South Star Metrics, Revisited
A Diagnostic Toolkit for North Stars Gone Wrong
Nearly 6 years ago, I published one of my earliest posts: South Star Metrics. The concept was simple: sometimes, over-rotating on a north star metric results in an anti-pattern with unintended side effects. I told the story of a Microsoft PM who, in the pursuit of getting Windows Update adoption above 90%, decided to just stop asking users for permission to reboot…
South. Star. Metric.
That post listed 7 scenarios where metrics go sideways, and they resonated enough that they have come up in conversations, interviews, and reader messages ever since. But looking back, I listed the scenarios without giving them proper names. And in the spirit of Goodhart’s Law (”when a measure becomes a target, it ceases to be a good measure”), I’ve been wanting to expand on the specific ways that north star metrics go south. A diagnostic toolkit, if you will.
So here it is: 7 types of south star metrics, each with a label, a definition, and a diagnostic to help you spot it in the wild.
