Skip to content
Go back

Token Usage Is an Input Cost, Not an Engineering Outcome

Updated:
Commentary

Originally published onLinkedIn on 15 Jul 2026.

Bryan O’Neill’s analysis of “tokenmaxxing” describes the latest form of engineering productivity theatre: teams generate measurable AI activity and then mistake that activity for delivered value. The pattern is familiar. Lines of code, story points, test counts, and now model tokens are attractive because they are easy to count.

The diagnosis is convincing. The proposed answer—define intent before generation—is necessary, but a specification alone is not enough. Specifications can also become targets, grow without increasing clarity, or preserve the wrong requirement with impressive precision.

What a token count actually tells us

Token usage is operational data. It can help estimate cost, latency, context pressure, and the efficiency of repeated workflows. It does not directly measure whether the resulting change is correct, maintainable, safe, or useful.

The same token count can describe very different outcomes:

Low consumption is not automatically efficient, and high consumption is not automatically wasteful. The measurement becomes misleading when it is detached from the evidence produced by the work.

Measure the chain, not one convenient number

AI-assisted engineering should be evaluated at several levels that answer different questions:

LevelUseful questionExample evidence
ResourcesWhat did the attempt cost?tokens, wall time, compute, human review time
FlowDid work move through the system better?lead time, review latency, rework, blocked time
QualityDid the change preserve or improve behaviour?tests, escaped defects, rollback rate, reliability signals
OutcomeDid the system solve the intended problem?user task success, reduced operational risk, measurable service improvement

No row should be promoted into a universal score. A faster change that increases rollback risk is not an improvement. A larger test count that repeats the same equivalence class is not broader verification. A lower token bill that transfers hours of work to a reviewer is not necessarily cheaper.

Specifications need executable evidence

The strongest part of spec-driven development is not that prose is written before code. It is that intent can be made inspectable before implementation begins. To resist metric gaming, the specification should be connected to evidence:

This shifts the unit of evaluation from “output produced” to “claim supported.” An agent may generate code, tests, diagrams, and documentation, but the engineering result is the justified relationship between the intended change and the observed evidence.

The practical conclusion

Track tokens when they help operate or budget an AI system. Do not use them as a proxy for productivity. The same warning applies to the volume of generated code, number of agents, number of tests, or size of a specification.

The useful question is not how much activity occurred. It is which claim was established, what evidence supports it, which uncertainty remains, and whether the result improved a real engineering or user outcome.


Share this note

Previous Note
Agentic Resource Discovery Still Needs a Verification Boundary
Next Note
Paper Diagrams Are Maps, Not Proof Certificates