You Don't Have a Measurement Problem. You Have an "Immediately" Problem.

health systems margin improvement physician groups tactics Oct 15, 2026

What It Costs You Monday — a 10-part series on healthcare margin at the operating level. Start with Part 1: The Opening. You're reading Part 2: The "Immediately" Problem.

Strategy one on the list reads like this: define clinical and financial goals, measure progress, and immediately address deviations from benchmarks.

Nobody argues with it. And almost nobody does it — though not for the reason the sentence implies.

Health systems and practices do not fail at measuring. Most of them have more reports than anyone reads. They fail on one word, and it is the shortest one in the sentence.

"That is not a reporting failure. It is an accountability failure wearing a reporting failure's clothes."

THE WORD THAT DOES ALL THE WORK

Here is what actually happens. A variance shows up. Somebody notices it. It gets discussed at the next meeting. And then it gets absorbed — folded quietly into the new normal, with nobody's name attached and no date. Six months later it is not a variance any more. It is the baseline.

That is not a reporting failure. It is an accountability failure wearing a reporting failure's clothes, which is why buying a better dashboard never fixes it. The organization could already see the number. What it could not do was make anybody own it by Thursday.

TWO WAYS TO LOOK, AND ONLY ONE IS OBVIOUS

There are two tactics I work through with clients, and only one of them is what people expect.

The full line-by-line review. Every department, one at a time, asking the same four questions of each: is there a physician incentive opportunity here, a productivity option, a cost I can get out, a service I could add that drives revenue? Then the same discipline against the financials underneath — every expense, every account, every time. Not a quarterly review skimmed for anything obviously wrong. That process is written up in full in The Red Queen's Trap, Part 3.

A whale curve. Start with the practice as a whole, add its average RVU rate, and that is your baseline curve. Then pull it apart — by physician, by service, by surgical versus medicine patient — and see what pops.

The difference between those two is the whole point of this piece. The department review slices by org chart. The whale curve slices by economics.

WHAT THAT DIFFERENCE IS WORTH

We ran a whale curve on a practice and found that the capitated patient group was the segment giving margin back.

They were being run as though they were fee-for-service. Which meant missing everything that maximizes payment and lowers cost under capitation — scheduling first among them — plus revenue cycle under-billing inside the capitated model itself.

It cleaned up quickly. $190,000 recovered.

And here is why the department review would never have found it. Capitated patients are not a department. A department-by-department pass looks straight at them and sees nothing wrong, because there is no box on the org chart to look inside. What the curve really showed was that nobody had been looking at details which would have jumped off the page if anyone had.

ONE HONEST CAVEAT

A whale curve is not a definitive tool and it is not a simple report. It is a very good tool for somebody who understands what they are seeing and how to manipulate it. If you pull one and it just looks like a curve, that is the tool working correctly and the reading not having happened yet.

So it is not a dashboard you buy. It is a question you learn to ask, and the question is: where does this organization's economics actually sit, as opposed to where does its reporting structure sit?

FIVE QUESTIONS WORTH ANSWERING IN WRITING

Name the last variance we caught. Who owned it, and by what date?
How many of our reports does anybody act on, as opposed to receive?
Which of our economics cut across departments rather than sitting inside one — payer class, service line, patient type?
If a whole patient population were being run under the wrong model, which report would have told us?
What is the longest a known variance has sat in this organization without an owner?

Measurement was never the hard part. Somebody's name and a date — that is the hard part.

Ready to talk about where your organization stands? Book a free strategy session · Connect with me on LinkedIn, where this series runs.

DR Advisors — Physician's Trusted Advisor