A measurement becomes useful when its limits remain attached to its authority.

Numbers travel further than their caveats. A figure gets produced with conditions attached, then quoted in a meeting, then repeated as a fact, and by the third telling nobody recalls what it measured, when, or how wide the uncertainty was. It has not become more accurate; it has lost its handles. Confidence in data tends to grow as the qualifications fall away.

Whenever you pass on a figure, carry one sentence of context with it: what it counts and how sure it is. Ask the same when you receive one. A number without limits is a rumour better dressed.