We do not sell revenue cycle management. We are not a billing company, we do not resell RCM software, and we have no referral arrangement with anyone who does. That is worth saying at the top, because most of what gets written about outsourcing revenue cycle management is written by people who sell it.
Clinic owners call us at a specific moment: they have decided their collections are worse than they should be, a vendor has quoted them a percentage of collections, and they want someone without a stake in the outcome to tell them whether the number makes sense.
Usually it does not, at least not for the reason they think.
The fee is the least interesting number
Every RCM conversation starts with the fee. Six percent of collections. Seven. Four and a half if you commit to three years. Owners compare quotes on that number and pick the low one.
The fee only matters relative to what you already spend. If your in-house billing costs 5% of collections once you count salaries, benefits, clearinghouse fees, software, and the portion of your practice manager's week that disappears into claims, then a 6% vendor is not 6% more expensive. It is 1% more expensive.
That reframes the whole decision. Here is what a vendor has to deliver just to break even:
| In-house billing cost | 4% fee | 5% fee | 6% fee | 7% fee | 8% fee | 9% fee |
|---|---|---|---|---|---|---|
| 2% of collections | 2.1% | 3.2% | 4.3% | 5.4% | 6.5% | 7.7% |
| 3% of collections | 1.0% | 2.1% | 3.2% | 4.3% | 5.4% | 6.6% |
| 4% of collections | 0.0% | 1.1% | 2.1% | 3.2% | 4.3% | 5.5% |
| 5% of collections | −1.0% | 0.0% | 1.1% | 2.2% | 3.3% | 4.4% |
| 6% of collections | −2.1% | −1.1% | 0.0% | 1.1% | 2.2% | 3.3% |
Each cell is the increase in collections the vendor must produce for you to end up level. The arithmetic is (1 − your cost) ÷ (1 − their fee) − 1, and you can check it on the back of an envelope.
Read the table twice. A practice spending 5% in-house that signs a 6% vendor needs a 1.1% lift in collections to break even. On $5M that is $55,000, and a competent billing operation should clear that on denial rework alone. A practice spending 3% in-house that signs an 8% vendor needs 5.4%, which is a different animal entirely. Same two vendors, same two practices, completely different decisions.
Negative cells mean you are paying more in-house than the vendor charges. If you land there, the vendor could collect slightly less than your current team and you would still come out ahead on cost. That happens more often than owners expect, usually at practices with one overqualified biller and not enough volume to keep them busy.
Nobody knows their in-house number
Which is the actual problem. Ask a clinic owner what billing costs them and you get a salary. The salary is maybe 60% of it.
Count all of it: biller and coder salaries plus benefits, the practice manager's time on claims and appeals, clearinghouse fees, the billing module of your PM system, statements and postage, collection agency fees, and the coverage cost when your biller takes vacation and the AR sits still for two weeks.
Then divide by annual net collections. We are deliberately not going to tell you what the answer "should" be, because the range across independent practices is wide enough that a published average would only mislead you. Your number is your number, and it is the one the vendor quote has to beat.
You cannot evaluate a quote without it. You are comparing a known price against an unknown one and calling it a decision.
Measure three things first
If you take one thing from this: do the measurement before you shop, not after. Vendors will run a "free revenue analysis" for you. It is a sales instrument. It is designed to find a gap, and there is always a gap.
Measure these yourself:
Net collection rate. What you collected against what you were contractually entitled to collect, over a rolling twelve months with a three-month lag so claims have adjudicated. This is the single best summary of whether your revenue cycle works. Our net collection rate benchmarks page covers what good looks like and how to calculate it without fooling yourself.
Denial rate, and the reason mix. Measure it the way the industry defines it rather than the way your billing software reports it. HFMA's MAP Keys are the standard set of revenue cycle KPIs, and two of them matter here: AR-5, the remittance denial rate, which counts denied claims against remitted claims and focuses on actionable denials; and AR-6, denial write-offs as a percentage of net patient service revenue, which captures what you gave up after appeals were exhausted. A vendor quoting you a denial rate that matches neither definition is quoting you a number you cannot compare to anything. The rate alone tells you little in any case. The reasons tell you everything. Denials for eligibility and registration errors are a front-desk problem and no billing vendor will fix them, because they happen before the claim exists. Denials for coding, documentation, or medical necessity are a different failure with a different owner. Our denial code reference maps the common codes to the team that can actually prevent each one, and what a healthy denial rate looks like covers the thresholds.
