Your Overhead is Growing Faster Than Your Revenue. That Isn’t a Bad Year – It’s the Model.

A lot of the groups we work with have a version of the same story. Volume held up. The providers worked as hard as they did last year, or harder. Collections were flat or even up a little. And distributions still came in lower than the year before.

The instinct in that moment is to look for the leak (i.e., a payer that slow-paid, a bad month in the surgery schedule, a staffing decision that didn’t work out). Sometimes there is one. More often, there isn’t. The practice is bailing faster than ever and the water keeps rising, because the problem was never the rate of bailing. It’s the hull.

This is not a story about mismanagement. Some of the most disciplined practices we see are losing ground, and they are losing it for reasons that have very little to do with how well they are run. What separates the practices that hold their margin from the ones that don’t is not effort. It is whether anyone is treating the cost structure itself as something to be designed.

Two numbers

Between 2001 and 2025, Medicare physician pay remained essentially flat while the cost of running a medical practice rose 59%. Adjusted for practice-cost inflation, Medicare physician payment declined 33% over that period.¹

That is the whole argument. Everything below is an explanation of that gap and what a practice can actually do about it.

The same divergence appears in benchmark data on the cost side. Median total operating cost per full-time-equivalent physician rose more than 63% from 2013 to 2022, while the Medicare conversion factor increased 1.7% over the same stretch.² And it did not start in the last few years. Reviewing five-year trend data back in 2018, MGMA found that median total operating cost per FTE physician in physician-owned multispecialty groups had climbed 29.3%, from $620,098 to $801,938, while total medical revenue per FTE physician grew only 16.6%, during a stretch when the Consumer Price Index rose 5.2%.³

So the trajectory has been visible for well over a decade. What has changed is that the cushion absorbing it is largely gone. In mid-2026, 84% of medical groups reported year-to-date operating costs higher than the same period in 2025, while only 47% reported higher year-to-date revenue, down from 56% the year before, with 36% reporting an outright decline.⁴ The population of practices gaining ground has been shrinking.

A caution before you look at your overhead percentage

Overhead as a percentage of collections is the number most practices reach for first, and it is the right place to start. But it is a ratio, and a ratio moves for reasons that have nothing to do with cost discipline. Before you draw conclusions from yours, understand what else is driving it:

• Payer mix. A practice with a heavier commercial mix will show a lower overhead percentage than an otherwise identical practice with heavier government payer mix, because the denominator is larger. Nothing about the cost structure differs.
• Ancillary revenue. Adding pathology, imaging, infusion, or a dispensary changes both sides of the ratio, usually favorably. Losing one works the other direction. Neither movement tells you whether your core operation got more or less efficient.
• Provider mix. Bringing on advanced practice providers typically raises collections faster than it raises overhead, but APP compensation itself has been climbing quickly, and where the practice books that cost determines whether the ratio looks better or worse.
• Ownership model. Physician-owned practices generally carry more support staff on site than hospital-owned practices, which can centralize business office functions across locations.⁵ Comparing your ratio to a benchmark drawn from a different ownership model will mislead you.

The practical takeaway is this: benchmark against your own specialty and ownership model, and pay more attention to your own trend line than to the peer comparison. A ratio that has drifted three points in two years is telling you something real regardless of where it sits relative to the median.

What it looks like on your P&L

Consider a three-physician specialty practice collecting $3.0 million a year with overhead at 55% of collections. That leaves $1.35 million to distribute, or $450,000 per physician.

Now run it forward two years with costs rising 7% annually and collections rising 3%. Collections reach roughly $3.18 million (i.e., genuine growth of about $183,000). Overhead reaches roughly $1.89 million. Distributable income falls to about $1.29 million, or roughly $431,000 per physician. Overhead has moved from 55% to just over 59%.

The partners collected more and took home less; and in many cases without anyone having made a bad decision along the way. There is often no leak to find.

Extend the same assumptions to year five and the picture gets harder. Collections reach about $3.48 million. Overhead reaches roughly $2.31 million. Distributable income falls to roughly $1.16 million, or about $388,000 per physician, a decline of more than 13% from where they started, on collections that grew 16%. The overhead ratio crosses 66%.

