Detecting unusual spend with AWS Cost Anomaly Detection - AWS Cost Management

Detecting unusual spend with AWS Cost Anomaly Detection

AWS Cost Anomaly Detection is a feature that uses machine learning models to detect and alert on anomalous spend patterns in your deployed AWS services.

Using AWS Cost Anomaly Detection includes the following benefits:

  • You receive alerts individually in aggregated reports either in an email message or an Amazon SNS topic.

    For Amazon SNS topics, create an AWS Chatbot configuration that maps the SNS topic to a Slack channel or an Amazon Chime chat room. For more information, see Receiving AWS Cost Anomaly Detection alerts in Amazon Chime and Slack.

  • You can evaluate your spend patterns using machine learning methods to minimize false positive alerts. For example, you can evaluate weekly or monthly seasonality and natural growth.

  • You can investigate the root cause of the anomaly, such as the AWS account, service, Region, or usage type that's driving the cost increase.

  • You can configure how to evaluate your costs. Choose whether you want to analyze all of your AWS services independently or analyze specific member accounts, cost allocation tags, or cost categories.

After your billing data is processed, AWS Cost Anomaly Detection runs approximately three times a day in order to monitor for anomalies in your net unblended cost data (that is, net costs after all applicable discounts are calculated). You might experience a slight delay in receiving alerts. Cost Anomaly Detection uses data from Cost Explorer, which has a delay of up to 24 hours. As a result, it can take up to 24 hours to detect an anomaly after a usage occurs. If you create a new monitor, it can take 24 hours to begin detecting new anomalies. For a new service subscription, 10 days of historical service usage data is needed before anomalies can be detected for that service.


You can opt out of Cost Anomaly Detection at any time. For more information, see Opting out of Cost Anomaly Detection.