Operating Theory · Sep 10, 2026
The Labor Curve
A rules-based operation has no mechanism to break the link between panel size and headcount. An AI-native one does, but the first two quarters of a migration cost more, not less, and the curve only bends where shadow-mode agreement supports widening autonomy.
Under the pillar Zero Dashboard Experiences
Thesis
The reason to move is that a rules-based operation cannot decouple labor from panel size, and the decoupling is the whole economic argument.
The argument
In a rules-based care operation, staff hours per thousand enrolled patients is roughly flat, so total labor rises with the panel. Every efficiency available inside that architecture, better templates, better queues, better scripts, moves the constant slightly and does not change the relationship. Value-based contracts require the opposite: margin that improves as the panel grows.
An AI-native operation can change the relationship, but only through autonomy, and autonomy is earned slowly. Implementation, parallel running, shadow mode and supervised operation all consume labor before returning any. The honest shape of the curve is a rise, then a plateau, then a bend, and the bend arrives only where measured agreement supports widening autonomy. If agreement stalls, the curve stays flat and the investment has bought reliability rather than cost.
This is also the argument's biggest exposure. An organization that budgets for a fall in the first quarter will cut the program in the third, before anything it paid for has had a chance to pay back.
What a legacy vendor would say
That their platform delivers documented efficiency gains today without a migration; that a curve that gets worse before it gets better is an unfundable proposition for a mid-sized group; and that the bend is asserted from models rather than observed in production.
The last point is currently correct and should be conceded plainly. The bend is a hypothesis. Nobody in this line of research has published a multi-year production labor curve, and the argument stands or falls on someone doing so.
What would settle it
Staff hours per thousand enrolled patients per month, tracked across a full migration and beyond, with panel size, acuity mix and autonomy level recorded alongside. Two years of that series from a real operation would settle this question more decisively than any argument in this pillar.
Open questions
Where the trough sits and how deep it is. Whether the bend is a property of the technology or of the operation's willingness to widen autonomy. Whether labor released is genuinely released or reappears as supervision, review and exception handling.
Put this to work
The Labor Curve challenge record
Test the claim behind The Labor Curve against a real case and look for where it fails.
- For
- Clinical, operational, product, reimbursement, and implementation leaders.
- What you keep
- A the labor curve challenge record you can review, revise, and send.
- What counts as sound
- Makes the claim testable
- Includes contrary evidence
- Preserves competing explanations
- States uncertainty
- Names what would change the conclusion
Clinical, regulatory, privacy, and release decisions remain with qualified people.
Nothing entered here is stored or sent. Review the prompt before sharing confidential, personal, patient, or privileged information.
Review the prompt
You can leave any field blank. The prompt will mark it as not provided.
If the record survives your review, send the question, evidence, unknowns, and requested next step.