Now consider the most common response to that trend, which is to add a provider and grow into the problem. If the new physician or APP generates proportional collections at the same cost structure, the arithmetic does not change: a bigger practice with the same design produces the same compression, just at scale. Growth only helps if the marginal provider carries a materially better cost ratio than the existing base, which is why adding an APP to an existing physician’s panel often works and opening a second location often doesn’t.

That is what a structural gap does. It doesn’t announce itself in any single line item. It shows up at year end as a number that doesn’t match the effort.

Why this is structural

• Labor reset, and it did not reset back. Wages in ambulatory care did not spike and settle; the floor moved. The increases were sharpest in exactly the roles a practice cannot operate without. MGMA’s compensation data showed double-digit year-over-year growth for supervisors and for senior and general managers.⁶ Meanwhile, 64% of medical groups budgeted base pay increases of only 1% to 3% for 2026.⁷ A practice can hold its budget or hold its people. Increasingly it cannot do both with a single across-the-board percentage.
• Technology became a permanent line item. A decade ago, IT was a rounding error on a practice P&L. It is now a standing percentage of revenue, and cybersecurity has made much of it non-discretionary.⁸ This is not an expense most practices can expect to shed. It is also an expense that tends to accumulate quietly, through renewals and add-on modules rather than through decisions anyone remembers making.
• Payers export their administrative cost onto your payroll. Prior authorization, credentialing, contract management, denials, and appeals all require staff. Physicians in AMA survey data reported completing an average of 39 prior authorizations per week, and more than a third of practices employ staff who do nothing else.⁹ Those positions are real salaries funded out of your collections to manage somebody else’s process. When a payer changes a policy, the cost of absorbing that change lands on you.
• Payment is moving the other direction. CMS released the CY 2027 Physician Fee Schedule proposed rule on July 14, 2026, proposing conversion factors of $33.1693 for qualifying alternative payment model participants and $32.8409 for everyone else (i.e., reductions of 1.19% and 1.68% respectively from CY 2026).¹⁰ Some of that reflects the expiration of the temporary 2.5% update Congress provided for 2026 alone.¹¹ Practice leaders are not imagining the squeeze: 80% of medical groups told MGMA in February 2026 that Medicare reimbursement is below their cost to deliver care, up from 73% in 2021.¹²
• More of your revenue now sits with the patient. This is the newest pressure and the one most practices have not yet priced in. Coverage tightened in 2026 when marketplace plan selections fell to 23.1 million, and reported effectuated enrollment (i.e., coverage people actually paid for) came in at 19.2 million as of February, roughly three million below the prior year’s paid total.¹³ For those who kept coverage, cost went up sharply; a Peterson-KFF analysis of insurer filings found a median proposed premium increase of 18% across 312 marketplace insurers, with subsidized enrollees keeping comparable coverage seeing annual premium payments more than double, from $888 to $1,904.¹⁴ Higher deductibles and more uninsured patients do not show up as a cost line. They show up as slower collections, larger patient balances, more write-offs, and softer demand for anything elective, another way of saying they show up as a revenue problem that behaves like a cost problem.

Four responses that don’t work

Most practices have already tried at least one of these.

• See more patients. This is the reflex, and it is not irrational; marginal volume on a largely fixed cost base does drop to the bottom line. But it buys a year, maybe two. It adds variable cost, consumes provider hours that are already scarce, and does nothing to the underlying cost structure. It also has a ceiling, and most practices are closer to it than they think. Where it does work is in filling genuinely unused capacity: an underutilized procedure room, an APP with open panel space, a fourth day in a location already paid for. Where it fails is when the practice is simply asking already-full providers to compress visit length.
• Cut staff. This is the response that most reliably makes things worse. MGMA’s own analysis has long held that many practices exercise a false economy in staffing (i.e., reducing headcount to the point that production is constrained and total profitability falls).¹⁵ Understaffing the front desk and the business office is how a practice converts an expense problem into a revenue problem: more scheduling errors, more eligibility failures, slower claim submission, unworked denials. The savings are immediate and visible on the P&L. The cost arrives ninety days later in your aging, where nobody connects it back to the decision.

  It is also worth noting that a stable staff-to-physician ratio can conceal real strain, because the ratio says nothing about whether the remaining people are doing the right work.¹⁶ A practice can hold headcount flat and still be badly understaffed in the two functions that determine whether claims go out clean.

