Sponsors negotiate CRO rates. CRO budgets are not driven by rates.
They are driven by assumptions — about how many sites you will need, how fast they will enroll, how many visits will happen, how long the study will run. The rate card is what gets argued over in contracting. The assumptions are what determine the invoice eighteen months later.
The three components
Almost every CRO budget resolves into direct fees, pass-through costs and investigator grants.
Direct fees are the CRO's labor, usually priced by unit — per site activated, per monitoring visit, per data point, per month of project management. This is the negotiable portion and the smallest source of surprise.
Pass-through costs are third-party expenses the CRO administers: travel, central lab, IRT, couriers, translations, archiving. These are estimated rather than fixed and they move with the study.
Investigator grants are payments to sites, built from a per-patient budget times a projected enrollment. They are frequently the largest single line and the least examined at contracting.
Where the budget actually moves
Four assumptions carry most of the variance.
Enrollment rate. Nearly every driver is downstream of this. Enroll at half the projected rate and you extend duration, add sites, add monitoring visits and add project management months, all at once.
Site count. More sites means more activation, more monitoring, more contracting and more grant commitments — and the added sites are usually the lower-enrolling ones.
Study duration. Project management, data management and vendor fees are largely time-based. Duration is the multiplier sitting underneath everything else.
Monitoring intensity. A plan built on 100% source data verification and one built on risk-based monitoring differ enormously, and the difference compounds with site count and duration.
Why the optimistic bid wins
A CRO that models enrollment optimistically produces a lower bid. It also produces the change order.
This is not usually deception. Bid grids are built from the information the sponsor supplied, and sponsors supply optimistic enrollment projections because their own plans depend on them. The optimism is shared and it is priced into nobody's expectations.
The correction is to compare bids on their assumptions rather than their totals. Normalize every proposal to the same enrollment curve, site count and duration, and see what happens to the ranking. It frequently changes.
Questions that hold a budget
What enrollment rate per site per month does this budget assume, and what is that based on?
Which lines move if enrollment runs at 60% of plan, and by roughly how much?
What triggers a change order, and what is the process and approval threshold?
Which costs are fixed, which are unit-based, and which are estimates?
What is the assumed monitoring model, and what would move it?
Change orders are not the enemy
A study that runs eighteen months longer than planned should cost more. The problem is not that change orders exist; it is that they arrive as surprises, unpriced, with no agreed process, in the middle of a delivery crisis.
Agree the mechanism at contracting: what constitutes a change, who raises it, what evidence supports it, who approves it, and within what timeframe. A change order process negotiated in advance is an operational tool. One negotiated in month fourteen is a dispute.
Benchmarking, and its limits
Sponsors often ask how to benchmark CRO pricing. Commercial benchmarking data exists and is useful for unit rates, but it is close to useless for total cost, because total cost is a function of your assumptions rather than their rates.
A more reliable comparison is internal. Ask three CROs to bid against an identical, explicitly stated set of assumptions — same enrollment curve, same site count, same duration, same monitoring model — and the spread tells you something real. Where a bid diverges sharply, ask why. The explanation is usually either a different view of the risk, which is worth hearing, or a different reading of the scope, which needs fixing before award.
What good looks like
A sponsor who can say, at any point, which assumptions the budget rests on, which of them are currently holding, and what the exposure is if they do not. That is a governance artifact rather than a finance one — and it belongs in the same forum where enrollment and risk are discussed, not in a separate conversation with procurement.
A condensed answer to this sits in the FAQ under How are CRO budgets structured?.
