“Where is the upside the seller left?” The usual answers are growth answers: another provider, another site, a long payer (insurance company) fight.
Those take quarters. They cost money before they earn.
Here is the answer: medical practice margin expansion starts as money you already earned and did not keep: leaks on the path from visit to bank.
No new patient required. That kept revenue shows up as EBITDA (earnings before interest, taxes, depreciation, and amortization), the operating profit buyers and boards watch.
Add providers later. Stop the leaks first.
A visit becomes a charge (the visit written as a billable line), then a claim (the bill to the insurer). The insurer pays or denies.
Payment posts in billing. Patient share collects (what the patient still owes after the payer pays). The patient returns, or does not.
Money leaks at each step. Reading those leaks as margin is the job across the group.
Related: portfolio oversight, fixes priced into the multiple, note-to-claim gap.
Margin from operations, not from growth
EBITDA is the number the group is measured on. The gap between a thin margin and a strong one is not always revenue. It is operational lift: the EBITDA improvement that comes from operations rather than from growth.
Same patients. Same insurers. Same providers. The practice keeps money it was already earning and losing.
That distinction matters to the model because the two kinds of improvement arrive on different clocks. Revenue growth takes quarters. Operational lift shows up in the numbers within one to three months, because the revenue already exists and only the leak changes.
Start with operations, and the growth work starts from a higher floor.
Where the EBITDA is usually hiding
- Visits never marked kept
- Notes unsigned (clinical documentation not finalized)
- Charge lag (days from visit to billable charge)
- Claims stuck or denied on the same causes over and over
- Payments unposted (received or deposited, but not recorded in billing)
- Patient balances never worked
- Follow-up patients who never rebook (no appointment created, so no alert fired)
Add-a-provider math can wait until these are named. Otherwise you grow the leak.
A completed visit with no charge behind it is not a denial and not an aging balance. It is nothing, and nothing does not appear on any report.
A clinician’s unsigned note holds a charge that cannot bill. Every one was care delivered and unbillable.
A patient’s share is easiest to collect while the patient is standing there and hardest weeks later by statement. The gap between those two collection rates is margin, at every desk in the group.
Working capital (the money tied up between the visit and the deposit) is financed on the group’s line of credit while it sits. Shortening the wait from visit to usable cash releases money that had been sitting in transit. None of it is new revenue.
Why the current owner left it
Not carelessness. Structure.
Each leak lives between two roles:
- Charge capture falls between the office manager’s day and the biller’s queue.
- The unsigned note sits between the clinician and the claim that cannot post without it.
- The patient balance falls between the front desk and the aging report (the list of unpaid bills by age).
- The patient who did not return falls between the last visit and a next one nobody scheduled.
Nobody at the practice owns the space between roles, so nobody is failing at their job while the money leaves.
The seller was not hiding the upside. The seller could not see it, for the same reason the diligence firm could not: a review of the financials cannot examine an event that produced no record.
That is also why the upside survives the transaction intact. Nothing about closing the deal changes where the roles stop. Revenue drops after an acquisition through exactly these gaps, and it keeps dropping until someone measures the space between roles weekly.
How to rank leaks without drama
Pick one closed month at one site. Estimate dollars stuck at three stops only:
- Status / note / charge (work done, bill never started)
- Claims / denials (bill started, money blocked or refused)
- Patient share (payer paid, patient balance untouched)
Rank by dollars, not by who complains loudest. Assign one owner to the top stop. Repeat at the next site.
Portfolio oversight is how you run that rank across the group every week.
The note-to-claim gap is usually where status, note, and charge hide the first pile.
Kept revenue vs new volume
New volume adds clinician cost, overhead, and ramp time before it adds cash.
Kept revenue on visits you already delivered uses capacity you already paid for. That is why path leaks are the fastest margin still inside the group.
Growth can still be right. Rank it beside the leak. Fund the leak that pays back first, then grow from a cleaner path.
Real situations that look like “we need growth”
A site asks for another clinician. Last month’s completed visits still wait on unsigned notes. Volume will not fix a note bottleneck.
Another site wants a new location. Charge lag and a hot denial cause on repeat (the same upstream gap producing refusals) already thin cash on the visits they have.
