Two Right Answers: Why Finance and Clinical Ops Disagree on Procedure Cost

Picture a monthly review at a surgery network. The CFO's slide shows what a knee replacement costs. The COO's slide shows a different number for the same procedure. Both teams pulled their numbers from real systems. Both did the math correctly. The meeting ends with a request to “reconcile the numbers and come back next month.”

Sound familiar? Nobody in that room is wrong. They are answering two different questions with the same word. And until someone decides what “procedure cost” means, every dashboard built on top of it will show the same argument, just faster.

Two Definitions, Both Defensible

Finance usually starts from the general ledger: supplies at what they cost to buy, plus payroll allocated across cases. Clinical Operations usually starts from the EHR: how many minutes the case spent in the OR, and who was in the room. Each view is built for a real job. Finance needs a number that ties to the books. Clinical Ops needs a number that reflects how the OR actually runs.

This isn't a rare problem. In a 2019 survey of 226 healthcare executives and directors by HFMA and Strata Decision Technology, cost accounting abilities ranged on average from “limited” to “moderate,” and about 63% of organizations still used the ratio of costs to charges, a broad estimate, to calculate labor costs. When the official method is an estimate, it's no surprise that the people running the OR build their own. The gap hasn't closed. In a 2026 Deloitte survey of healthcare finance leaders, 73% said they were expected to be regularly or heavily involved in enterprise decisions, but only 49% felt well equipped to contribute, in part because information is fragmented across financial and clinical domains.

Why a New Platform Doesn't Settle It

A common response is to invest in a new data platform or dashboard tool, hoping that one place for the data will produce one answer. It rarely does on its own. Move both definitions into the cloud, and you get both numbers on a faster, more expensive screen. The platform can store the data. It can't decide what the data should mean.

I wrote about the same pattern outside healthcare in The Cloud Migration Nobody Owns. The technology moves the data perfectly. The meaning gets left behind.

The Same Patient, Three Different Records

Definitions are only half the problem. To cost a procedure, you have to connect the case across the EHR, billing, inventory, and payroll systems. If each system knows the patient by a different ID, the costs never fully line up, no matter which formula you use.

This is common, too. In a 2021 Black Book Research survey of 1,485 health technology managers, respondents estimated that 35% of all denied claims came from inaccurate patient identification or information, costing the average hospital about $2.5 million in 2020. These are self-reported estimates, but they show how quickly small identity gaps turn into real money.

Multiple Versions of the Truth

Everyone wants a single source of truth. In reality, Finance and Clinical Operations may each have a definition that is correct for their specific needs. The challenge is not deciding who is right, but determining which definition should be used, when, and for what purpose. That is the role of data governance.

Good governance answers three questions: who owns the number, what the rules are, and who settles it when teams disagree. When they do, a cross-functional data governance council resolves the conflict by setting the rules: one shared definition, or clearly named versions for different uses. The decision is then applied consistently across the organization.

Master Data Management (MDM) applies those decisions to the records. Using matching rules the business defines, it links the same patient, provider, or supply item across systems to one golden record, so every system is talking about the same case.

Governance decides the rules. MDM makes them actionable.

How to Get to a Number Everyone Trusts

The fix is less about technology and more about order. Agree on the meaning first, then build it into the data, then build the dashboard.

1. 1. Map Every Version (Clarity)

   Write down every version of the number that is in use today. For each one, note the formula, the systems it pulls from, and the decision it is used for. Most teams have never seen their definitions side by side.

2. 2. Agree on Meaning and Owners (Alignment)

   Put Finance, Clinical Operations, and IT in the same room and agree on what each number means and who owns it. Sometimes that ends in one definition. Often it ends in two clearly named metrics, such as direct case cost and fully loaded case cost, each with an owner and a clear use. This is data governance in practice.

3. 3. Build It Into the Data (Delivery)

   Build the agreed definitions into the data itself, not into each report. Match the same patient and case across the EHR, billing, inventory, and payroll systems, check the data where it enters, and only then build the dashboard.

The goal isn't always one number. It's numbers with names, owners, and a clear purpose, so the next meeting is about what to do, not which figure to believe.

Five Signs Your Cost Numbers Need an Owner

If two or more of these sound familiar, the issue is likely definitions, not your people or your tools.

1. Finance and Clinical Ops bring different cost numbers for the same procedure to the same meeting.
2. A margin question takes weeks to answer because someone has to pull and match data from several systems by hand.
3. Nobody can say which definition a dashboard is using without asking the person who built it.
4. The same patient or case shows up under different IDs in different systems.
5. A new platform or dashboard went live, and people still check the numbers in a spreadsheet before they trust them.

The Key Takeaway

When Finance and Clinical Ops disagree on cost, it's tempting to ask who is right. Usually both are. The real gap is that nobody owns the definition, and nobody has connected the systems the definition depends on. Close those two gaps, and the same data that started the argument can end it.

So the next time two leaders bring two numbers for the same procedure, who in your organization decides what the number means?