Definition version

A dated version number on a measure's written definition, cited on every table and answer that uses it. When the definition changes, the version changes, and prior periods are restated under the new one or marked as under the old. Without it, a definition change reads as a movement.

DefinitionA dated version number on a measure's written definition, cited on every table and answer that uses it. When the definition changes, the version changes, and prior periods are restated under the new one or marked as under the old. Without it, a definition change reads as a movement.

Annually, not silently

Definitions change once a year, versioned. A change in April makes March and May incomparable unless the version says so.

On the board pack

In the footer, per table. A board member's question about a number goes to the version first.

How it is used

Each measure has a written definition with a version number and a date. Every table and answer cites the version used, and when a definition changes, history is recomputed or the break is marked.

Example

Win rate version 2 excluded stalled deals; version 3 counts them as lost from 1 July. The board pack shows 27 percent under version 2 and 22 percent under version 3, labelled.

Where it goes wrong

A definition changed quietly. The number moves, nobody knows why, and a quarter is spent explaining a change in arithmetic.

Articles that use this term

Board and management reporting · Finance and FP&A teams

Bookings, billings and revenue: one reconciliation before anyone reports a number

Why sales, finance and the board each quote a different figure for the same quarter, the three definitions behind them, the bridge from bookings to billings to recognised revenue with the timing and the adjustments that separate them, the identity that ties the three together, and the governance rule that every report names which of the three it is showing.

16 Sept 20262 min read AI and self-service analytics

Citations on every number: what a citation from an analytics tool must contain

A number in a report or an assistant's answer is a claim until it carries a citation, and a citation has five parts: the table, the row, the period cut, the definition version, and the upload it was computed from. Why each part is there, what a reader does with it, the difference between a citation and a link, how a citation survives a re-upload, and the three shortcuts that turn a citation into decoration.

16 Sept 20263 min read AI and self-service analytics

Deterministic first: the architecture of an analytics product that uses a model and stays right

How an analytics product uses a language model without letting it touch a number: deterministic code computes every measure from stated definitions, a tool registry names each computation, the model plans which tools to call and reads their outputs, every output carries its citation, and the model's own text is labelled as a reading. The four boundaries that make this hold, what crosses each, and what never does.

16 Sept 20263 min read AI and self-service analytics

How to ask an analytics assistant a question it can answer reliably

What separates a question an AI analytics assistant answers correctly from one it answers plausibly: naming the measure, the level, the period and the comparison, asking for the table before the sentence, and checking the citation. Six rewrites of common questions, and the four kinds of question an assistant should refuse.

16 Sept 20263 min read Board and management reporting

How to read a movements page: the line to check first, and the line that is not a finding

A reading guide for the movements page of a monthly pack: why the first line to check is any failed identity, not the largest movement, how to tell a movement caused by a definition change or a re-upload from one caused by the business, the citation as the thing to click, the threshold column as the reason a line is there, the within-threshold line as the one that closes the question, and how long the page should take.

16 Sept 20263 min read Forecast and pipeline

Pipeline stage definitions: exit criteria that make a weighted forecast mean something

Why a stage-weighted forecast is only as good as the stage definitions, what an exit criterion is, five stages with criteria a rep can verify and a manager can audit, how to check from the CRM that deals in a stage actually meet its criteria, the stage probabilities derived from the team's own history rather than set in a workshop, and the versioning that keeps last year's pipeline comparable.

16 Sept 20262 min read