• Give everyone the same small raise. Spreading a 2% pool evenly across the organization protects the budget and loses the two or three people whose market value reset by double digits. Replacing an experienced billing manager or office manager costs a multiple of the raise that would have kept her, and the practice absorbs months of degraded throughput while the replacement learns the payers, the workflows, and the physicians. A differentiated pool (i.e., maintenance-level increases for stable roles, catch-up increases for the specific positions where you are demonstrably below market) costs about the same and does something entirely different.
• Sell. For a growing number of practices this is the answer, and sometimes it is the right one. The share of physicians in private practice fell from 60.1% in 2012 to 42.2% in 2024, and the share with any ownership stake fell from 53.2% to 35.4% over the same period.¹⁷ But selling is a decision about who bears the cost problem, not a solution to it; and the valuation you receive is a direct function of the margin you are producing at the time you sell. A practice that spends three years letting its overhead ratio drift and then goes to market is negotiating from the worst possible position. If a transaction is anywhere in your five-year thinking, the cost work is not an alternative to preparing for sale. It is the preparation.

Five lines worth an honest look

The productive question is not “where do we cut.” It is “where is our cost structure poorly designed.” Five places to start:

• Staffing ratio and skill mix. The useful question is not how many FTEs you have per physician but whether people are working at the top of their roles. Is a clinical staff member spending her afternoon on scheduling and records requests? Is your highest-paid administrative person doing work a coordinator could do? Mix problems are usually larger than headcount problems, and correcting them rarely requires anyone to leave. It requires someone to sit down and map what each person actually does against what the position was designed to do. Those two lists diverge quickly in a practice that has grown without redesigning anything.
• Occupancy. Compare your cost per square foot to specialty benchmarks, then read the lease (i.e., the actual document, not the summary someone made when it was signed). Look at escalators, CAM pass-throughs, renewal and termination mechanics, and space that serves a single purpose a few days a week. If the building is owned by a physician-owned entity leasing back to the practice, confirm the terms are documented at fair market value; that matters for compliance today and for valuation whenever ownership changes. Occupancy is usually the second- or third-largest line on the P&L and the one least likely to have been examined in the past five years.
• Cost to collect. Very few practices know this number in full. It is the billing salaries or the vendor percentage, plus clearinghouse fees, plus billing and clearinghouse software, plus the staff time spent working denials and appeals. Most practices know one component and assume it is the total. Calculate it as a percentage of collections and compare it to your specialty. If you outsource, compare the all-in percentage to what an internal function would cost at your volume. The answer is not always what the vendor’s rate card implies, in either direction.
• Technology spend. Print the subscription list. There is almost always duplication between the EHR’s native functionality and something purchased separately, at least one product nobody has logged into in a year, and a renewal that auto-escalated without review. Assign a named owner to every line and a renewal date to every contract. This is the least glamorous item here and frequently the fastest payback.
• Payer contract terms and underpayment recovery. This one sits on the revenue side, but it belongs in the same exercise, because a dollar recovered from an underpayment behaves exactly like a dollar of cost removed. Many practices are operating under agreements signed a decade ago with evergreen renewal provisions and no one tracking whether payments actually match the contracted fee schedule. Load your contracted rates into your practice management system and audit a sample of remittances against them. Practices that have never done this are frequently surprised, and not pleasantly.

What to measure, starting this month

None of the above works as a one-time project. It works as a set of numbers your administrator produces monthly and your partners actually look at:

• Overhead as a percentage of collections, trended monthly rather than reviewed annually
• Total operating cost per FTE provider
• Support staff cost per FTE provider
• Cost to collect as a percentage of collections
• Days in accounts receivable and the percentage of A/R over 90 days

The value is in the trend line, not the single reading. A practice that sees its overhead ratio drift a point and a half over two quarters can respond while the options are still cheap and reversible. A practice that discovers the same drift at year end is choosing among worse ones.

Plenty of practices have never built these metrics, and the reason is usually mundane: the chart of accounts wasn’t designed to produce them. Expenses get booked to categories that made sense to whoever set up the file years ago, which means overhead cannot be separated cleanly from provider compensation, or occupancy from general administrative. That tends to be the first thing worth fixing, since the other measurements depend on it.

