerate recruitment, and advance ophthalmic research.