AI service
Amazon Lookout for Metrics
Detects anomalies in business and operational metrics.
Key points
- Detects anomalies in time-series business and operational metrics.
- Can help identify root causes and related metrics when unusual metric behavior occurs.
- Targets metrics such as revenue, transactions, page views, or operational KPIs.
- Connects to supported data sources and monitors metric streams over time.
- AWS discontinued support for Lookout for Metrics on October 10, 2025.
When to use it
- Choose Lookout for Metrics in legacy exam-style scenarios about automatic anomaly detection in business KPIs.
- Use it conceptually for spikes, drops, or unusual patterns in operational metrics.
- Use SageMaker or other current analytics approaches for new custom anomaly detection planning.
Exam tips
- Lookout for Metrics is for metric anomalies; Lookout for Equipment is for industrial sensor behavior.
- Lookout for Vision is for visual defects, not time-series KPI anomalies.
- Because support has ended, avoid presenting it as the current best choice for new workloads.
- CloudWatch alarms detect threshold breaches; Lookout for Metrics focused on ML-based anomaly patterns.