The fee schedule states the intention. The comp plan sets the price.
In February, the owner of a three-doctor small-animal practice collecting $2.8M approved the annual fee increase. 6% across the schedule, loaded into the practice software March 1. She had the regional benchmarks, the appointment book was full, and nobody on the team pushed back. By June, revenue per invoice was running below the prior year. The fee schedule went up. The prices came down. Nobody edited the schedule in the software, and nobody had to. The one field that grew was the discount field. The practice is a composite drawn from engagement work; the arithmetic below is the arithmetic that surfaced when we decomposed it.
The number the owner trusted was the fee schedule. It records what the practice intends to charge, it gets reviewed once a year against the benchmarks, and it feels like the pricing policy because it has the fees printed in it. The document that actually sets the practice’s prices is the compensation plan. The schedule states an intention. The comp plan decides, invoice by invoice, what the practice accepts, because the comp plan is the document that tells each doctor what a discount costs her personally. In most owner-run practices those two documents were written in different years, for different reasons, and reviewed by different people. When they disagree, the comp plan wins. It is the one attached to money.
One estimate, two answers
Take a $2,400 dental treatment plan at schedule fees, extractions included. The associate is paid 20% of her production, the bottom of the 20 to 22% band that practice appraisers and the ProSal literature both describe as standard for companion-animal general practice. The client winces at the estimate, and the associate reads the room: hold the number and the plan gets approved six times in ten; take 10% off and it gets approved seven and a half times in ten. Composite figures, rounded for the demonstration, and the structure survives any reasonable substitution.
The associate’s decision. At full fee her production credit pays $480. At the discounted $2,160 it pays $432. Holding: 0.60 x $480, an expected $288. Conceding: 0.75 x $432, an expected $324. The discount is worth $36 to her. She grants it, and on her math she should.
The practice’s decision. Delivering the dental costs $1,300 in technician time, anesthesia, drugs, and allocated overhead. At full fee the practice clears $2,400 less $480 in doctor comp less $1,300 in delivery: $620. Discounted, it clears $2,160 less $432 less $1,300: $428. Holding: 0.60 x $620, an expected $372. Conceding: 0.75 x $428, an expected $321. The discount costs the practice $51 in expectation. On the practice’s math, the number holds.
Same patient. Same client. Same probabilities. Opposite correct answers. Neither party ran the math wrong, and nobody misbehaved. The comp plan handed one pricing decision to two parties and turned it into two different problems, and the person standing in the exam room is solving the one that pays her. A single case is small money. A doctor makes this call several times a day, every doctor in the building makes it, and the exam room is exactly where the fee schedule has no one to defend it. That is how a schedule rises 6% while revenue per invoice falls.
Why the plan prices this way
The plan splits every discount unevenly. A 20% production share means the doctor absorbs 20 cents of each discounted dollar and the practice absorbs 80. That $240 concession cost her $48 and the practice $192. This is incidence, the oldest question in price theory: who actually bears a cost, as opposed to who nominally grants it. Whoever bears the smaller share of a price cut will grant more of them. That follows from the split. The instinct is to see a soft-hearted associate who can’t hold a number in front of a crying client. The split predicts the behavior regardless of who is in the room, which is why replacing the associate changes nothing. Mark Opperman, whose ProSal model is the ancestor of most of these plans, has estimated for years that a discounted dollar takes roughly four dollars of new revenue to replace. The warning and the plan have coexisted for decades. The incidence arithmetic above is the reason the warning keeps losing.
Production comp is blind to margin. The plan pays the same 20 cents whether the dollar cost the practice 55 cents to deliver or 85. A doctor paid on gross production will rationally pursue gross, and the practice’s profit lives entirely in the spread the plan can’t see. The services with the thinnest margins are often the ones where a client balks and a discount closes the gap, so the blindness compounds: the plan steers its holders toward conceding exactly where a concession does the most damage.
And a production share prefers volume at any workable fee. More approved plans always pay the doctor more. The practice runs on finite doctor hours and finite surgery slots, and the discounted dental consumes the same anesthesia block as the full-fee one it displaced. Capacity is the constraint the plan never met. AVMA benchmarking has the average practice losing roughly 95 active clients a year since 2019, and a shrinking client file is exactly the backdrop that makes a walk-out feel expensive in the exam room and a concession feel cheap. The plan and the fear point the same direction.
The number your reporting can’t produce
Call it the discount incidence report. For every doctor with discretion at the invoice: twelve months of her discounts, waived line items, and comfort adjustments, each split by the plan into her share and the practice’s share. Your software already totals the discount field, and that total is the practice-wide leak. It doesn’t show whose hand granted it, and it doesn’t show what fraction of each conceded dollar that doctor personally paid. Those two columns are the ones that explain the behavior. At the composite practice, roughly 70% of the year’s discounting traced to two of the three doctors, and their personal share of the dollars they conceded was 20 cents on the dollar. They were spending practice money at a four-to-one exchange rate, with authorization, per the plan. The report takes an afternoon to build from data the software already holds. Most owners have never sorted the leak by the hand that opened it.
The governing question
The owner arrived asking whether the fees were right. That was the wrong question. The fees were fine; the benchmarks said so, and clients kept approving full-fee plans six times in ten. The governing question is who holds pricing authority under the comp plan, and what each concession costs the person holding it. Get that split wrong and no fee schedule survives contact with the exam room. This is why RIDA, the proprietary economic discipline B.L. Sheets & Co. runs on, sequences decision architecture ahead of any fee work: pricing guardrails only hold when the comp plan stops paying people to breach them.
The fix is rarely dramatic. Sometimes it is a production credit computed on contribution instead of gross. Sometimes it is a discount threshold above which the concession comes out of the doctor’s credit at the practice’s ratio instead of her own. In every version, the move is the same: make the person who sets the price bear the price’s consequences in the proportion the practice does. Alignment does the enforcement, and the fee schedule starts meaning what it says.
A practice prices the way it pays.
Sources
Mark Opperman, CVPM, in dvm360, on the ProSal compensation model, standard associate production percentages, and the revenue cost of discounting. Byron Farquer and David McCormick, What It’s Worth: Veterinary Practice Value, on small-animal associate compensation at 20 to 22% of production. AVMA Veterinary Economics Division benchmarking, 2024 data, on active-client trends. The practice in the demonstration is a composite drawn from engagement work, and its figures are labeled as such in the text.
The Decision Layer publishes on Thursdays. Each essay takes a number the operator trusts and shows what it was actually measuring.