The Number That Runs the Company

What this pack runs.

Somebody picks a number because it approximates something that matters and can be counted weekly. Targets get written in it, then bonuses, then the person who chose it moves on — and with them goes the only thing that made it a proxy rather than a goal. Five years later the organization is optimizing the number, competently and in good faith, and nobody in the building can say what it originally stood for. The check never happens because the number and its outcome usually both go up, which reads as confirmation; drift is slow by construction, because a proxy that diverged fast would have been noticed and replaced. This pack turns that into arithmetic. It computes the worth of one unit today and the worth of one unit when the metric was adopted, scales the old number by how much the outcome actually grew, and returns the hollow share — the part of the number carrying none of what it was chosen to approximate — with what producing it cost. It stops on five conditions: nobody has said what the number is, what it was at adoption, how long it has been the number, what it stands for, or what is spent driving it.

The casebooks these cases come from.

The pack reads nine sourced decisions through the framework. They are drawn from these casebooks, each one a set of verified records on a single decision type.

A verified record has passed a mechanical claim-versus-source check, a second model's read, and human approval. The count is what exists today, not a target.