Where AI actually belongs in an operations dashboard

Every operations dashboard we're asked to add "AI" to already has the data it needs to answer a narrower, more useful question than the one being asked. The request is usually "can it predict X," when the actual gap is "nobody notices when Y drifts outside the normal range until it's already a problem."

Anomaly detection on data you already collect is a better first AI feature than a forecasting model on data you don't yet trust. It's cheaper to build, easier to validate, and it fails safely — a false alert costs someone two minutes, a bad forecast costs someone a quarter's planning.

Our rule of thumb: an AI feature earns its place on a dashboard when removing it would make someone's job measurably harder, not when it makes the product page read as more sophisticated.