Credentialing level is where the money goes
A BCBA supervises for two hours. The claim goes out at the RBT rate.
It pays. Full balance, no denial, no short payment, no flag anywhere. It will keep paying, at that rate, until somebody opens the contract and the system side by side — which, in most practices, is never, because there is no reason to. Nothing is broken. That is the whole trick.
Why this one is different from a typo
Most rate errors are one-offs. A single claim keyed wrong, caught at reconciliation, corrected, done.
A credentialing mapping error is not a one-off. It is a rule, and rules are tireless. Once a service line is pointed at the wrong level in your fee schedule, every claim generated from that line inherits the error automatically, and nothing downstream questions it. The claim is internally consistent. It matches what your system believes the rate to be. The payer adjudicates against its own schedule, pays what it owes, and moves on.
You are not underpaid. You are underbilled. The payer did nothing wrong.
That distinction matters for recovery, and we will come back to it.
Where the mapping breaks
ABA prices by who delivered the service. BCBA, BCaBA and RBT carry materially different rates under the same contract, and the codes that can be delivered by more than one level are where the mapping goes wrong.
97155 is the usual offender. Protocol modification can be delivered by a BCBA or, in some contracts and states, by a BCaBA under supervision. If the service is mapped once at the lower level and then delivered predominantly by the higher, every hour of that work bills short.
97153 goes the other way. Direct treatment is an RBT service almost all the time, but when a BCBA covers a session — a call-out, a new client, a difficult transition — the entry can carry the BCBA’s credential while the rate stays at the RBT default. Now you are billing under on a service you delivered at a higher cost.
Supervision codes are the third. 97156 and 97157 are delivered by the analyst, priced accordingly, and routinely mapped to a generic default when a new payer is loaded.
None of these are visible on a claim. They only show up when you put the service line’s loaded rate next to the contracted rate for that credential — which is not a report anybody ships, because it is a report about a thing that is supposed to be true.
What it costs
One payer, trailing twelve months, illustrative.
Illustrative arithmetic, not a client engagement.
The shape of this is different from the cents-per-unit drift in the last chapter. A credentialing error is not a rounding difference. It is the gap between two price tiers, which is usually dollars rather than cents, and it lands on the codes with the highest rates rather than the highest volume.
A rate that drifts costs you a little on everything. A credentialing error costs you a lot on one thing, and the one thing is usually the code your best-paid people spend their days on.
The recovery problem
Here is the part nobody enjoys.
When a payer underpays against your contracted rate, you have a dispute. There is a window, it is usually 90 to 180 days from remittance, and inside that window you can challenge the adjudicated amount and generally be made whole.
When you underbill because your own system had the wrong rate loaded, there is no dispute. The payer paid what you asked for. Recovery means corrected claims, and corrected claims are bounded by timely filing, not by the dispute window — which is usually shorter, and which has already been running since the date of service.
So the recoverable portion of a credentialing error is almost always smaller than the total. On a variance that has run eleven months, expect to recover the most recent quarter and to write off the rest as the cost of finding out.
That is not a reason to skip the work. Fixing the mapping stops the bleed on the next claim, and the next claim is where most of the money is. You are not doing this to recover the past. You are doing it so there is nothing to recover next year.
The audit is a table, not an investigation
Good news: nobody has to cooperate with you. No payer call, no biller meeting, no asking anyone to explain themselves. Two documents and an hour.
On one side, the contracted rate by code, by credential, from the current fee schedule. On the other, the loaded rate by code, by credential, from your practice management system. Put them in the same sheet and subtract.
Every non-zero row is a finding. Multiply each by trailing twelve-month volume for that combination and you have the annual number, ranked.
Then check the direction. Anything paying above contract is not a win — it is a recoupment scheduled for a date you do not control, and it belongs in chapter 08.
Five checks you can run this week
Start with your largest payer by volume. You do not need help with any of these.
1. Produce the current fee schedule, by credential. Current, not the one attached to the original agreement. If it takes more than ten minutes to find, that is the first finding and it is common.
2. Export your loaded rates by code and credential from the practice management system. Not what claims paid — what the system thinks it should bill.
3. Subtract, and rank by annual dollars, not by variance per unit. The largest per-unit gap is rarely the largest loss.
4. Pull delivered units by code and by rendering credential for the trailing twelve months, and check that the credential on the claim matches the credential that actually delivered the session. A mismatch here is a mapping error even when the rates are right.
5. Check both directions. List anything paying above contract separately. That is not a finding to celebrate, it is a liability you have not booked.
If any row on that sheet is larger than a month of one clinician’s wages, it is worth fixing this week rather than at close.
The number to actually track
Loaded rate against contracted rate, by code and by credential, with every non-zero row ranked by annual dollars.
Pair it with delivered units by rendering credential, so a mapping error shows up even where the rates themselves are right.
Ranking by variance per unit will point you at the wrong row. The largest per-unit gap is rarely the largest loss.
Where this sits
This is the seventh of eight stages where ABA revenue leaves. It is the quietest one, because nothing denies and nothing ages. The claim pays, at the wrong number, forever.
Nothing in this stage denies, ages or short-pays. The claim is internally consistent and the payer owes exactly what you asked.
A Leak Map Assessment measures all eight — the reservoir, authorization burn, cancellations, session conversion, conversion to billed, clean claims, rate integrity and recoupments. A claims audit is a different instrument. It starts once a session has been converted and billed, and reconciles the back half of the cycle to cash.
Aimline closes the books, oversees the revenue cycle, and sits in the CFO seat for ABA practices. We run claims audits scoped to a payer book, and Leak Map Assessments that measure the full cycle — three weeks, $7,500 fixed. Either one gives us a verified number to run oversight against, and the report is yours whether or not you engage us to fix what it finds.
