Question

A data engineering team publishes a daily sales dashboard in Amazon QuickSight backed by a SPICE dataset. Business analysts want to be automatically alerted when daily revenue for any product category deviates significantly from its expected pattern, without writing custom statistical code or maintaining external models. Which QuickSight capability should the team enable to meet this requirement with the least operational overhead?

Accepted Answer

Configure QuickSight ML Insights anomaly detection on the revenue measure grouped by product category

Answer

Create a CloudWatch alarm on the QuickSight SPICE ingestion metric and notify analysts via SNS

Answer

Export the dataset to Amazon S3 nightly and run an Amazon SageMaker training job to score outliers

Answer

Add a calculated field that flags rows exceeding two standard deviations using QuickSight functions