A reasonable ninety days

If this describes your practice, the sequence matters more than the ambition:

• Weeks 1-3. Get the chart of accounts into a structure that produces the five metrics above. Pull three years of history and restate it into the new structure so you have a real trend line rather than a starting point.
• Weeks 4-6. Build the four cost diagnostics: staffing mix, occupancy, cost to collect, technology inventory. This is mostly assembling information the practice already has in five different places.
• Weeks 7-9. Audit a remittance sample against your contracted rates. Identify the two or three payer agreements worth renegotiating and confirm their notice and renewal dates.
• Weeks 10-13. Bring findings to the partners with specific options and dollar figures attached, not general concerns. Decide what changes, who owns each change, and when it gets reported on.

Nothing on that list requires a large capital outlay or a reorganization. It requires someone to own it and a partner group willing to look at the output.

The point

Rowing harder does not fix a hull problem. The practices that hold their margins over the next three years will not be the ones that worked the most weekend hours. They will be the ones that stopped treating overhead as an annual budget exercise and started treating it as a design question.

That work is not glamorous and it does not happen in a single meeting. But it is entirely doable, and it is considerably less painful than the alternative a lot of practices are drifting toward.

DoctorsManagement has worked with independent physician practices for more than sixty years on operations, financial performance, and benchmarking against specialty peers. If the numbers above sound like your practice, we offer a complimentary consultation to talk through what you’re seeing and whether an engagement makes sense.

References

1. American Medical Association comment letter on the CY 2026 Medicare Physician Fee Schedule proposed rule, as reported by Medical Economics, March 2026; see also AMA, “Medicare physician pay has plummeted since 2001,” and AMA’s 2025 Medicare updates and inflation chart (sources: Federal Register, Medicare Trustees’ Reports, Bureau of Labor Statistics, Congressional Budget Office).
2. MGMA Stat, “2026 Medicare reimbursement changes: Tracking what matters,” Feb. 26, 2026. The same analysis notes that 90% of medical groups reported increased operating costs in 2025.
3. David N. Gans, MSHA, FACMPE, “Data Mine: The administrative burden of operating a medical group,” MGMA, Aug. 21, 2018, reporting five-year trend data from MGMA DataDive Cost and Revenue for physician-owned multispecialty groups with primary and specialty care.
4. MGMA Stat, “Revenue growth narrows as costs climb: the 2026 squeeze on medical practices,” July 2, 2026, reporting a June 30, 2026 poll (221 applicable responses) alongside the prior-year June 17, 2025 comparison.
5. MGMA Cost and Revenue Survey comparative data, reporting a difference of one to three more support staff per FTE physician in physician-owned versus hospital-owned practices across specialties.
6. MGMA, 2025 DataDive Management and Staff Compensation data report.
7. MGMA Stat poll, Sept. 9, 2025 (349 applicable responses), reported in “Budgeting competitive, sustainable 2026 staff pay for your medical practice,” Sept. 10, 2025.
8. MGMA benchmark data on information technology spend as a share of revenue for outpatient groups.
9. American Medical Association, 2024 prior authorization physician survey.
10. Centers for Medicare & Medicaid Services, “Calendar Year (CY) 2027 Medicare Physician Fee Schedule Proposed Rule” fact sheet, July 14, 2026. Comments on the proposed rule are due Sept. 14, 2026.
11. AMA, “MedPAC signals need to bolster Medicare physician payments,” January 2026, noting that the temporary 2.5% update provided for 2026 expires in 2027.
12. MGMA Stat poll, Feb. 24, 2026 (166 applicable responses).
13. MGMA Stat, “Revenue growth narrows as costs climb,” July 2, 2026, citing marketplace plan selections of 23.1 million and HHS-reported effectuated enrollment of 19.2 million as of February 2026.
14. Peterson-KFF analysis of 2026 marketplace insurer rate filings, as reported in MGMA Stat, July 2, 2026.
15. David N. Gans, MSHA, FACMPE, “Cost-efficiency with medical group staffing,” MGMA.
16. MGMA Stat, “Stable staff-to-physician ratios might conceal real strain in your practice,” July 8, 2026.
17. Carol K. Kane, PhD, “Physician Practice Characteristics in 2024: Private Practices Account for Less Than Half of Physicians in Most Specialties,” AMA Policy Research Perspectives, May 29, 2025, based on the AMA Physician Practice Benchmark Survey.