Decides what a business should and should not build with AI, and in what order.
Why this one is hard to judge
Everyone can talk about strategy. The tell is whether they have killed a project that was already funded.
What to ask for
Ask for a specific decision they changed by advising against a popular option.
Ask for a one-page plan they wrote that a team actually followed.
Ask them to name a project they recommended stopping, and why.
Ask how they would spend the first budget allocation and what they would skip.
The mistake most hirers make
Hirers often choose someone who talks fluently about the field but has never made a call that cost something. They mistake broad awareness of tools for the ability to say no to a popular idea. The result is a plan that pleases everyone and commits to nothing.
What good looks like after 90 days
A clear written view of where the organisation should and should not use AI. At least one project stopped or redirected on their advice. A short list of decisions that others can now make without asking them.
How we assess it
Against a rubric that is published in full, on evidence the practitioner supplies and a reviewer checks. Where something has not been verified, the profile says so.