N—WORK / Decision filesQuestion · Boundary · Position

The shape of the work

Clarity should be visible in the record.

Client material remains private. The files below show how a consequential decision is framed, challenged and turned into a written next move.

File 01
AI investment

Value before scale.

The central question is not whether an AI system can be built. It is what evidence would make the investment rational—and what evidence should stop it.

Situation

A compelling technology case is gathering organisational momentum before the commercial threshold is explicit.

Boundary

Separate capability from value. Fix the owner, admissible evidence, cost of error and conditions for further investment.

Position

Proceed only to the smallest test capable of changing the investment decision; narrow or stop when the threshold cannot be met.

  • Decision brief
  • Value threshold
  • Assumption ledger
  • Stop conditions
File 02
Pilot governance

Success before expansion.

A pilot becomes dangerous when being live is treated as evidence of being useful. Success must be written before the work can redefine it.

Situation

The pilot is operating, but evidence, validation and release decisions are not tied to predetermined gates.

Boundary

Name observable success, unacceptable failure, minimum evidence and the person accountable for the release decision.

Position

Expand only when every release gate is satisfied. Correct, narrow or stop without treating sunk effort as proof.

  • Release protocol
  • Evidence ledger
  • Failure states
  • Review point
File 03
Platform commitment

Reversibility before lock-in.

A platform choice often becomes permanent through integration, incentives and habit long before anyone declares it strategic.

Situation

A short-term implementation choice is quietly acquiring switching cost, operational dependence and internal sponsorship.

Boundary

Expose dependencies, exit cost, data portability, control points and the last credible moment to choose differently.

Position

Commit only where the evidence justifies permanence. Preserve optionality elsewhere through explicit technical and commercial limits.

  • Dependency map
  • Exit conditions
  • Option comparison
  • Commitment boundary

Bring the decision while it can still change.

Three concise answers are enough to start a private review. Every engagement is led directly by Can Unal.

Open the private brief