A quoted risk adjustment ROI describes a vendor’s client list, not their platform. Here is where the return actually separates, and why it matters.
Ask ten risk adjustment vendors what their return on investment is and you will get ten ratios. Five to one. Seven to one. Occasionally something bolder, usually with a footnote.
Every one of those ratios has the same thing behind it. It is an average of what happened at other organizations. It describes the clients that vendor happened to sign. It does not describe the software.
That distinction disappears in most evaluations, and it is expensive. Two organizations can license the same platform, staff comparable coding teams, follow the same rules, and finish the year far apart. The distance between them was set before anyone opened a chart.
A quoted ratio is a fact about the client list, not the product
A published ROI is the output of a portfolio. It blends organizations with complete, reconciled data feeds together with organizations carrying three years of supplemental files nobody has matched. It blends Medicare Advantage books with Medicaid books. It blends first-year programs, where the easy capture is still sitting there, with mature programs that already took it.
When a vendor reports the average of that portfolio, what they are reporting is the composition of their client list. A vendor with a favorable mix will post a better ratio than a competitor with better technology and harder accounts.
So the comparison most buyers run, ratio against ratio, is not a comparison of capability. It is a comparison of who each vendor sold to.
Where the return actually separates
We run coding and encounter operations across Medicare Advantage, Medicaid, and commercial populations, for health plans, health systems, IPAs, and provider groups. That vantage point makes it possible to test something the industry rarely gets to look at directly. How much of the spread in return is explained by which population you serve, and how much is explained by the state of the data underneath it?
That question says something uncomfortable about how this market talks about itself. The line of business is treated as the main variable in every ROI conversation. It may not be.
Want to see where your own numbers land on this?
The peer group you benchmark against is probably the wrong one
The default comparison is line of business and membership size. A regional Medicare Advantage plan looks at other regional Medicare Advantage plans. A provider group in a risk arrangement looks at groups of similar size.
That grouping is convenient because it is the only one the industry publishes. It also assumes the population is what drives the difference.
The organization whose results actually predict yours is the one whose data looks like yours. Same feed completeness. Same reconciliation state. Same share of submitted work that gets accepted. Those organizations are almost never the ones you are being compared to on a conference panel.
The revenue number and the audit number come from the same place
There is a second question sitting next to the ROI question this year, and it is the one people actually lose sleep over. Does the number survive an audit?
These two questions share a root. A condition with a broken trail behind it is both revenue you did not collect and exposure you cannot defend. When RADV asks for support, the weak points that surface are the same ones that were suppressing the return all year.
Which means an ROI conversation that never touches the data layer is also an audit conversation that never happened.
What we will and will not tell you about our own numbers
We will not quote you a ratio in a first meeting. Not because the number is bad, but because a number given without seeing your population mix, your data sources, and your current acceptance picture is a guess with a decimal point on it.
What we will do is tell you where you are likely sitting and why, and be specific about what would have to change to move you.
Our platform is built around the data layer for that reason. Risk Analytics shows where conditions concentrate and where the underlying data is thin. Coder Workbench puts the full clinical picture in front of the coder in one place. Encounter Submissions tracks what was submitted, what was accepted, and what has to be corrected and resubmitted.
See where you land
If you want a real number instead of an industry average, it starts with your data. We will walk through your population mix, your feeds, and your acceptance picture, and show you what the return looks like from there.
Frequently asked questions
What is a good ROI for risk adjustment?
There is no single benchmark. Published ratios are portfolio averages that reflect the mix of clients a vendor signed, including their populations and the state of their data. A useful estimate has to start from your own population mix, feed completeness, and submission acceptance rather than from someone else’s average.
How is risk adjustment ROI calculated?
Program cost is compared against revenue tied to newly captured or corrected conditions. The variable that gets least attention is the denominator. Conditions that never reach the system, or that fail at submission, never enter the calculation, which makes a constrained program look more efficient than it is.
Why do risk adjustment ROI estimates vary so much between vendors?
Because the ratio reflects the vendor’s client list. A portfolio weighted toward mature programs with clean data will produce a different average than one weighted toward first-year programs or unreconciled feeds. Two vendors can quote very different numbers while offering comparable capability.
Does risk adjustment ROI differ between Medicare Advantage and Medicaid?
Yes. The models, revenue mechanics, and member relationships differ, and so do chronic condition prevalence and documentation frequency. Population is a real variable. The open question is whether it explains more of the difference in return than the state of the data underneath each program.
What is risk adjustment ROI for commercial ACA plans?
Commercial ACA risk adjustment is budget neutral, so returns depend on how a population scores relative to other plans in the same market rather than on absolute capture. That changes the calculation substantially compared with Medicare Advantage, where capture translates more directly into revenue.
Does data quality affect risk adjustment revenue?
It sets the size of the opportunity. Missing feeds, mismatched identifiers, and unreconciled supplemental files reduce the pool of conditions available to work. Coding performance determines how much of that pool gets captured, which makes data the first constraint and coding the second.
How should I benchmark my risk adjustment program?
Start with HCC prevalence in your own population and benchmark against your prior year case mix. Performance in ACA populations, or anywhere member churn is high, deserves particular attention. Provider level documentation gaps are also critical to measure. Those inputs are more predictive than comparing against other organizations by line of business and size.
What causes missed HCC capture?
The primary reasons are gaps in care coordination for high risk members, where the conditions exist but are not treated consistently, and provider documentation that lacks the required specificity. Transplants and amputations are common examples where documentation gaps appear. Captured conditions that fail at submission are a separate cause.
Do rejected encounters affect risk scores?
Yes. Diagnoses on an encounter that is not accepted do not count toward the risk score, so the coding work behind that encounter produces no revenue. Record-level visibility into acceptance is the only reliable way to know how much completed work actually landed.
Why do encounters get rejected?
Rejections generally trace to ingestion problems, mapping errors, reconciliation gaps, or failed handoffs between systems. Identifier mismatches, outdated code sets, and missing required fields are common examples. Most repeat on a pattern until someone has visibility into the pattern itself.
What does RADV audit defensibility mean?
Every submitted diagnosis can be traced to documentation that meets CMS requirements. Defensibility depends on the same trail that drives revenue, so a condition that cannot be traced to its source represents both uncollected revenue and audit exposure at the same time.
Can IPAs and provider groups measure risk adjustment ROI?
Yes, and the same variables apply. Provider organizations and IPAs in risk-bearing arrangements are exposed to feed completeness, reconciliation, and submission acceptance the same way plans are. The contractual relationships differ, but the inputs that set the return do not.
