Sheet 10
The same program, the same intake form, and the same nine questions, applied to two requests that deserve very different treatment. Read together, they are the argument the whole playbook is making: proportionality is not a slogan, it is a measurable difference in how long a request waits.
Both scenarios are fictional training examples built for demonstration. Nothing here is legal advice or an employment-law position, and the recruiting example in particular would be handled with qualified counsel in a real organization.
Both numbers below are measured between two dates already in the record set: the day the request was registered and the day the last related decision was recorded. Neither is an estimate of how long a review “should” take.
75days
04 Mar 2026 registered to 18 May 2026 decided.
Screening declined with documented reasoning. A reduced-scope coordination pilot approved with four conditions, addressing the larger share of the capacity problem with none of the employment-decision risk.
9days
28 Jan 2026 registered to 06 Feb 2026 decided.
Approved as standard practice with published guidance, nine days from submission to publication, with no pilot and no committee debate.
The difference is not that one team was treated better. It is that one request touched employment decisions and restricted personal data and the other did not, and the model says so out loud before either requester submits. A requester who knows the elevated pathway takes months can decide whether the consequential part of their idea is worth it — which is a better conversation than discovering the cost after waiting for it.
Every step shows what happened, who owned it, and which artifacts were produced or used. The artifact links go to the template and its worked example.
People & Culture asks to use a recruiting platform's AI screening to score and rank roughly 900 early-career applicants for 40 openings. The problem is real: two recruiters cannot keep up, response times slip, and strong candidates accept other offers while they wait.
Why this example
This is the hardest kind of request to handle well. Saying no protects the firm but leaves a real problem unsolved and teaches the requester that the program is an obstacle. The useful outcome is neither approval nor refusal — it is finding which part of the problem can be solved without putting a consequential decision inside a tool nobody can explain.
Step 1 of 11 · S-01
Two recruiters, 900 applications, 40 openings, and a nine-day median response time.
Screening declined with documented reasoning. A reduced-scope coordination pilot approved with four conditions, addressing the larger share of the capacity problem with none of the employment-decision risk.
Scale, modify, pause, or retire at the December 8 governance review