Radiology Billing Software Pricing: What Imaging Centers Pay Per Study

The short answer: outsourced radiology billing is priced as a percentage of collections, quoted in the market at roughly 5.5% to 10% for hospital-based groups billing the professional component only and 2% to 5% for imaging centers billing globally. On a 30,000-study outpatient center that works out to somewhere near $4 to $5.50 per study. The fee is not what decides the outcome. Using Medicare's own 2024 national figures, about 20% of the allowed amount on a normal outpatient imaging mix is patient coinsurance and deductible, and the spread between a vendor that collects most of that and one that does not is several times larger than the spread between a 4.5% fee and a 6% fee.

Billing proposals are unusually hard to compare because the two models are quoted in different units. Software comes with a per-user or per-provider monthly price. A billing company quotes a percentage of what it collects. Put them side by side and the software looks cheap, right up until you add the payroll that has to sit behind it.

The way out is to convert everything to dollars per study and then check the assumption underneath the conversion, which is how much money each option actually brings in. That second part is where the decision really lives, and it is almost never in the proposal.

How much do radiology billing companies charge?

Almost all of them charge a percentage of collections rather than a flat fee, and the percentage depends on what you are billing rather than on how good the firm is. The ranges quoted publicly, including by AuntMinnie contributor Kyle Tucker in an October 19, 2017 piece that is still the most repeated benchmark in this market, are about 5.5% to 10% of collections for hospital-based groups billing the professional component only, and about 2% to 5% for imaging centers billing globally.

That gap looks unfair and is not. A global claim covers both the technical component, meaning the equipment and technologist time, and the professional component, meaning the radiologist's read. It brings in roughly twice the money for a similar amount of billing work. Medicare's national data makes the size of the gap concrete: for MRI of the lumbar spine without contrast, CPT 72148, the average allowed amount in 2024 was $68.51 in a facility setting against $144.78 in an office setting. Same code, same read, more than double the allowed amount, so a smaller percentage still pays the firm more per claim.

If a proposal quotes you a percentage without first asking whether you bill globally or professionally only, that is a signal about how much radiology work the firm actually does.

How much does radiology billing software cost?

Nobody in this category publishes a rate card either, and the reason is that the license is a minority of the cost. Products shortlisted by United States imaging practices, including ImagineSoftware, AdvancedMD, Tebra, athenahealth, DrChrono and eClinicalWorks, are generally priced per provider or per user per month, with clearinghouse fees charged separately per claim or per batch. Expect the vendor to quote you a range and then scope it.

The number that matters more is the payroll. The staffing rule of thumb in the same AuntMinnie piece is that a billing office should be staffed at about 60% of the FTE radiologist count. A center served by five full-time-equivalent radiologists is therefore looking at roughly three billing FTEs. At the Bureau of Labor Statistics median annual wage for medical records specialists, $51,140 as of the May 2025 OEWS survey, that is about $153,420 in base wages before benefits, payroll taxes, software or clearinghouse fees.

Three FTEs at $153,420 in base wages against 30,000 studies a year is $5.11 per study, and that is the floor rather than the total. Hold that number, because it is what the percentage models have to beat.

What does that work out to per study?

Here is the whole calculation on one illustrative outpatient center, built entirely from Medicare's published national figures for calendar year 2024 so you can reproduce every line. The center reads 30,000 studies a year and bills globally, which is why every allowed amount below is the office place-of-service figure.

Now price the two models against that $2,710,530.

• Outsourced at 4.5% of collections: $121,974 a year, or $4.07 per study
• Outsourced at 6.0% of collections: $162,632 a year, or $5.42 per study
• In-house base wages alone, three FTEs: $153,420 a year, or $5.11 per study, before benefits, taxes, software and clearinghouse fees

Two things fall out of that. First, at this volume in-house is not the cheap option, which surprises most people running the comparison for the first time. Second, the entire distance between a 4.5% quote and a 6% quote is $40,658 a year, or $1.35 per study. Keep that figure in view, because the next section is about a number roughly three times larger that appears in neither proposal.

The number that beats the fee

Medicare does not pay the full allowed amount. It pays its share, and the rest is beneficiary coinsurance and deductible that the practice has to collect from a person rather than from a payer. Run that split across the same mix, using the 2024 paid-versus-allowed figures for each code, and the patient responsibility comes to $558,240 a year, or 20.6% of allowed revenue.

