A large retail organization is leveraging Amazon Athena for ad-hoc SQL querying of their multi-petabyte e-commerce transaction dataset. The dataset, housed in Amazon S3, is updated nightly through an AWS Glue job. To align with the retail analytics team's needs, the dataset must be refreshed in the BI tools every two hours. The data engineer aims to optimize Athena query costs while maintaining operational simplicity and adhering to the refresh frequency requirements. Which approach should the data engineer adopt to minimize Athena querying costs without increasing infrastructure complexity?