The short answer: when choosing a contract research organization, the single most important factor is enrollment capability in your specific indication, because almost every clinical program that runs late runs late for one reason: the patients did not arrive on schedule. Weight enrollment heavily, weight the bid price lightly, and ask every candidate the one question almost nobody asks: what was the final contract value versus the original bid on their last five comparable studies.
You are not really choosing a CRO. You are choosing an enrollment forecast.
Why enrollment is the whole game
Data quality, monitoring, and regulatory support all matter, but they are rarely what sinks a timeline. Enrollment is. When patients arrive slower than the plan assumed, everything downstream slips with it, and the cost overruns follow.
Yet most selection processes treat enrollment as one line item among fifteen and spend most of their energy negotiating the rate card. Then, eighteen months later, the study is behind and the change orders have arrived, and everyone is surprised.
The four CRO delivery models
Choose the model before you choose the vendor, because it changes who you should be asking.
- Full service. The CRO runs the trial end to end. Right for companies without a large internal clinical operations team.
- Functional service provision. You buy specific functions and run the study yourself. Right for companies with a strong internal team who want control and lower cost.
- Hybrid. Full service for the study, with one or two functions such as biostatistics retained. Increasingly the default for mid sized companies.
- The specialist. A CRO focused on one therapeutic area or modality. Right for difficult indications where finding patients is the risk.
The criteria that predict success
Score every candidate on the same sheet:
- Enrollment capability in your indication (30%). Their actual enrollment performance in studies like yours, with the numbers.
- Therapeutic area depth of the actual team (20%). The experience of the people assigned to you, not the company’s overall record.
- Project management quality and stability (15%). Who your project manager is and how many studies they carry.
- Change order behavior (15%). The gap between their original bids and final invoices on past studies.
- Data and technology (10%). Who owns the data and how quickly you can access it.
- Quality and regulatory track record (5%).
- Price (5%). The bid is a marketing document. The final invoice is the price.
The change order trap
This is the part of the CRO relationship that damages trust most, and it is predictable. A competitive bid is priced to win, which means it is built on optimistic assumptions, and the easiest assumption to make optimistic is enrollment. An optimistic enrollment assumption produces a cheaper bid and a more expensive study.
The defense is to ask, before you sign, for the original and final contract values on their last five comparable studies. A CRO with a disciplined record will answer and be pleased to. A CRO with a bad record will explain why the question is unfair. That reaction is the answer.
Questions to ask before you sign
- Show me your enrollment performance on your last three studies in this indication, against plan.
- Name my project manager and lead monitor, and how many other studies each carries.
- On your last five comparable studies, what was the original contract value and what was the final one?
- Which of my eligibility criteria do you think will cause the most screen failures?
- Who owns the data, and can I see it in real time?
Frequently asked questions
What is the most important factor when selecting a CRO?
Enrollment capability in your specific indication, backed by real performance data from comparable studies. Because most delayed trials are delayed by slow enrollment, this predicts success better than price or overall company reputation.
What is the difference between full service and functional service provision?
In a full service model, the CRO runs the entire trial and you oversee it. In functional service provision, you run the trial and buy specific functions such as monitoring or data management, which gives you more control at a lower unit cost.
Why do CRO costs exceed the original bid?
Competitive bids are priced to win, often on optimistic enrollment assumptions. When enrollment runs slower than assumed, the study takes longer and costs more, and protocol amendments trigger change orders. Asking for the bid to final invoice history of past studies helps predict this.
How do you evaluate a CRO’s enrollment capability?
Ask for actual enrollment curves from their last several studies in your indication, against the original plan, by month and by site. A vendor that cannot show this is offering a hope rather than a forecast.
Get the full evaluation scorecard
We built a complete guide with the four models compared, a weighted scorecard you can print and fill in, all 26 questions to ask, the change order defense, seven red flags, and the governance numbers to track once the study starts.
