How to Benchmark Your Payer Reimbursement Rates Against Medicare

By Clayton Severson, Principal Consultant · Published July 23, 2026

To benchmark a payer contract, restate every rate as a percentage of Medicare: your contracted rate divided by the Medicare allowable for the same code, locality, and year. Medicare is public and code-level, so it turns a wall of fee schedules into one comparable scale — then you volume-weight it to rank your payers best to worst.

Most provider organizations walk into payer negotiations with anecdotes; the payer walks in with actuaries. Percent of Medicare is how you close that gap. It is the lingua franca of rate negotiation — a single, defensible scale that a practice administrator, a CFO, and the payer's own network team all already understand. This article is the method we run: how to build it, where the data comes from, and the mistakes that quietly make the number wrong.

What does it mean to benchmark reimbursement rates against Medicare?

It means expressing what each payer pays you as a share of what Medicare pays for the same work. Medicare publishes an allowable amount for nearly every procedure code, specific to geography and year. That makes it the one reference every payer can be measured against. A commercial contract at 150% of Medicare on a code pays half again what Medicare pays; one at 95% pays below Medicare. Put all your rates on that scale and two questions you could not answer before become obvious: which of my payers pays worst, and on which of my codes.

How do you calculate percent of Medicare?

Per code, it is one division: your rate over the Medicare allowable for that exact code, locality, and year. The care is all in the denominator — getting the right Medicare allowable. Below is an illustrative benchmark for a single payer across a few common codes. The rates are made up to show the shape of the output; your locality's real Medicare allowables and your real contracted rates will differ.

Code	What it is	Payer A contracted rate	Medicare allowable (your locality)	% of Medicare
99213	Established patient office visit, level 3	$95	$82	116%
99214	Established patient office visit, level 4	$120	$116	103%
99204	New patient office visit, level 4	$185	$167	111%
93000	Electrocardiogram, complete	$18	$16	113%
20610	Major joint injection/aspiration	$54	$62	87%

Illustrative figures only — not current CMS values. The point is the last column: 20610 is the weak cell, paying below Medicare while the E/M codes clear 100%.

One code paying 87% of Medicare is a data point. To turn a table like this into a strategy you need two more things: the right benchmark source, and a way to weight codes by how much they actually matter.

Where do you get the Medicare benchmark numbers?

For physician and outpatient professional services, the source is the Medicare Physician Fee Schedule (PFS), and it is fully public. Medicare's payment for a code is built from three relative value units — work, practice expense, and malpractice — each adjusted for local cost, then scaled by a national dollar multiplier:

Each RVU component carries its own geographic index, because the three cost drivers don't move together across the country. Indexed so the national average equals 1.000, the three GPCIs are:

• Work GPCI — adjusts the physician-work RVU for local differences in the cost of physician labor. It varies the least of the three, and Congress has periodically applied a 1.0 floor so lower-cost areas aren't paid below the national average for the work itself.
• Practice Expense (PE) GPCI — adjusts the practice-expense RVU for local overhead: commercial rent, non-physician staff wages, supplies. It varies the most, and it is the main reason the same code pays materially more in San Francisco or Manhattan than in a rural locality.
• Malpractice (MP) GPCI — adjusts the malpractice RVU for the local cost of professional-liability insurance, which swings sharply by state with the malpractice environment.

Three practical notes. First, practice-expense RVUs differ for facility versus non-facility settings — the same code has two Medicare allowables, and mixing them up is the most common benchmarking error. Second, if you bill facility or hospital-outpatient work, the professional PFS is the wrong table; use OPPS Addendum B for hospital outpatient and the ASC rates for surgery centers. Third, the conversion factor changes every January, and Congress has repeatedly passed mid-year adjustments, so pin the benchmark year to each contract rather than benchmarking last year's rates against this year's Medicare.

How do you benchmark your payer rates against Medicare, step by step?

The full method is six steps. It moves from raw contracts to a ranked list of which payers and which codes to fix first.

1. Structure your fee schedules. Get every payer's contracted rates into one table: payer, plan, CPT/HCPCS code, modifier, place of service, contracted rate, effective date. Fee-schedule exhibits usually arrive as PDFs; extract them into rows you can join on.
2. Pull your volume. Export twelve months of units and charges by code from your practice-management system. This is what tells you which codes are worth benchmarking and how to weight them.
3. Build the Medicare benchmark. For your locality and the contract year, compute the Medicare allowable per code from the PFS (RVUs × GPCI × conversion factor), using facility or non-facility rates to match how you actually bill.
4. Join and divide. Match contracted rate to Medicare allowable on code, modifier, locality, and year, and compute percent of Medicare for every code.
5. Volume-weight per payer. Weight each code's percent of Medicare by that code's share of your volume and sum, giving one volume-weighted percent of Medicare per payer — the number that ranks payers best to worst.
6. Find the weak cells. Sort by high volume and low percent of Medicare. Those are your negotiation targets: the codes where a rate increase moves the most annual dollars. Model the dollar impact at current volume before you ever sit down.

Which codes should you benchmark first?

The ones that move your money — not all of them. Sort your twelve months of volume by allowed or charged dollars and you will almost always find that roughly twenty codes drive 80% or more of professional revenue, typically the office-visit E/M codes plus a short list of procedures. Benchmark those first and negotiate those. Spreading effort evenly across 400 codes dilutes the analysis and hands the payer a negotiation over rates that do not matter. Concentration is the whole point: negotiate the twenty-five codes that carry 80% of the dollars, not the long tail.

How do you compare one payer to another?

With the volume-weighted percent of Medicare. A raw table lets a payer look good on a handful of codes you rarely bill and bad on the ones you live on. Weighting each code by its share of your actual volume fixes that: it collapses an entire fee schedule into one figure that reflects your real mix. Rank your payers by it and the conversation with the worst one writes itself — you can state, in a single sentence backed by their own rates and your own volume, that they pay you materially below your other contracts across the codes that carry your practice.

Should you benchmark against market rates too?

Percent of Medicare gives you a floor and a common scale. It does not tell you where you sit among your peers — and "you pay me below the market" is a sharper argument than "you pay me below some multiple of Medicare." That comparison is now possible because of price transparency.

The catch is data quality. MRFs are enormous and notoriously dirty — ghost rates for services a provider never renders, stale files, duplicate records. A market-percentile claim is only as defensible as the cleaning rules behind it, so treat MRF benchmarking as a data-engineering problem, not a spreadsheet download. When it is done properly, the combination — percent of Medicare for the common scale, market percentile for the peer comparison — is what turns a rate request into an evidence package.

Benchmarking versus getting paid what you negotiated

One more distinction, because it is where money hides. Benchmarking asks whether the rate you agreed to is fair. It says nothing about whether the payer is actually paying that rate. That second question is underpayment detection: compute the expected allowed amount from the contract for each claim line, compare it to the actual allowed amount on your 835 remittance, and flag the systematic gaps. In revenue-cycle work I have built the parsing that reads 835 remittances down to the claim-line allowed amount, and the same pipeline that benchmarks rates should also be checking that those rates are honored. A fair contract that is being underpaid still bleeds; a rate paid exactly as written can still sit far below market. Benchmarking finds the second problem, underpayment detection finds the first, and a serious payer strategy runs both off the same data.

Frequently asked questions