A third site chases a payer rate fight. Patient shares on paid claims sit untouched. The closer dollars are already in house.
A fourth reads “soft month” on the pack. Cut the same analysis by clinician and the loss is not spread evenly. A follow-up booked before the patient left, at a few desks in particular, brings the recurring revenue back on the next patient cycle.
If the growth deck ignores those piles, you are pricing a leak into the plan.
What this means for the value creation plan
The value creation plan (the post-close roadmap for improving the practice) usually leads with growth because growth is what the model was built on. Put operational lift first instead, for three reasons:
- It is faster. Weeks to months, against quarters.
- It is cheaper. No headcount, no new site, no contract fight.
- It is structural when it is done right, which is what gets priced in at exit.
A leak closed by an instrument that keeps running holds its value across the hold period (the years between buying and selling). A one-time cleanup fades.
See fixes priced into the multiple for how buyers tell those apart.
What a buyer will keep
A weekend note blitz lifts EBITDA once. A note SLA with a named owner is what persists. Tell the margin story with what stays fixed.
How to estimate a leak without inventing precision
You do not need a perfect model. You need an honest pile.
- For status, note, and charge: count completed visits with no charge yet, then multiply by a careful average allowed amount you already trust from recent paid claims. Direction is enough.
- For claims and denials: sum open stuck dollars and denied dollars still workable at the hottest denial cause.
- For patient share: sum balances on paid claims that nobody worked last month.
If the top pile is obvious, stop estimating and assign the owner. False precision is how the growth deck wins the week.
Growth-deck red flags
Watch for these lines:
- “We need more providers” while unsigned notes and charge lag are unnamed.
- “Payer rates are the problem” while patient shares sit untouched.
- “Another site will dilute overhead” while the current path leaks on every visit.
Those bets can still be right later. They are expensive first moves when kept revenue is sitting inside the path you already run.
What to do this week (simple check)
- Pick one practice.
- Estimate last month’s lost dollars at three stops only (status/note/charge, claims/denials, patient share).
- Rank them by dollars, not drama.
- Assign one owner to the top stop.
- Repeat at the next site.
One site first, then the group
Do not boil the ocean. One closed month at one site is enough to prove the method.
Once the top leak has an owner and a weekly count, copy the same three-stop rank to the next site. Portfolio oversight is the group rhythm. This page is the margin lens you bring to that rhythm.
Where this sits on the money path
Before you add volume, read the path you already run.
Visit marked kept. Note signed. Charge created. Claim out. Denial worked or prevented. Payment posted. Patient share collected.
Each break is margin you already paid to earn.
Margin expansion here is kept revenue first. Growth second.
What this means for you
Before you add a provider or open a site, find out how much of the margin is already inside the practices you own. Every one of the leaks above has a dollar figure at every practice, and every figure comes from records the practice already keeps.
Grab 30 minutes with us. Prep nothing. You will see the first margin pocket still sitting inside a practice you already own.
Questions people ask
What is medical practice margin expansion here?
Raising EBITDA by keeping more of the revenue you already produced, not only by adding volume. In a practice it means billing every completed visit, signing every note, collecting the patient’s share at the desk, bringing patients back, and shortening the wait between visit and deposit.
How fast does operational lift show up?
Within one to three months, because the revenue already exists and only the leak changes. Working capital moves in weeks. Recurring revenue from patients who stopped returning takes a full patient cycle. Revenue growth takes quarters.
Is this the same as cutting staff?
No. It is stopping unpaid work and uncollected dollars on work already done. The leaks live between roles, so another hire looks into one more territory and misses the same gaps.
Why did the seller not fix this?
Because the seller could not see it. Each leak produces no record, so it appears on no report the seller ran, and financial diligence confirmed the records rather than the operation.
Will a buyer care?
They care what persists. Structural path fixes beat one-time backlog heroics. See fixes priced into the multiple.
Should we pause all growth?
No. Rank leaks beside growth bets. Fund the leak that pays back first, then grow from a cleaner path.
How do we know a leak is real?
Count visits or dollars stuck at a stop for one closed month. If the biggest backlog sits with nobody owning it, it is real.