The short version
A typical family practice, mapped process by process. AI does the work in the background, graded against what your team actually did. The dentistry stays human. Only the paperwork that passes gets switched on.
Includes a live graded run — real grades, made-up documents
42%
42% of the practice's staff hours are insurance paperwork, claims chasing, recall, and notes. We graded AI on that work — 90% right on its first run (Fig. 3).
The whole report in three lines
42% of practice hours are paperwork. AI can carry most of it. The dentistry: none of it.
We graded it: 90% right, half a cent per item. Real run, receipts below.
Nothing turns on until it beats your bar. Below the bar, it keeps practicing.
01 · Why now
Private-equity-backed dental groups are rolling up independent practices, and the hardest hire in the building keeps getting harder. The way through is the same either way: take the paperwork off the desks you already have — and make the practice worth more than the founder's chair.
Dentists in DSO groups
13%
And roughly double that among dentists under 35. Consolidation runs one direction.1
Hygienist recruiting
9 in 10
Practices recruiting hygienists call it very or extremely difficult.2
Dentists age 55+
~40%
A wave of practices changes hands this decade. Buyers pay for process, not a founder's chair.3
The pricing reality: solo practices typically change hands around 60–80% of one year's collections; DSO platforms are priced on profit multiples.4 The gap is transferable process — which is exactly what the measured scorecard below documents.
02 · Your week
Representative family practice — two dentists, hygiene, front office. In the engagement, every row is measured, not modeled.
Fig. 2 — Where the hours go · % of all staff hours, representative model
42% of all staff hours sit in the paper layer, where AI carries 35–75%. Chairside stays at 0% — the dentistry is not the automation target. The paperwork is. Nobody gets cut; the front office stops drowning and the chairs stay full.
03 · How we test it
AI redoes your practice's past paperwork. We grade it against what your team actually did. It touches nothing until it passes.
The answer key
200–500 pieces of your past work per task — EOBs you posted, claims you filed, recalls you sent. What your team did is the right answer.
Practice runs
AI works alongside your team on real work. Patients never see it. It just gets graded.
The bar
You set it — e.g. 98% right, and it earns the job. Below the bar, it keeps practicing.
04 · The report
We didn't model this section — we ran it. The documents are made up (no patient data, no PHI, ever). The grades, the costs, and the mistakes are real, scored case by case on the Recursiv platform.
Fig. 3a — The tape · every graded case from the live run, in order
Reading the misses: one sorting miss garbled its reply on a HIPAA form; the other filed a CBCT imaging report under LAB instead of REFERRAL — a genuine judgment miss. Both EOB misses were the grader cutting the answer off mid-number. The hand-backs were a smudged fax. Every one of these is exactly what the grading exists to catch.
Fig. 3b — The scorecard · bar: 98% to run alone, 90% to run with review
| The work | Right | Handed back | Cost each | Verdict |
|---|---|---|---|---|
| Sorting incoming documents20 docs: EOBs · referrals · cards · lab slips · traps | 90.0% | 0% | $0.005/doc | On, with review |
| Posting EOB numbers20 fields, graded one field at a time | 90.0% | 10% | $0.005/field | On, with review |
| Insurance verificationeligibility · frequencies · remaining maximums | — | — | — | Measured in pilot |
| Recall & no-show outreach draftsright = sent without edits | — | — | — | Measured in pilot |
| Denial appeal letters, first draftright = filed without material edit | — | — | — | Measured in pilot |
| Clinical note first draftsfrom dictation — always doctor-signed | — | — | — | Measured in pilot |
Right = matches the answer key · Handed back = gave it to a person instead of guessing. Neither task cleared the 98% run-alone bar on day one — that's the point of grading before trusting. The pilot fills in the remaining rows on your practice's work.
The point: nothing on this page asks for trust. Every row either has a grade or says it doesn't have one yet. A pilot runs thousands of cases on your practice's own work and re-grades every month.
05 · Your data
PHI lives under HIPAA. Recursiv runs on your own servers — the AI comes to the records, the records don't travel. Every action is logged for your HIPAA security documentation, and no vendor touches PHI without a BAA.
Fig. 4 — Where everything lives. Default: everything inside the green line. Outside AI services are a choice you make, behind a BAA and no-storage terms — never a default.
06 · The path
Who does what, hour by hour. Output: your version of Fig. 2 — real hours.
≈ 2 weeks
AI works alongside your team on real jobs. Patients never see it.
≈ 4–6 weeks
Your version of Fig. 3. What passed, what didn't — against your bar.
End of practice cycle
Behind review gates. Everything else keeps practicing until the numbers earn it.
Ongoing
What's real here: Fig. 3 rows 1–2 are an actual graded run (40 cases, July 29 2026) on made-up dental documents — no patient data, no PHI, ever. Fig. 2 is a representative model of a family practice, and the green shares there are estimates until a pilot measures them. Nothing here is legal or clinical advice. The engagement replaces every estimate with a measured number from your practice.
The next step
A 6–8 week pilot: we map your hours, AI practices on your actual paperwork — patients never see it — and you get this exact report with measured numbers. If the grades don't earn it, you switch nothing on.