Why CROs Are Slow, Part 2: The Economics of Billable Hours

Your CRO isn’t slow by accident; they are slow by design. In the economics of billable hours, every day your project sits in a queue is a win for their bottom line.

In Part 1, I deconstructed the operating model of the traditional CRO. We saw how manual processes and fragmented handoffs create physical bottlenecks, and how utilization, the percentage of time resources are actively working, acts as a major driver of delay. The math is brutal: as utilization climbs, the time work spends waiting in a queue increases exponentially.

This presents a massive paradox. Every CRO markets themselves on speed—from marketing copy to the names of the companies themselves. (Dash is no exception!) We all know that even small gains in turnaround time offer a massive competitive advantage to sponsors. So, why do CROs push so hard to maximize the very utilization numbers that sabotage their speed?

The answer isn't found in the lab; it's found in the ledger. While the operating model dictates how a CRO works, the business model dictates why they work that way. In this post, we’ll explore how traditional revenue structures are fundamentally incentivized to favor high utilization over your deadlines.

The Economics of Cost-Plus

Most CROs operate in a “cost plus” model, where revenue scales with utilization. The hard costs of running work (reagents, consumables, etc.) are passed through, people’s time is billed hourly or daily to cover salary, and a markup is added on top to cover overhead and profit. While these billable hours are often dressed up as deliverables and other milestones, underneath the hood you are being charged for hours. The moment something unexpected happens, a change order is issued, revealing the truth.

In a cost-plus model, profit per unit time depends on keeping these hourly billable resources engaged. We can see how this plays out by writing out the math for this model. Revenue per time unit is the product of bill rate, b, and utilization, u. If utilization is zero, revenue is zero; at 100% utilization, revenue equals the full billable rate:

If we call the cost of a resource per unit time r, and the markup percentage of that cost that we’re applying, m, then billable rate can be rewritten as the cost plus a markup:

So profit per unit time can be written as:

This form makes the incentives explicit. To maximize profit per unit time in a cost-plus model, a CRO has two levers: markup (m) and utilization (u).

Let’s consider m first. In practice, markup is constrained by market position and the fundamental competitiveness of the market a a particular CRO operates in. A CRO with a strong brand, long track record, and/or unique specialization can command higher rates; a newer or less-differentiated CRO simply can’t be competitive this way and must charge a lower rate. As with other consulting businesses, markup is not a freely-adjustable lever: it is shaped by reputation, experience, and competitive pressure.

That leaves utilization (u) as the more immediate and controllable variable. Here’s what happens as we increase u, keeping m steady. (I’m showing m=50% as an example, a markup that isn’t unheard of):

You can see that as utilization increases, so do profits. Thus in order to maximize profit margin, utilization must be pushed as high as possible. And since wait time is a function of utilization, we can plot profit margin as a function of wait time as well:

This figure explains our current dynamics in the market.

For a CRO using a cost-plus model, maximizing utilization is the rational objective. The more fully resources are utilized, the greater the fraction of billable hours that can be realized and the greater the profits. But as we discovered from Kingman's formula in part 1, the greater the utilization, the more you wait.

From the sponsor’s perspective, the dynamic is reinforced. Sponsors seeking cost-effective outsourcing gravitate towards CROs with more modest markups. Those CROs, in turn, must aggressively maximize billable hours to sustain profitability. The result is, again, high utilization and longer queues.

Under this model, the only way for a sponsor to guarantee rapid turnaround is to pay even more for it. We see this with reserved capacity, ring-fenced resources, and accelerated pricing. Sponsors with deep enough pockets and acute enough need can afford to pay more. Those who can't are pushed further down the queue to wait.

A Different Incentive Structure

Traditional CROs need to maximize utilization. High utilization causes delays. How do we fix it?  

To escape the historical dynamic, Dash Bio built our business model on “outcome-based pricing,” where revenue is tied to the completed work rather than the hours billed.

To see how this changes the incentives, we can again do the math. Imagine a CRO as a set of parallel production lines. Each “line” is a staffed and equipped workflow (scientists, instruments, software, etc) capable of taking a project from start to finish. Whether that line processes one project a week or five, its underlying costs (salaries, lab space, depreciation, overhead, etc.) are largely fixed.

Under an “outcome-based” model, those fixed costs remain, but revenue is realized only when a project is completed, not when hours are logged. So, each production line carries a fixed cost, f, regardless of how heavily it is used:

If we take u, the utilization rate from the Kingman Formula, c, the project completion rate, and p, the average price you charge per project becomes:

If you complete a project on average every 5 days, c = 0.2 projects/day. If an average project costs say $20,000 and you are 50% utilized, your revenue/day works out to $2,000. The critical difference here is that c, the project completion rate, is itself a function of utilization: as utilization increases, wait time increases and you can get fewer projects completed in the same calendar time. 

I won’t go into the algebra here, but you can again plot profit margin as a function utilization and of waiting time, but get a very different outcome from the cost-plus model:

You can see that in an outcome-based model, you are no longer incentivized to push utilization to the max. Yes, if utilization is too low, you don’t make enough money to cover your fixed costs. But on the other hand, if utilization is too high, you get so backed up with work that you never complete anything to get paid.

Again, plotting this as a function of waiting time, you can see that the outcome-based model is penalized for waiting:

The longer sponsors wait, the less money you make. Furthermore, you can see from the different-colored lines which represent different service times, you are rewarded with higher profits when you reduce service time as well. So to maximize profits in an outcome-based model, one must maximize the velocity of projects, which is only done by reducing service time, wait times, and variability.

To be clear: an outcome-based model does not automatically eliminate queues or variability. Instead, it aligns our incentives with what sponsors actually care about: speed and predictability. This model pushes us to continuously drive turnaround times down and invest in improvements in quality that reduce rework and variability.

Beyond the Billable Hour

CRO delays are rarely the result of individual or organization failure in effort or intent. Instead, they are the predictable, structural byproduct of a system functioning exactly as it was designed. In a world where demand drives utilization, and utilization drives wait times, "slow" isn't an accident. It’s the default outcome of success for the traditional business model.

But the stakes of this delay extend far beyond a missed quarterly milestone or a frustrated project manager.

Every day a drug candidate sits in a queue is a day it isn't in a clinical trial. Every week lost to "billable hour" friction is a week of burned R&D capital that could have been reinvested into the next breakthrough. When the industry accepts slow as the status quo, we place a heavy, invisible tax on the timeline of every new therapy.

To change the result, we had to change the math.

At Dash Bio, we didn’t set out to simply "work harder" within an old framework. We built a system where speed and predictability are emergent properties of the model itself. By aligning our revenue with successful outcomes and replacing manual handoffs with integrated automation, we’ve removed the structural incentives that prioritize "being busy" over "being finished."

Our goal isn't just to run a faster assay; it’s to build a fundamentally new architecture for drug development. We are doing far more than optimizing a legacy CRO model. We’re engineering a faster path from the bench to the patients whose lives depend on the next breakthrough.