The eval auth that expires mid-plan-of-care denies every visit after it.
Money leaks across your locations where auth windows, visit limits, and MCO rules sit in systems that do not talk to each other. Relay builds AI employees wired into the systems you already run. The work catches it overnight, before the denial lands, all HIPAA-compliant.
See where you're leakingWhere your pediatric occupational therapy clinic is leaking revenue.
Authorization windows, visit-count limits, and MCO-specific rules sit scattered across CentralReach, WebPT, SimplePractice, and a biller's spreadsheet. Multiply that by every location you run. When an eval auth expires mid-plan-of-care, the denial lands weeks later and nobody saw it coming. AI employees pull every expiration date, every remaining-visit count, and every pending credentialing gap into one picture. You can act on it before the visit, fully HIPAA-compliant.
AI employees go to work inside your back office.
In the first 2 to 3 weeks we map your operation and build your first working AI employee inside your stack, so you see it before we build the rest. Then our engineers embed in your clinic. They build AI employees inside the systems you already run. Prior auth renewals, eval auth expirations, MCO-specific visit limits, and overnight billing reconciliation stop falling through the cracks. Anything that matters routes back to your staff for approval first.
Auth expirations stop slipping overnight
An AI employee sweeps every active authorization each night. It matches each one against scheduled visits. Then it surfaces anything that expires before the next appointment. Eval auths for 97165 through 97168 and treatment auths for 97110 and 97530 are tracked separately, because payers treat them differently. Staff see a reviewed queue in the morning. The ones that need action are already drafted.
Visit limits close their own loop
When a patient nears the authorized unit count on an MCO plan, the AI employee flags it. It checks whether a renewal request has gone out, and drafts one if it has not. At two or more locations, this runs at the same time across every active patient. So a silent overage does not become a batch of denied visits. A staff member approves before anything leaves the building.
How we build it.
We start from the problem you feel, then build the fix on the systems you already run. Discovery and your first working AI employee take 2 to 3 weeks. The full build runs 8 to 12 weeks.
Start with a free 30-minute call
A short call about where the work is piling up and what that is costing you while it stays manual. No commitment, and you leave knowing where you would start.
Discovery and your first AI employee (weeks 1 to 3)
A few working sessions with your team. We map your operation end to end, every workflow across your locations, and find where the money leaks and what closing it is worth. You do not walk away with just a document. By the end of discovery we have built your first working AI employee on top of the systems you already run, so you see it pay off in your real setup before the full build starts.
The full build (8 to 12 weeks, start to finish)
We build the rest of the AI employees you mapped and wire them across every location. Nothing goes out until your team approves it, so you stay in control the whole way. One pediatric therapy client had all seven locations live within 90 days.
Common questions.
Does occupational therapy require prior authorization?
Yes. Most major payers require prior authorization for OT, and the rules vary by payer and plan. Evaluation codes 97165 through 97168 each carry their own auth requirements. Treatment codes like 97110 and 97530 carry visit-count limits that reset on MCO timelines. When a multi-location practice tracks all of this by hand, authorizations expire silently. The denials follow weeks later.
What is a good claim denial rate for an OT practice?
Most well-run OT practices target a denial rate under 5%. The gap almost always traces to auth lapses, missing modifiers like the GO modifier, or MCO-specific rule variations. Billing software does not catch these on its own.
Why do OT claims get denied?
The most common causes are expired or missing prior authorizations, missing modifiers, exceeding the authorized visit count, and documentation that does not match the coded complexity of the evaluation. The GO modifier is required on every OT claim line. Each cause is detectable before a claim goes out. That is exactly where Relay's AI employees work.
What happens when a prior authorization expires mid-treatment?
Every visit billed after the expiration date denies, often in a batch weeks after the sessions occurred. By then the window to appeal may have closed and the revenue is gone. The fix is catching expirations before the visit. AI employees run that sweep overnight across every location and active patient.
Can one system handle billing across multiple OT clinic locations?
Yes. Relay builds custom AI employees that sit between the platforms you already run, such as WebPT, CentralReach, and SimplePractice. They automate billing rules across every location, and your existing systems stay in place. One pediatric therapy client had all seven of its locations running on a custom Relay build within 90 days.
What is the therapy cap for Medicare Part B OT billing in 2026?
The KX modifier threshold for 2026 is $2,480 for OT services. Above it, Medicare requires the KX modifier on each OT claim line to document medical necessity. It is not a hard cap, but a missing modifier triggers an automatic denial. At volume across multiple locations, that is a systematic exposure the AI employees track automatically.
Find the leak before the next denial batch does.
It's a free 30-minute intro call. We'll show you where the revenue is leaking and walk through a couple of workflows where AI closes the gap. At Sensory Speech and OT, the build turned into 3 formal referral partnerships generating $24,000 a month in recurring revenue. The AI employees reconcile billing there overnight. Bring your payer mix and your biggest back-office headache.