You need to connect to and transform data to be loaded into a Fabric lakehouse using Dataflows Gen2. How would you complete this task?

• A. Connect to Data Factory workload > Create Pipeline to copy data and load to lakehouse > Transform directly in the lakehouse.
• B. Connect to Real-time intelligence workload > Create Pipeline to copy data > Transform data with an Eventstream.
• C. Connect to Data Factory workload > Create a Dataflow Gen2 to transform data > add your lakehouse as the data destination.
• D. Connect to Data Factory workload > Create a Dataflow Gen2 to transform data > export the transformed data to a separate data store.

Answer: (D)

Using Dataflows Gen2 to transform data inside a Data Factory workflow is about shaping and cleansing data before it lands in the Fabric lakehouse, often via a staging or intermediate sink. In this approach, you connect to the Data Factory workload, create a Dataflow Gen2 to apply the desired transformations, and then export the transformed results to a separate data store that serves as a staging area. This staging step then feeds the lakehouse, ensuring data quality and compatibility before final ingestion.

This path leverages the transformation capabilities of Dataflows Gen2 and keeps the ingestion process modular and verifiable. The other options skip the proper transformation step with Dataflows Gen2, rely on real-time/event streaming workflows that aren’t aligned with a typical batch transformation-to-lakehouse pattern, or suggest loading directly into the lakehouse without clarifying how the data is transformed within Dataflows Gen2.