Goodhart’s law

Charles Goodhart; this formulation by Marilyn Strathern · 1975

When a measure becomes a target, it stops being a good measure.

Its place in the frameworkData Core›KPI management

What it does

Originally an observation about monetary policy, generalised since to any managed indicator. The mechanism is not dishonesty: a measure is a proxy for something you care about, and optimising the proxy will eventually pull it away from the thing it stood for, whether or not anyone intends it.

Reach for it when: Before attaching consequences to any indicator, and whenever a metric has improved dramatically while the outcome behind it has not.
Where it stops: It is a warning, not a method. It does not say which measures will corrupt or how to prevent it — only that they will, and that unmeasured judgement remains necessary.

Charles A. E. Goodhart, 1975; this wording from Marilyn Strathern, “Improving Ratings”, European Review, 1997.

Why it sits at KPI management

Choosing the few measures the organisation is actually run on, defining them so two people would agree, setting what good looks like, and reviewing them honestly.

A model is only useful when you reach for it at the right moment. This one answers a question that arises here — so it is filed here, and nowhere else. These are the working areas it serves:

Choosing what to measureKaplan & Norton’s balanced scorecard, and the leading-against-lagging distinction that determines whether a measure permits a response.
DefinitionsMeasurement-system practice on operational definitions, and the reproducibility requirement that separates a measure from an impression.
Targets and thresholdsGoal-setting research on specific and demanding targets, and statistical process control on reacting to signal rather than noise.
The reviewManagement-control research on review as the point where measurement becomes management rather than reporting.

All of KPI management

What it touches elsewhere

Nothing in a business is decided on its own. A conclusion reached with this model at KPI management lands in these other cores, whether or not anyone follows it there.

Goal CoreA KPI without a goal behind it is a number; a goal without a measure is an intention.
Data CollectionA measure that cannot be collected reliably is not a measure, whatever the definition says.
Business CoreMost operational KPIs are process measures, and they belong to whoever owns the process.
Time CoreThe review rhythm has to sit inside the organisation’s existing cycle rather than beside it.

Filed at the same place

These answer questions that arise at KPI management too. Where they disagree with this one, the disagreement is the useful part.

The balanced scorecardMeasure across four perspectives at once, so financial results are read alongside what produces them.

Elsewhere in Data Core

These are other people’s models, named here so you can go to the source and use them properly. The Omnigoal is not affiliated with their authors and is not endorsed by them; nothing of theirs is reproduced here — no canvas, no diagram, no wording. Each is described in our own words, with the originator credited, because the framework is a place to put thinking, not a replacement for the people who did it. Model names and trademarks belong to their respective owners and are used here only to refer to the work itself.

All 125 models