The calibration programme

Nobody in this category has published validation data.

Not Gartner. Not McKinsey. Not Deloitte. Not this instrument, yet. Every framework in this space asserts its weights.

That is uncomfortable in a book whose central argument is that you should not believe claims you cannot evidence. So the weights are published as a hypothesis with a stated test, and this is the test.

The design

What is actually being tested.

ElementDetail
Sample50–100 organisations, initially within a single sector
MethodFull thirty-five-question assessment, same protocol, same evidence standard
BaselineScore by dimension, plus revenue per employee, headcount and service-line mix
Follow-upThe same operating metrics at twelve months
TestDoes the baseline score predict the change, and which dimensions carry the prediction?
OutputPublished distribution, sector medians by size band, and re-derived weights
The predictions

Four are testable from assessment data directly.

1Coordination cost per unit of work declines with cumulative execution
2Management-to-executor ratios fall over time
3Human approvals decline as precedent coverage increases
4Score at baseline predicts revenue-per-employee movement at twelve months

The falsification, stated plainly

If management-to-executor ratios stay flat across a decade of serious adoption, the theory underneath Part VIII is wrong — and that is computable from public filings by anyone, without my cooperation.

A prediction you can only evaluate with the author's help is not really a prediction.

Participating

What you get, and what you give.

You receive

Your own calibrated readout

The full assessment, the coherence matrix, the economics baseline and the deflation map. Plus your sector median — which does not currently exist anywhere.

You receive

A named credit

Contributing organisations are credited in the book's appendix, and get early access to the manuscript and to revisions.

You give

Anonymised results

Dimension scores, sector, headcount band, and where provided, subsequent revenue-per-employee movement.

Protections

How the data is handled

No identifying detail retained. No individual score ever published. Reported only in aggregate at a minimum cohort size of ten. Withdrawable at any time, including retrospectively.

Status

First cohort.

The programme is in its first cohort. Fifteen completed assessments produces a publishable sector median; fifty produces a defensible one.

Current sample size, the distribution as it stands, and any weight the data has already forced to change will be published here and mirrored in the source ledger. Including — especially — where it contradicts the argument of the chapter that defends the current weighting.

Live: 0 consented self-assessed results in the pool. The median publishes at a minimum cohort of ten.