The Reservoir: Why Your Marketing Budget Is the Wrong Question
A practice owner asked me recently how much they should be investing in marketing to bring in new patients.
Fair question. Wrong first question.
That practice was losing an average of $42,000 a week to provider cancellations and mismanaged schedules. Roughly $2.2 million a year in authorized, approved, payable revenue that never became a session.
Adding new clients to that system would have added more authorized revenue to leak.
The Reservoir
At the top of every ABA revenue funnel sits something most practices never measure. I call it the Reservoir: revenue you are already entitled to, that the patient could actually receive, and that has not entered the funnel yet.
Two forms:
Deliverable units that never get scheduled. The payer approved them. The patient can tolerate them. The family has the availability. The calendar never claimed them, and then the authorization period ends.
Whole service lines that never get delivered. Parent training, supervision, reassessment. Authorized, budgeted by the payer, and quietly not happening.
This is the cheapest leak in the entire business to fix. There is no denial to appeal. No underpayment to chase. No payer to argue with. The money was already approved. It just needs to get on a calendar and get delivered.
Why nobody sees it
Every other leak in a practice announces itself.
A denial shows up as a rejection. A recoupment shows up as money leaving the bank. An underpayment shows up when someone compares the remittance to the contract.
Unused authorization produces no artifact at all. Nothing gets rejected, because nothing was ever submitted. Nothing was submitted, because nothing was ever delivered. Nothing was delivered, because nothing was ever scheduled.
The loss is invisible in the accounting records. You cannot find it in QuickBooks, because it is not a transaction. It is the absence of one.
That is precisely why it grows.
Measure against what is actually deliverable
This is the part most utilization advice gets wrong, and getting it wrong is why these numbers usually get ignored.
Authorized is not the same as deliverable.
A payer may approve 25 hours a week. The patient may only tolerate 15. The family may only have availability for 18 once you work around school, siblings, and two working parents. Pushing hours a child cannot handle in order to satisfy a utilization number is bad clinical practice, and it is not what this measurement is for.
So the denominator is not authorized units. It is schedulable capacity: what this patient can realistically receive, given clinical tolerance and actual availability.
Three numbers, not two:
Authorized capacity. What the payer approved.
Schedulable capacity. What the patient can realistically receive. Set by clinical judgment and family availability, not by the authorization letter.
Delivered. What actually happened.
True leakage is schedulable capacity minus delivered. That is the gap a scheduler or clinical director can actually close, which is exactly why it is the number worth reporting.
The other gap, authorized minus schedulable, is not leakage. It is information. If a large share of an authorization is not schedulable, that is worth raising at reassessment, with the payer, or with the family. It is a planning conversation, not an operational failure. Sometimes it means the authorization was written to a level the family's life cannot absorb, and that is useful to know before the next auth period.
Measuring against raw authorization produces a large, alarming number that clinical will dismiss in the first meeting, correctly, because it includes hours nobody could have delivered. A metric that gets dismissed once never gets looked at again. Measure what is truly possible to deliver and the number survives contact with the people who have to act on it.
The two numbers to start with
Auth Burn. Units delivered divided by schedulable capacity. A reasonable starting target is 85 to 95 percent. Below that, you are leaving deliverable revenue in the reservoir.
Service Line Coverage. Service types delivered divided by service types authorized, with a documented reason for any line not being delivered. Parent training and supervision going undelivered is usually an operational gap rather than a tolerance issue, but if the family declined, document it and move on.
Treat these as starting points, not gospel. Replace them with your own baselines once you have a few periods of real data. A target you did not measure your way into is just someone else's number.
The raw data is visible in CentralReach under Billing, then Authorizations, in the Worked, Remaining, and Utilization columns. Schedulable capacity is not in there. It comes from clinical, and it has to be captured somewhere consistent, because without it you are measuring against the wrong denominator.
The second leak: provider cancellations
Sitting right below the Reservoir is the leak that made up most of that $42,000 a week.
A cancellation is a booked session that does not happen. Split them by who caused it, because the economics are completely different.
A client cancellation costs you the revenue.
A provider cancellation costs you the revenue and, because clinical staff are hourly and often still get paid or backfilled, can cost you labor on top of it. You paid for capacity that produced nothing. It hits cost of services and it hits utilization at the same time.
A reasonable starting target is under 5 percent provider-driven. Track it separately from client cancellations, and size it in dollars, not percentages. Units times rate. A percentage gets nodded at. A dollar figure gets acted on.
Then track the recovery rate. A cancelled session that gets rescheduled inside the authorization period is a delay. One that does not is a loss. Those are not the same event and should not be counted the same way.
Back to the marketing question
Here is the arithmetic that owner had not run.
New client revenue costs money to acquire. Marketing spend, intake time, authorization work, scheduling capacity, onboarding. Every new client carries an acquisition cost and a ramp.
Recovered authorization revenue costs none of that. The client is already enrolled. The authorization already exists. The rate is already contracted. The only input is management attention.
So before spending a dollar on acquisition, the question is: how many deliverable hours are we already sitting on, and what would it take to schedule them?
If a practice is delivering 70 percent of its schedulable capacity, getting to 90 recovers a fifth of the revenue it could already be earning. No marketing spend, no new clinicians, no payer negotiation, and no clinical compromise, because by definition those hours were always deliverable. That is the highest-return work available to most ABA practices, and it is almost always the work nobody owns.
Marketing is the right investment eventually. It is the wrong investment while the reservoir is leaking, because all it does is increase the volume flowing into a system that is losing a fifth of it.
Fill the bucket before you turn up the tap.
What this looks like for each audience
For owners: one question. What percentage of what we could actually deliver are we delivering?
For investors: auth burn against schedulable capacity, and provider cancellation rate, are capacity-utilization metrics. They tell you whether growth can come from operations or only from acquisition, and operational growth is far cheaper.
For operators: a weekly list. Which authorizations are burning too slowly and expire soonest, which service lines are not being delivered at all, which providers are driving cancellations.
Do this, in this order
Capture schedulable capacity per client, from clinical. Without this you have no valid denominator.
Pull auth burn against that number. Sort by authorization end date, soonest first. Those are your expiring dollars.
Pull service line coverage. Find authorized lines producing zero sessions.
Split cancellations by provider-driven and client-driven, and size the provider side in dollars.
Give the numbers an owner. Scheduling and clinical book the sessions. Finance surfaces the number. Neither works without the other.
Separately, review where authorized and schedulable are far apart. That is a reassessment conversation, not an operations one.
Only then, price your marketing.
The cheapest revenue in your practice is revenue you are already approved for and the patient can already receive. It is sitting in the reservoir waiting for someone to look.
