Digital Recruitment for Clinical Trials: What Works, What Doesn't, and Why

Every recruitment vendor pitches digital as the answer. The promises are consistent: more patients, faster enrollment, precise condition-level targeting. What they don't explain is what the numbers should actually look like, or how to tell whether a campaign is working.

This guide is written for research sites evaluating or running a digital recruitment campaign. Not how to buy ads (that's the vendor's job). How to read the funnel, benchmark the numbers, and know whether the math is moving in the right direction.

What Digital Recruitment Can and Cannot Do

Digital recruitment has one genuine advantage over passive channels: reach. Your internal database degrades at roughly 25% per year. Physician referrals depend on how much time the referring physician has that week. Word of mouth is noise. Digital campaigns reach patients in condition-specific environments (search results, social feeds, health content sites) where they're already thinking about their condition.

That reach is real. Studies using digital-first recruitment strategies have reported over 150% more prequalified inquiries compared to traditional approaches, with enrollment timelines shortened by four or more months in favorable cases.

What digital cannot do is override a bad patient-population match, an eligibility profile that disqualifies most patients who express interest, or slow site-side follow-up. Most campaigns that underperform fail at one of these three constraints, none of which the vendor controls.

The distinction matters because sites often evaluate digital recruitment by the vendor's output (leads, impressions, form completions) rather than by what the site delivers on the back end. The campaign can be running exactly as designed while enrollment stalls, because the bottleneck is at the site, not the ad. Understanding which part of the funnel is leaking is where the useful work happens.

How a Digital Recruitment Funnel Actually Works

A digital recruitment funnel for clinical trials moves through five stages. Each stage has a conversion rate. The product of all five is your cost per randomized patient.

The two places funnels leak most often: stage 2 (pre-screen form drop-off, often a landing page problem) and stage 4 (slow site response). Both are fixable. Stage 4 is always the site's responsibility.

What Cost-Per-Patient Math Actually Looks Like

The number that matters is cost per randomized patient (CPP). Not cost per lead, not cost per pre-screen completion: CPP. Everything else is a step toward it.

Here is what CPP looks like from Clinical Enroll's published case study data:

The lowest recorded CPP across Clinical Enroll campaigns is $577. The published average across the documented case study set is $1,167.

What drives CPP in either direction: indication prevalence (rare disease costs more per patient), eligibility complexity (more restrictive criteria means more spend per conversion), geography (competitive markets and rural areas both raise CPP), and site response speed (slow follow-up directly raises CPP by reducing stage 4 conversion).

Put these numbers against site revenue to understand the ROI math. Phase III trials pay research sites between $5,000 and $20,000 per randomized patient in grant payments, with a Journal of Clinical Oncology analysis finding an average of approximately $6,100 per patient. Phase II trials run higher: $15,000 to $40,000 per patient (ProRelix Research, Sofpromed).

At $2,500 CPP against a $6,100 per-patient site fee, recruitment cost is roughly 40% of the per-patient grant. At $1,167 CPP, it's about 19%. At $577, it's under 10%.

Sites that track CPP as their primary recruitment metric can evaluate any vendor or channel against the same standard. Sites that track leads or impressions cannot.

What Each Digital Channel Actually Does

Not all digital channels produce the same type of patient, and not all indications perform the same way across channels.

The Two Levers Sites Control

Digital recruitment vendors control the campaign. Sites control two things that determine whether the campaign produces results: response speed and pre-screening quality.

The guide on clinical trial enrollment challenges covers the broader structural picture, including why pre-screening and response speed matter across all recruitment channels. The guide on feasibility assessment covers the upstream step: knowing whether the patient population supports the campaign before it starts.

Three Questions to Ask Before Signing with Any Recruitment Vendor

Most recruitment proposals describe a campaign. What they don't describe is an outcome. Before signing anything, a site should have clear answers to three questions.

1. 1

   What specific number of randomized patients are you committing to in this agreement?

   If the answer is a range, a projection, or a "we'll do our best," the vendor is not committing to an outcome. They're committing to activity. The contract should specify a randomization number. That number is what accountability is built around.

2. 2

   What is your projected CPP for this indication and geography?

   A vendor that has run campaigns in comparable indications should be able to give you a CPP range based on actual historical performance. If they project leads or clicks but not CPP, they may not be tracking the metric that matters, or they don't want to be held to it.

3. 3

   What happens if you don't hit the number?

   The answer reveals the structure of the relationship. If the vendor has no position on what happens when targets are missed, the financial risk sits entirely with the site. A vendor with a genuine commitment to outcomes has an answer: a refund, an extended campaign, additional spend, or a contractual make-good provision.

These three questions filter for accountability before a campaign starts. Vendors that answer all three clearly are structurally different from vendors that don't. The difference shows up in enrollment results.