AI / ML
Anomaly Detection
Automatically identifying statistically unexpected changes in marketing metrics: sudden CPA spikes, unusual CTR drops, spend anomalies, or conversion volume changes that deviate from expected patterns. Catches problems hours or days before manual review would. Uses statistical models to establish 'normal' ranges for each metric and flags deviations beyond a confidence threshold. Critical for brands managing high daily spend where a day of undetected problems can waste thousands.
Why it matters
Most metric damage is done in the days before anyone notices. Detection value is almost entirely in the lag it removes.
In practice
Tune the threshold against your own volatility. An alert that fires on ordinary daily variance gets muted within a week, which is worse than no alert.
Related terms