end the day rate
End the day rate. Price Units of Work instead.
AI makes the day rate economically incoherent. Its replacement is the Unit of Work: a bounded, audited deliverable the client pays for by outcome, not by hour.
Why the day rate became inverse-aligned with AI productivity
The day rate made sense when professional output was roughly linear with time. An experienced consultant produced more per day than a junior one, and the market priced that differential in the daily fee.
AI breaks that linearity. A consultant who uses a well-configured agent to produce a technical audit in four hours has not delivered half a day of value. They have delivered the same audit, potentially with better coverage and fewer errors, in half the calendar time. Under a day-rate model, they earn half as much. The more efficient they become, the more the pricing model penalizes them.
This is not a theoretical problem. It is already happening to consultants who have significantly increased their output velocity with AI. The smart ones have started selling Units of Work. The rest are quietly discounting their own productivity.
What the Unit of Work means precisely
A Unit of Work is a bounded professional object with a defined input, a repeatable method, and a certified output. It is not an open-ended engagement. It is a scope.
Examples: a codebase audit for a defined repository, a contract review for a specific agreement, a migration plan for a named system, a compliance analysis for a given regulation, a product specification for a scoped feature. Each unit has a clear entry point, a method the expert applies, and a deliverable the client can receive, read, and use without further explanation.
The Mission Plan that opens each unit captures the scope, the constraints, and the agreed outcome. The Agent handles the first pass of production. The expert performs the Audit: reviewing, correcting, and certifying the result. The Certified Deliverable closes the unit. That is what the client purchases. Not a day of the expert's time, but a supervised professional object.
The simple arithmetic
Consider a senior consultant who currently bills 800 euros per day. Under the day-rate model, one day of calendar work generates 800 euros of revenue, assuming they can sell every day and have no unpaid qualification time.
Under the Unit of Work model, the same consultant packages four audits per type at 250 euros each. With AI handling Agent production, each audit requires roughly ninety minutes of expert Audit time: reviewing the output, correcting errors, certifying the result. Four units can comfortably be completed in a single calendar day, generating 1,000 euros of revenue, with cleaner client expectations and less unpaid pre-sales work.
The arithmetic is not magic. It works because the Unit of Work separates the production cost, which AI has reduced dramatically, from the expert judgment cost, which remains scarce and valuable. The client is not paying for the consultant's Tuesday afternoon. They are paying for a Certified Deliverable they could not produce themselves.
The Mission Cycle in five phases
The Bidirectional Mission Cycle in Temet structures the Unit of Work as a five-phase sequence.
Phase one is Prepare: the expert and their agent draft the Mission Plan, defining scope, deliverable format, and method. Phase two is Push: the Mission Plan is sent to the client through the A2A Network via a signed envelope. Phase three is Receive: the client reviews the Mission Plan, makes modifications, and returns it. The agent ingests the client's changes and produces a semantic report of what they kept, modified, and asked for. Phase four is Report: the agent delivers the semantic report to the expert, who reviews what the client wants and audits the final version of the deliverable. Phase five is Orchestrate: the expert accepts, adjusts, or rejects the iteration and prepares the next version if needed.
This five-phase cycle replaces the open-ended brief, the vague email chain, and the unbounded back-and-forth that currently consume a consultant's unpaid time before work even begins.
The first concrete step
The shift from day rate to Units of Work does not require a new business entity or a revised website. It requires defining one repeatable Unit of Work clearly enough to price it and deliver it consistently.
Start by identifying the service you deliver most often. What does it actually produce? What does the client receive? What did the last three engagements have in common? That common structure is a Unit of Work waiting to be named and priced.
Install Temet. Let it read your existing sessions and surface the Encoded Method already present in your practice. Use that as the basis for your first Mission Plan. The day rate is not abolished on day one. It gives way over time to a portfolio of Units of Work that price the expert's judgment correctly rather than discounting their efficiency.
FAQ
Is the day rate completely obsolete?
Not immediately. Open-ended advisory work still fits a time-based model. But for repeatable deliverables with defined scope, the Unit of Work is more aligned with the actual value the expert provides.
What if clients resist fixed-scope pricing?
Clients who have experienced scope creep on hourly engagements often prefer fixed-scope units once the scope is clearly defined. The Mission Plan makes the scope explicit before work begins, which reduces the disagreements that arise at invoice time.
How does Temet help with this transition?
Temet structures the Mission Inbox to receive scoped requests, helps the expert prepare Mission Plans, and ensures every deliverable carries a supervision trace. The infrastructure supports Units of Work rather than open-ended engagement logs.
Can I run multiple units in parallel?
Yes. That is the point. When Agent production handles the first pass and the expert supervises the Audit, parallel units become operationally realistic. The expert is not producing each unit from scratch; they are the control layer across several concurrent missions.
Next step
Use this guide in practice with Temet's audit, tracking, and profile workflow.
Connect your agentPublished May 29, 2026