A marketing company collects clickstream data. The company sends the data to Amazon Kinesis Data Firehose and stores the data in Amazon S3. The company wants to build a series of dashboards that…

Storing and querying raw clickstream data directly in Amazon Redshift for daily updates and hundreds of users can be very expensive in terms of storage, compute, and ETL processes, especially when a serverless, pay-per-query model for S3 data exists.

Defining a QuickSight dataset using a direct SQL query (against a suitable source like Amazon Athena or Redshift Spectrum on S3) provides flexibility to select and transform specific clickstream data for analysis. This approach allows precise control over the data ingested for dashboards.

Amazon Athena is a cost-effective, serverless option for querying S3 data. However, for dashboards accessed by hundreds of users, consistently direct querying Athena for every interaction can accumulate high query costs due to the pay-per-query model based on data scanned. It is often more efficient as a source for SPICE.

The Amazon QuickSight SPICE engine provides a highly scalable and cost-effective solution for serving dashboards to hundreds of users with fast, interactive performance. Importing the defined dataset into SPICE with a daily refresh minimizes direct queries to the underlying data source, reducing costs and resource consumption for frequent dashboard access.

Amazon S3 Analytics is a storage class analysis tool for optimizing S3 storage costs by identifying access patterns, not a service used for querying data to build interactive dashboards.