Two codes in the table break the pattern in a way worth knowing. On screening mammography (77067) and low-dose CT lung cancer screening (71271), the average allowed amount and the average Medicare paid amount are identical, because cost sharing is waived on those preventive services. Every other line carries roughly a quarter of its value in patient responsibility.

So: suppose one billing operation collects 85% of that $558,240 and another collects 60%. The difference is $139,560 a year. That is 3.4 times the entire gap between a 4.5% fee and a 6% fee, and it is invisible in every proposal that leads with the percentage.

This is why patient-balance performance is the question to open with rather than close with. Ask each candidate for their patient collection rate as a percentage of patient responsibility billed, ask how many statements go out before an account moves to a different process, and ask whether balances are pursued by text and email or only by mailed paper. The mechanics of chasing an outstanding balance by email and SMS until it is paid are the same in imaging as in any other business, and they are the difference between a 60% number and an 85% one. If a firm will not give you the figure, that is itself the answer.

Is in-house radiology billing cheaper?

At the volume above, no, not on cost alone. It becomes competitive when claim volume rises far enough that the fixed billing team is spread over many more studies, because payroll does not scale with collections and a percentage fee does. A practice collecting $8 million on the same three-FTE team is paying about $1.92 per study in base wages against $360,000 at 4.5% outsourced. That crossover is the real argument for bringing billing in-house, and it arrives at scale rather than at a particular feature.

The arguments that hold at any size are different ones: the accounts receivable and the data stay under your control, you are not negotiating a runout when you change vendors, and denial patterns are visible to you directly rather than through a monthly report. Those are governance reasons, and they are legitimate. They are just not cost reasons.

The case for outsourcing that survives scrutiny is narrower and more practical. If you cannot hire and keep a certified radiology coder, if a single resignation would take your billing knowledge with it, or if patient balances are visibly aging, a firm with a rules engine and a collections process will beat anything you buy. The four operating models and how each one fails are compared in detail on the radiology revenue cycle management and billing software comparison.

What the 2026 fee schedule does to all of this

CMS finalized the CY2026 Medicare Physician Fee Schedule on October 31, 2025, with a conversion factor of $33.4009 for clinicians outside advanced alternative payment models and $33.5675 for qualifying APM participants. Both are increases. The specialty picture is not: the American College of Radiology reports estimated impacts of minus 2% for diagnostic radiology, minus 1% for nuclear medicine and radiation oncology, and plus 2% for interventional radiology, because an efficiency adjustment reduces work RVUs on non-time-based services and most diagnostic imaging sits in that bucket.

A higher conversion factor with lower work RVUs can still mean less money per study, which tightens the arithmetic above in both directions. It makes a percentage fee slightly cheaper in absolute dollars and makes the patient-responsibility collection rate matter slightly more, since the collectable pool shrinks while the effort to collect it does not.

Where the report fits into the price

The cheapest denial is one that never happens, and a large share of radiology denials trace back to documentation rather than to billing. A study coded at a level the report does not support, a laterality that was never stated, a comparison that would have justified medical necessity but was not mentioned: none of those are billing errors, and no billing vendor at any fee can fix them after the fact.

That is the part of the revenue cycle Radiological.ai touches. It flags suspected findings for a second look, prioritizes the worklist and drafts the structured report into your own template, so what reaches a coder is complete and consistent between readers. It does not assign CPT codes, does not submit claims and does not touch accounts receivable, and the radiologist reviews, edits and signs every study. If report consistency is the gap, structured radiology reporting is the closer fit, and reimbursement for AI-assisted imaging specifically is covered in radiology AI CPT codes and reimbursement.

What to ask before you sign

Do the per-study conversion yourself with your own mix and your own payer split rather than accepting a headline percentage, then hold every candidate to the patient-responsibility question. The wider mechanics of renewal uplift and fully loaded first-year totals are in the contract terms worth negotiating, and radiology software pricing by category works through what the surrounding systems cost. If the schedule and charge capture upstream are the weak link rather than the billing itself, that decision is radiology information system software.

Sources: CMS, Medicare Physician and Other Practitioners by Geography and Service, national rows for calendar year 2024, released May 21, 2026 and pulled through the data.cms.gov API on August 25, 2026. Bureau of Labor Statistics, Occupational Employment and Wage Statistics, May 2025, medical records specialists (29-2072). American College of Radiology summary of the CY2026 Medicare Physician Fee Schedule final rule. Percentage-of-collections and staffing benchmarks as published by AuntMinnie, October 19, 2017. The illustrative study mix is our own construction and is labeled as such; substitute your own volumes and payer mix before using any of these figures in a decision.