Days in AR, split by payer and by bucket. A blended average hides everything. One slow payer or one aging bucket usually explains most of the problem, and that is often a contract or a workflow issue rather than a billing-effort issue.
Those three numbers tell you whether you have a collection problem or a process problem. Only one of them is for sale.
Automation compounds whatever process you already have
The pitch for RCM automation and AI-assisted coding is that it removes manual work from the claim lifecycle. That part is broadly true, and the tools have improved.
The catch is that automation is a multiplier, and it multiplies in both directions. If your front desk is inconsistent about verifying eligibility, automating downstream means you now submit clean-looking claims carrying bad eligibility data faster than before. The denials arrive faster too. We wrote about where this goes wrong in AI billing tools: what they save, what they miss.
The practices that get real value from automation are the ones whose process was already sound and whose constraint was throughput. The practices that get burned are the ones who bought automation to paper over a process nobody wanted to fix.
Ask any software vendor a simple question: which of my denial reasons does this actually prevent? If the answer is a list of features rather than a list of your denial codes, you are talking to someone who has not looked at your data.
The contract terms that matter more than the fee
Once you have decided the economics work, the risk moves into the paperwork. The clauses we see cause the most damage:
Who works the legacy AR. When you switch, someone has to chase the claims already in flight. If the new vendor will not touch them and your in-house biller has left, that AR quietly dies. It is usually the single largest cost of switching and it rarely appears in the quote.
What "collections" means in the fee. Is the percentage taken on net collections or gross charges? Does it apply to patient payments you collect at the front desk? Does it apply to capitation or value-based payments the vendor had nothing to do with? These definitions move the real price by a lot more than the headline rate.
Who owns the data. Your claims history, your denial history, your payer correspondence. If leaving means losing your own history, you are not really free to leave.
Term and exit. A three-year term at a lower rate is only cheaper if the vendor performs. Look for the exit terms and any performance floor. If there is no performance language at all, the low rate is buying you nothing but a longer commitment.
Who talks to patients. Outsourced statements and collection calls carry your name. That is a patient-experience decision as much as a financial one, and it deserves more thought than it usually gets.
When outsourcing is genuinely the right answer
It is often the right call, and the case is strongest at the small end. A one-to-three-provider practice usually cannot keep a specialty-trained biller fully occupied, cannot cover them when they leave, and cannot afford the concentration risk of a single person holding the entire revenue cycle in their head. We laid out the scale thresholds in outsourced versus in-house medical billing.
It also makes sense when you are growing faster than you can hire, when you have just added a specialty whose coding your team does not know, or when your biller is retiring and the knowledge is about to walk out of the building. If the person leaving sat above the billing team rather than in it, that is a different gap, and we set out the sequence for it under interim CFO.
What it does not do is fix a front desk that does not verify eligibility, a provider who documents thinly, or a payer contract negotiated badly in 2019. Those problems follow you to the new vendor, and six months later you conclude the vendor is failing when the vendor never had the ability to touch them.
The order of operations
Measure your three numbers. Add up what billing actually costs you today. Put your denial reasons in front of any vendor and make them tell you which ones they prevent. Then run the break-even table and decide whether the lift they are promising is one you believe.
If the honest answer is that most of your leakage happens before a claim is ever submitted, no vendor is going to sell you the fix. That is the finding we deliver most often, and it is the cheapest one to act on.
We look at this in the context of the 12 to 18% gap between billing and collections that shows up at most multi-location practices we review. Before you outsource the symptom, it is worth knowing which part of the gap is actually yours to close.
Sources
- HFMA, MAP Keys: industry-standard revenue cycle KPIs — definitions for AR-5 (remittance denial rate) and AR-6 (denial write-offs as a percentage of net patient service revenue)
- HFMA, 7 KPIs providers should be tracking
- MGMA, Practice operations data and benchmarking resources
The break-even table is arithmetic, not a benchmark: required lift = (1 − your in-house cost as a share of collections) ÷ (1 − vendor fee) − 1. Substitute your own two numbers and check it. The 12 to 18% billing-to-collections range reflects what we have found across the practices we have reviewed, not a published industry figure.
