What Is Alert Fatigue?

Alert fatigue is the desensitization that sets in when alert volume exceeds meaning. When most notifications require no action, operators adapt rationally: they mute channels, batch reviews, build mental filters — and the one alert that mattered dies in the noise. The postmortem then records, accurately and uselessly, that "an alert did fire."

Why it matters

Alerting exists to move scarce attention to real deviations. A noisy alerting system is worse than none, because it consumes attention and provides false assurance that the operation is being watched. Fatigue is not an operator discipline problem; it is a design failure in what was deemed alert-worthy — the argument of Alert Fatigue Is a Product Design Failure.

Design correctives

Alert on business impact, not raw thresholds: "orders aging past promise" beats "queue depth over 1,000." Route to a named owner with the context to act — an alert without an owner is a broadcast. Give every alert a disposition (acted, suppressed, false) and prune relentlessly on that data. Prefer exception queues with ages and costs over point-in-time pings, per Build an Exception-First Control Tower.

Related concepts

See COO dashboard, AI control tower, and business observability.