Clinical Trial Feasibility: Rethinking Site Selection

By Maria Marques, Senior Director of Clinical Solutions

Clinical trial feasibility is one of the earliest stages of clinical development. It's also one of the phases that teams tend to underestimate.

Throughout my career in clinical research, I've built research departments, set up ethics committees, and planned the systems for clinical trials.

Industry benchmarks show that feasibility assessment takes an average of 29 days, followed by another 31 days for site qualification. That's nearly two months before the first patient is enrolled, assuming the right sites were selected the first time.

After years of building research operations from the ground up, these three lessons still influence my perspective on feasibility:

Lesson 1: Good questionnaires don't always lead to good decisions

Feasibility questionnaires remain a cornerstone of clinical trial site selection. These tools help sponsors understand experience, patient populations, and operational capabilities.

But questionnaires only capture a moment in time.

Sites and investigators complete them in good faith, using the information available at that moment. The limitation isn't the questionnaire itself. The challenge is that the data is often historical, sometimes optimistic, and rarely reflects the specific demands of the protocol or how a site's capacity may change by the time the study begins.

As a result, site selection decisions are often based on incomplete information rather than a forward-looking view of site readiness.

Predictive feasibility offers an opportunity to move beyond historical questionnaires by combining live operational data for more informed site selection decisions.

Lesson 2: Delays often begin before recruitment

When a study falls behind, recruitment is usually the first area to receive attention. In my experience, recruitment often reveals problems rather than creating them.

The earliest warning signs tend to appear only after the study is underway. An investigator who was enthusiastic during selection becomes harder to reach. Screen failure rates begin to rise. A study coordinator is suddenly supporting more trials than anticipated.

By then, the feasibility process is already over. Contracts have been signed, staff have been trained, and supplies have been shipped.

Changing direction at that stage creates rework, delays execution, and can threaten the entire programme.

That's why I see site feasibility in clinical trials as one of the most important opportunities to reduce downstream risk. Better decisions made before enrollment can prevent months of operational delays later.

Lesson 3: AI should improve judgement, not add complexity

Technology has transformed many aspects of clinical research. Remote visits, eConsent, and decentralized trial elements have expanded access and improved the patient experience.

Operational workflows, however, haven't evolved at the same pace. Digital tools have often been layered onto existing workflows instead of redesigning them.

Sponsors, CROs, and sites often operate using different systems and different operational languages, forcing teams to recreate the same feasibility assessments every time a new study begins.

This is where I believe artificial intelligence can make the greatest impact. AI shouldn't become another tool layered onto an already fragmented process. It should help redesign how feasibility decisions are made by bringing together operational performance, investigator experience, patient availability, and competing trial activity into a more complete picture of site readiness.

On the workflow side, I believe the manual feasibility process as we know it will largely disappear. It has to. Predictive feasibility, drawing on live data rather than historical surveys, will become the baseline expectation.

Why Clinical Trial Feasibility Matters More Than Ever

Clinical trial site feasibility shouldn't be viewed as an administrative requirement. It's one of the earliest and most important decision points in clinical development.

The better those decisions become, the more efficiently studies will run, and the sooner patients can benefit from the treatments they're waiting for.