Study inputs
Draft study plan or indication + treatment type
Add the protocol, synopsis, assumptions and constraints that are available today.
Study Design and Protocol Optimisation
Estimate whether the proposed design can meet its primary endpoint, then pressure-test recruitment, protocol burden, countries, sites, timelines and costs before the plan becomes an expensive commitment.
See how feasible your proposed study really is.
Bring your indication, treatment type and draft protocol or study plan, if available.
Study inputs
Add the protocol, synopsis, assumptions and constraints that are available today.
Historical evidence guides the agent as it tests the study's connected assumptions.
Estimate P(success) and test recruitment, burden, footprint, duration and cost together.
Human checkpoint
Challenge the evidence, refine assumptions and rule out impractical options.
Decision-ready output
Compare predicted success, feasibility risks and practical next steps.
Real-world constraints
Budget · Timeline · Regulatory requirements · Geography · Patient burden
Probability of success + feasibility
TrialKey estimates the proposed design, explains the model drivers, generates alternatives and tests whether the stronger candidates can be delivered in the real world. The recommendation balances predicted success with historical evidence, study priorities and feasibility constraints.
Example study-plan optimisation
TrialKey compares every alternative with the same reference plan, estimates its probability of meeting the efficacy objective and checks whether the changes are supported by historical trial patterns.
The recommended plan balances modelled probability of success, evidence alignment, study priorities and feasibility constraints—it is not selected on probability alone.
Candidate progression
Predicted probability of success
Optimisation progress
Plan comparison
Eighteen alternative study plans assessed. Historical evidence is used as a separate grounding check alongside the modelled probability of meeting the study objective.
Recommended only after feasibility and expert review
What changed
Generated design changes were compared with 1,200 historical eligibility criteria before the recommended plan was surfaced.
What looks better
The recommended candidate increases the modelled probability of meeting the efficacy objective by 21.4 percentage points while retaining the core disease-confirmation and outcome anchors.
What to check
Confirm that the refined eligibility criteria remain clinically appropriate and do not create a new patient-access or recruitment constraint.
Recommended next step
Review the changed eligibility lines with the clinical and operational subject-matter experts before adopting the candidate as the study-plan recommendation.
Eligibility comparison
Reference plan
Recommended plan
Clinical promise and operational feasibility are not the same thing.
A study can have a promising modelled probability of meeting its endpoint but still be difficult to recruit or deliver. A highly recruitable study can still have weak statistical or clinical prospects. TrialKey keeps both questions visible and ranks alternatives across them.
Model the probability that the proposed design will meet its primary efficacy endpoint, with a confidence range rather than a single unsupported score.
See which study-design variables, evidence signals and assumptions contribute most to the predicted outcome.
Generate and compare design options that may improve predicted success without creating an impractical protocol or operating plan.
Compare probability, recruitment, patient burden, timeline, cost and geographic footprint before selecting a preferred scenario.
Clinical trial feasibility assessment
Clinical trial feasibility depends on the interaction between protocol design, patient availability, recruitment assumptions, site capability, country selection, timelines and budget. TrialKey evaluates those decisions together so one optimistic assumption does not undermine the entire study plan.
Test whether the proposed population and eligibility criteria leave a credible pool of recruitable participants.
Benchmark recruitment rates, screen-failure assumptions and enrolment duration against relevant completed and competing studies.
Identify visits, assessments, endpoints and operational requirements that may create avoidable patient or site burden.
Compare patient access, investigators, competing trials, regulatory conditions and operating complexity across countries.
Evaluate whether the proposed footprint can support activation, recruitment, data quality and delivery expectations.
Pressure-test whether start-up, recruitment, treatment and follow-up assumptions fit the programme's time and capital constraints.
Clinical trial delay evidence
Historical trial outcomes show that delayed completion is common—and that delay rises sharply as protocols accumulate amendments. TrialKey makes those risks visible while study alternatives can still be tested.
