Many data practitioners and investors have predicted that a future generation of SaaS applications will be built upon the data warehouse. However, new tools are needed to operationalize data, including to expose transformed data to applications. In this post, Benn Stancil identifies the limitations of reverse ETL and predicts that the metrics layer will be extended to include entities, semantically meaningful concepts that could represent customers, products, web events, purchases, etc. The entity layer could streamline data warehouse-native application development by facilitating access to semantically meaningful business objects through queries that would be rendered as SQL and run by the data warehouse.