The Revenue Quality Scorecard Every Medical Group Should Build

Ask most practice administrators how the group is doing financially and you will get a revenue number. Ask how the group is doing compared to what it should be collecting and the room gets quiet.

Revenue is a comforting number because it always sounds like an answer. It is not. It is a symptom. Two groups can bill the identical volume, see the identical patients, and land eleven percent (11%) apart on what actually reaches the bank. The difference is not effort. It is revenue quality, and almost nobody measures it.

A revenue quality scorecard fixes that. It is not a dashboard for the sake of a dashboard. It is a short list of numbers that tell you whether the money you earned is actually showing up.

Six numbers, and what good looks like

Most scorecards fail because they have forty metrics. Nobody reads forty metrics. Build six, put a target next to each one, and review them on a schedule.

Net reimbursement per encounter. Average reimbursement per visit after contractual adjustments, tracked monthly and split by payer. This is your earliest warning system. When a payer quietly reprices a code family, this number moves before anything else does.

Net collection rate. What you collected divided by what you were contractually entitled to collect. Not what you billed. Billed charges are a fiction everyone agrees to ignore. Target ninety-five percent (95%) or better. Anything under ninety percent (90%) is money walking out the door through late filings, unworked denials, small balance write offs nobody approved, and bad debt. That last one deserves its own conversation, and it is the reason a single blended net collection rate will mislead you. See below.

Denial rate, by payer and by reason. The aggregate number is nearly useless. A five percent (5%) denial rate that is entirely one payer and one reason is a solvable problem. The same five percent (5%) spread evenly across everything is a documentation problem. Track both dimensions or do not bother tracking it.

Days in accounts receivable. MGMA’s Cost and Revenue Survey puts the median around thirty six days for better performing practices and closer to forty seven for everyone else. HFMA’s MAP Keys target is under forty. Above fifty and you have a structural problem, not a bad month. Pair it with aging. HFMA suggests keeping accounts receivable over ninety days under ten percent (10%) of the total, and a healthy looking thirty eight day average can hide a very unhealthy tail.

Underpayment recovery rate. Expected reimbursement versus actual payment, at the claim level. More on this below, because it is the column almost no group builds.

Cost to collect. What you spend to get paid, as a percentage of collections. Most groups have never calculated it. They should.

Bad debt is a payer mix problem wearing a billing problem costume

Here is how a good revenue cycle team gets blamed for something it did not do.

The group’s net collection rate falls from ninety-five percent (95%) to ninety-one percent (91%) over four quarters. Finance flags it. Somebody suggests the billing team has gotten sloppy. Meanwhile the group opened a second location last year in a market with a very different payer mix, and the entire decline is patient responsibility that was never going to be collected from anyone.

Bad debt is not evenly distributed. It follows the product, and the products behave nothing alike.

Kodiak Solutions data published by HFMA in June 2026 shows how wide the spread gets. Point of service cash collections as a share of patient payments ran 50.2% for traditional Medicaid, 27.6% for Medicare Advantage, 24.7% for commercial and managed care, 23.0% for Medicaid managed care, and just 15.0% for traditional Medicare. Same practice, same front desk, same script, wildly different outcomes depending on what card the patient hands over.

The commercial trend is the one most groups have not internalized. Kodiak found the provider collection rate on commercially insured patients falling from 37.6% in 2023 to 34.4% in 2024, driven by high deductible designs pushing more of the bill onto the patient. Commercial used to be the easy money. It is now a growing collection risk that arrives disguised as a good contract.

Two more findings worth putting in front of leadership. Kodiak’s analysis found that bad debt write offs equaled roughly 1.54% of total claim charges in 2023, which sounds like a rounding error until you learn it added up to more than 17 billion dollars, and that fifty-three percent (53%) of those write offs came from patients who had insurance. Bad debt is not primarily an uninsured phenomenon anymore. It is an insured people with deductibles phenomenon. Kodiak also identified a fairly clean behavioral line at 500 dollars. Below that number most patients paid. Above it, most did not.

Broader industry ranges tell the same story at the organizational level. Providers serving heavily Medicaid populations commonly run bad debt in the range of four to seven percent (4% to 7%) of gross revenue, while organizations with predominantly commercial populations tend to land closer to two to four percent (2% to 4%). Treat those as directional rather than precise, because the definition of bad debt varies by organization, but the shape of the finding holds.

None of this means Medicaid heavy practices are managed worse. It means measuring them against a commercial benchmark is malpractice of a bookkeeping variety.

What to actually do with this. 

Split net collection rate three ways on the scorecard. Insurance net collection rate, which is what your billing team genuinely controls. Patient net collection rate, segmented by product. And a blended number for the board, clearly labeled as blended so nobody mistakes it for a performance metric.

