At Idorsia, we have developed a large R Shiny application supporting a metadata-driven approach to shell and output creation. We will demonstrate some key features of the app and discuss how this approach led to huge efficiency gains when delivering results from a Phase III trial earlier this year. Additionally, we will discuss the value of the metadata collected within the app, and highlight the reuse of this in driving further Shiny apps supporting data review and exploration around the specified outputs. Our hope is to show the versatility of Shiny for this work, particularly when coupled with well-designed metadata management.