Skip to Content
GuidesAdvanced analyticsAnomaly detection

Anomaly detection

Active Reach continuously monitors your key metrics and alerts you when something unusual happens — a sudden drop in delivery rate, an unexpected spike in bounces, or a trend reversal in engagement.

How it works

The platform compares current metric values against your historical baseline. When a metric deviates beyond normal variance, it generates an anomaly alert. The detection adapts to your workspace’s unique patterns — seasonal trends, send schedule, and growth trajectory are all factored in.

What’s monitored

Metric categoryExamples
DeliveryDelivery rate, bounce rate, complaint rate per channel
EngagementOpen rate, click rate, reply rate
VolumeSend volume, event ingestion rate
RevenueConversion rate, attributed revenue, average order value
InfrastructureAPI latency, worker queue depth

Viewing anomalies

Anomalies surface in two places:

  • Measure → Insights → Alerts — dedicated alert feed with severity and recommended actions
  • Measure → Analytics → Overview — anomaly badges appear on the time-series chart where the deviation occurred

Each anomaly shows:

  • What changed — which metric, by how much, over what period
  • Severity — critical, warning, or info
  • Likely cause — AI-generated hypothesis (e.g., “Provider outage”, “Campaign targeting too broad”, “New segment drift”)
  • Recommended action — what to do about it

Configuring alerts

Go to Measure → Insights → Alerts to configure:

  • Which metrics to monitor — toggle individual metrics on/off
  • Notification channels — email, push, in-app, or Slack webhook
  • Frequency — real-time, daily digest, or weekly summary

What’s next