Then hold the segments accountable to different targets. If insurance net collection rate is ninety-seven percent (97%) and blended is ninety percent (90%), your revenue cycle is running well and you have a patient collections problem. Those are two entirely different projects with two entirely different owners, and a single blended number hides which one you have.

This also belongs in your contracting math. A commercial plan with an aggressive deductible design is quietly transferring collection risk onto your balance sheet, and that risk has a price. When you model what a contract is worth, model it net of the patient responsibility you will realistically never see. The rate on the page is not the rate you get.

One more reason to build this now. HFMA reported in June 2026 that coverage changes under OBBBA are expected to push significant numbers of people off Medicaid and into high deductible exchange plans or out of coverage entirely, which means more patient responsibility flowing through the same billing infrastructure. Groups that already segment their net collection rate will see that shift as it happens. Groups that do not will see a blended number sag and spend a year investigating the wrong department.

The metric nobody builds

Every group assumes payers pay what the contract says. Payers assume nobody is checking. Both assumptions are usually correct.

Underpayments are not fraud. They are usually a fee schedule loaded wrong, a modifier priced incorrectly, or a code that gets bundled every single time because a system rule says so. No single instance is large enough to escalate. In aggregate they fund somebody’s partner distribution, and it is not yours.

Building the underpayment column means loading your contracted rates into your system and comparing expected to actual on every claim. It is tedious. It is also the highest return work in the entire revenue cycle, because you are not asking anyone for more money. You are asking them to pay what they already agreed to pay.

Layer everything by payer or you have wasted your time

A scorecard that shows only practice level numbers tells you that something is wrong. A scorecard layered by payer tells you who is doing it.

That distinction is the entire point. Payers do not behave the same way and should not be measured as if they do. One plan pays fast and denies rarely. Another pays market rates and then claws back a third of it through medical necessity denials that mostly get overturned on appeal, which is a strategy, not an accident. MDaudit’s 2025 data showed the average denied amount on a Medicare Advantage claim rising more than twenty-two percent (22.4%) to roughly one thousand dollars, and denials tied to requests for information and medical necessity climbing seventy percent (70%) to about four hundred fifty dollars. Those are not billing errors. Those are business decisions being made about you.

Add the administrative drag column too. The AMA’s 2024 survey of one thousand physicians found practices completing an average of thirty nine prior authorization requests per physician per week, consuming roughly thirteen staff hours. Forty percent (40%) of physicians reported having staff who do nothing but prior authorization. A contract that pays three percent (3%) above market and generates twice the authorization volume is not a good contract. It just looks like one on a rate sheet.

To learn more about strengthening payer agreements, explore our Payor Contracting Services, where we help healthcare organizations analyze reimbursement data and negotiate stronger contracts.

A scorecard nobody reads is just a spreadsheet

Here is the unglamorous part. The scorecard does not create value. The meeting does.

Same six metrics. Same room. Every month. Finance, operations, revenue cycle, and at least one physician leader who can explain why documentation looks the way it does. Trend lines, not snapshots, because a single month of anything in healthcare billing is noise.

The March 2024 MGMA Stat poll is the case for the discipline. Sixty percent (60%) of medical group leaders reported denial rates rising, and only eleven percent (11%) had managed to bring them down. Everybody knew denials were a problem. Very few had a standing process for doing something about it. The gap between those two groups is a recurring calendar invite.

Four quarters of that and the conversation changes. People stop arguing about whose number is right and start arguing about what to do next, which is the only argument worth having.

Better decisions come from better questions

A good scorecard does not just report. It tees up the questions leadership should be asking anyway. Which payer is costing us the most in ways that never show up on the rate sheet. Where is the leakage concentrated. Did the renegotiation we finished eight months ago actually produce what we modeled, or did we just feel good in the meeting.

That last one matters more than most groups admit. Plenty of practices sign an improved contract and never verify the improvement landed. Loading the new rates and watching net reimbursement per encounter move is how you find out. Sometimes it does not move at all, and that is a conversation worth having with the payer while the ink is still fresh.

How Accretive Health Advisors Can Help

Accretive Health Advisors builds these scorecards for physician groups, specialty platforms, and health systems, and then uses them. We load your contracts, quantify what you are owed versus what you were paid, benchmark your rates against the market, and take that analysis directly into negotiations on your behalf. Our team spent decades on the payer side, so we know which numbers actually move a negotiation and which ones get politely acknowledged and ignored.

If you do not know your net collection rate, your underpayment exposure, or which of your contracts is quietly your worst, that is not a criticism. It is just an opportunity nobody has gotten to yet. Contact Accretive Health Advisors and we will build the scorecard, find the money, and go collect it.