A company wants to shorten the time to develop new features. It uses CodeBuild and CodeDeploy to build and deploy applications and has a separate CodePipeline CI/CD pipeline per microservice. The company needs more visibility into metrics such as average time between feature releases and mean time to recovery after failed deployments. Which solution gives that visibility with the least configuration effort?

The correct solution uses EventBridge to trigger a Lambda function immediately after each successful or failed pipeline run. This ensures real-time capture of relevant events. The Lambda then emits custom CloudWatch metrics, which are ideal for tracking performance indicators like deployment frequency and mean time to recovery. CloudWatch dashboards can then easily visualize these metrics with minimal configuration. The incorrect options either poll for events (which is less efficient and real-time than event-driven triggers) or use DynamoDB and QuickSight. While DynamoDB and QuickSight are powerful, they introduce more complexity and configuration effort than directly leveraging CloudWatch for metrics and dashboards, which is designed for this type of monitoring.