Amazon Lookout for Metrics
Amazon Lookout for Metrics uses machine learning to automatically detect anomalies in business and operational metrics such as revenue performance, customer engagement, and user activity. It continuously monitors data from various sources including Amazon S3, CloudWatch, RDS, Redshift, Athena, and AppFlow, providing root cause analysis and alert notifications when anomalies are detected.
APIs
Amazon Lookout for Metrics API
The Amazon Lookout for Metrics API provides programmatic access to create and manage anomaly detectors, anomaly groups, alerts, and datasets for automated anomaly detection in b...
Capabilities
Amazon Lookout for Metrics - Anomaly Detection Operations
Workflow capability for data science and operations teams to manage anomaly detectors, monitor metric anomalies, configure alerts, and provide detection feedback using Amazon Lo...
Run with NaftikoFeatures
Uses ML to automatically detect anomalies in business and operational metrics without requiring ML expertise.
Identifies the top contributors to each anomaly to help determine root causes quickly.
Connects to Amazon S3, CloudWatch, RDS, Redshift, Athena, and AppFlow as data sources.
Continuously monitors metrics and sends real-time alerts when anomalies are detected.
Configure alerts via Amazon SNS, Lambda, or other AWS services when anomalies occur.
Provide feedback on detected anomalies to improve future detection accuracy.
Tag anomaly detectors and related resources for cost allocation and organization.
Use Cases
Monitor revenue metrics and detect unexpected drops or spikes that could indicate fraud or system issues.
Track customer engagement metrics and alert teams when patterns deviate from expected ranges.
Monitor operational metrics such as system performance, error rates, and throughput for anomalies.
Detect anomalies in e-commerce metrics like conversion rates, cart abandonment, and sales volume.
Analyze user activity patterns and detect unusual behavior that may indicate security incidents.
Integrations
Use S3 buckets as a data source for metric data in CSV or JSON format.
Ingest CloudWatch metrics directly for anomaly detection.
Connect to RDS databases to retrieve metric data for analysis.
Use Redshift data warehouse as a source for business metrics.
Query Athena tables to feed metric data into anomaly detectors.
Use AppFlow connectors to ingest data from SaaS applications.
Send alert notifications via SNS topics when anomalies are detected.
Trigger Lambda functions in response to detected anomalies for custom workflows.