Revenue per Visit and per Patient: What They Really Tell You

Revenue per visit and revenue per patient answer two different questions — and there's no universal benchmark for either. Per visit tells you what one encounter is worth (pricing and intensity); per patient folds in how often people come back (frequency and loyalty). The number that matters is net — cash collected, not charges billed — and because it's driven by your specialty and payer mix, no "the average is $X" figure will fit you. So instead of chasing a benchmark, measure your own and watch the trend — with two anchors for scale.

The two formulas — and the gross-vs-net trap

Both are simple division. The trap is in the numerator.

The word "revenue" is doing all the work, and it's where practices fool themselves. Gross revenue per visit uses your billed charges — a list price almost no payer actually pays, so the number comes out inflated. Net revenue per visit uses the cash you actually collected. That's the real one, and it connects straight to another metric:

Now the distinction almost every article misses. Revenue per visit is about a single encounter — how it's coded, what it includes, what the payer allows. Revenue per patient multiplies that by how often the patient returns. Two clinics can collect the identical amount per visit and be worlds apart per patient: if one sees each patient once a year and the other four times, the second earns four times as much per patient with the same per-visit economics. Per visit is the price of the encounter; per patient is that price times loyalty.

A note on two neighbours: revenue per provider (revenue ÷ FTE physicians) is the version MGMA actually publishes, and it's really revenue per visit × visits per provider. And patient lifetime value is roughly revenue per patient per year × the years you keep them — we work that one through in how to calculate patient lifetime value.

Is there a benchmark for revenue per visit? (Short answer: no)

People want a target — "what should my revenue per visit be?" — and the honest answer is that a single universal number can't exist. It's dominated by two things you can't average away:

• Specialty. A behavioral-health session, a primary-care visit and an orthopedic consult are not the same product and were never priced to be comparable.
• Payer mix. The exact same visit pays differently under Medicare, Medicaid and commercial plans. Shift your mix and the number moves with zero clinical change.

What is real and public are two government anchors — not targets to hit, just reference points for what a visit is even worth:

Medicare's Physician Fee Schedule pays roughly $89 for a routine established-patient office visit (code 99213) and about $125 for a more involved one (99214) in 2025 — a real, published price, though only for one payer and before your local adjustment. And the AHRQ Medical Expenditure Panel Survey put the all-payer average at about $265 per office visit — but that's 2016 data, it spans every office visit including procedures, and it hides an enormous specialty spread (roughly $186 for primary care up past $400 for orthopedics — illustrative 2016 figures, not targets). And both anchors sit a notch above your net collected dollar — Medicare's is an allowed amount, MEPS is total expenditure — so use them to sanity-check your own number, not to grade it.

What actually moves your revenue per visit

If you want the number to go up for real (not just re-billed), these are the levers, strongest first:

Coding level and documentation

The clearest lever, because the dollars are fixed and public. On Medicare in 2025, the gap between a 99213 and a 99214 is about $36 — so systematically documenting one level low adds up fast. AAFP has illustrated it plainly: 250 Medicare visits coded 99213 when they should have been 99214 is nearly $9,000 of revenue left behind.

But here's the half most "you're under-coding!" pitches skip. The most-cited evidence — a study by King and colleagues — found family physicians agreed with expert coders on only about half of established-patient notes, and where they differed the most common error was under-coding; yet on new-patient notes they agreed just 17% of the time, and there the predominant error was over-coding. Coding is inaccurate in both directions. And that evidence is a 2001 study — a survey using hypothetical vignettes, under the old E/M rules that were replaced in 2021. So the honest claim is "code accurately, because a level is worth about $36 and the errors run both ways" — not "everyone is leaving money on the table."

Payer mix and visit mix

Two levers that move the number without touching a single clinical decision. Payer mix: the same 99214 pays more from a commercial plan than from Medicaid, so a shift toward better-paying contracts lifts revenue per visit on its own. Visit mix: a new-patient work-up (a 99204 allows about $163 on Medicare in 2025) is worth more than a routine follow-up (99213 at ~$89), and a visit with a procedure attached is worth more than one without. None of this is about seeing patients faster; it's about what each encounter actually contains.

Collection rate and no-shows

The last two are about not losing what you've earned. If you bill well but collect poorly, your net revenue per visit falls — which is the whole gross-vs-net point again, and why net collection rate sits right next to this metric. And an empty slot earns $0: no-shows are pure lost revenue per visit. They're common — a systematic review of 105 studies put the average outpatient no-show rate around 23% (though that's a global figure, not US-specific) — and the sober way to price the damage isn't the "$150 billion a year" folklore you'll see quoted, but simple: your no-shows per week × your net revenue per visit. We cover the fix in how to calculate and cut your no-show rate.

What moves your revenue per patient

Everything above still applies — but the lever unique to revenue per patient is frequency. A patient who comes back is worth a multiple of one who doesn't, so the biggest driver of revenue per patient isn't price at all; it's retention. The others: how many services a typical visit includes, and the makeup of your panel — a practice managing chronic conditions naturally earns more per patient than one seeing mostly one-off acute complaints, because those patients return on a schedule. If revenue per patient is climbing while revenue per visit holds steady, that usually means more visits per patient — often patients choosing to come back, though a sicker or more chronic panel can drive it too.

How to measure your own baseline

Since no benchmark will fit you, the whole exercise is your own trend. It needs three numbers you already have:

• Collected revenue for the period (net, not charges).
• Visit count — completed encounters.
• Unique patients seen in the period.

Divide two ways, then do the one thing a benchmark can never do: segment. Revenue per visit by provider surfaces a coding or scheduling gap; by payer shows you which contracts actually pay; over time tells you whether a change helped. The number in isolation is noise — next to last quarter, and split by provider and payer, it's a decision.

Frequently asked questions

Written by

Olha · clinic data analyst

I build the reporting our managers open every morning at a multi-branch medical clinic — and package it so other practices don't have to start from scratch.

Dollar figures here are labelled by what they are — a government price schedule (Medicare), an all-payer survey (MEPS), or an illustrative example — with the data year noted. Medicare amounts are national figures before local adjustment and can shift with mid-year rule changes. Where no reliable benchmark exists, we say so rather than borrow one.