Completion delay is measured by comparing the date a trial registered as its planned completion with the date it actually finished.
median completion overrun
finish more than a year late
finish more than two years late
Amendment burden
27 months late
Median delay for trials with 11 or more amendments, versus about 4 months for trials with one or two.
Approximately 6.7× longerProtocol amendments
Median delay
1–2 amendments
4.0 mo
3–5 amendments
7.4 mo
6–10 amendments
14.1 mo
11+ amendments
26.6 mo
How it works
TrialKey combines historical clinical-trial evidence, protocol features, competing-study activity and operational assumptions to identify where the proposed plan is credible—and where it needs refinement.
Start with a protocol, synopsis or draft study plan—or with the indication, treatment type and development concept when the design is still taking shape.
Compare eligibility, enrolment, sites, countries, visits, duration and other feasibility drivers with relevant historical and competing trials.
Estimate the proposed design's probability of success, then test how changes to the population, endpoints, burden, footprint and timeline affect both clinical prospects and execution risk.
The system compares evidence-supported options, then clinical and operational experts challenge the results and convert the strongest feasible scenario into recommendations a team can defend.
From protocol to operations
A credible feasibility plan separates evidence-supported assumptions from numbers that are merely convenient for the business case. TrialKey shows where the proposed protocol and operating model are most exposed.
Assess whether eligibility criteria, visit schedules, procedures, endpoints and follow-up requirements make participation or delivery unnecessarily difficult.
Compare recruitment rates, screen failures, activation timelines and study duration with evidence from relevant trials rather than relying on a single-point forecast.
Identify where patient access, investigators, competing studies and operational capability are most likely to support high-quality enrolment.
Feasibility outputs
Clinical trial feasibility should explain why the plan is credible, where uncertainty remains and what can be changed. TrialKey produces a traceable assessment that clinical, operational, regulatory, finance and governance teams can review.
Where feasibility fits
TrialKey uses feasibility to turn a proposed design into realistic operating assumptions, then carries those assumptions into country and site selection.
Product capability
Develop and compare clinically defensible study designs.
Explore clinical trial designFeasibility layer
Test whether the design can recruit and operate within real constraints.
Next product capability
Identify the countries and sites most likely to deliver the feasible plan.
Explore country and site selectionQuestions
Clinical trial feasibility is the assessment of whether a proposed study can recruit the required participants, operate across its intended countries and sites, meet its timeline and remain deliverable within clinical, regulatory and budget constraints.
A study feasibility assessment tests the assumptions behind a clinical study before execution. It can examine patient availability, eligibility criteria, recruitment rates, protocol burden, country and site capacity, competing trials, timelines and costs.
Protocol feasibility asks whether the study design itself is practical for patients, investigators and sites. Site feasibility asks whether specific research sites have the participants, capabilities, capacity and operating conditions needed to deliver that protocol.
Yes. TrialKey can begin with an indication, treatment type, development phase and draft study concept. A protocol, synopsis or study plan can be incorporated when available, but it is not required to begin an early feasibility assessment.
AI and machine learning can compare the proposed study with historical and competing trials, identify influential feasibility drivers and test alternative assumptions. TrialKey combines this analysis with expert review rather than relying on an automated score alone.
They answer connected but different questions. Probability of success estimates how likely a study design is to meet its primary efficacy endpoint. Clinical trial feasibility tests whether that design can recruit, operate and finish within real-world constraints. TrialKey evaluates both so a clinically promising design is not recommended if it is impractical to deliver.
No feasibility assessment can guarantee recruitment or delivery. TrialKey provides evidence-based ranges, identifies material risks and compares alternatives so teams can make more defensible planning decisions before committing resources.
Higher protocol-amendment counts are associated with longer clinical-trial completion delays in TrialKey's readiness analysis. Amendments can add review cycles, site retraining, document and system changes, resupply work and recruitment disruption. TrialKey uses historical patterns as evidence—not as proof of causation—to identify design pressure points and compare alternatives before execution.
Evaluate study feasibility
Bring us your indication, treatment and proposed study assumptions. We will show how TrialKey can structure the evidence, expose the material feasibility risks and compare practical alternatives.