How to Connect Your Semantic Models to Microsoft Fabric | Power BI Tutorial

In this comprehensive Power BI tutorial, Zane Goodman walks viewers through how to connect existing Power BI semantic models to Microsoft Fabric using OneLake integration. While this integration doesn’t transform Power BI models into native Fabric semantic models, it does enable seamless data connectivity across platforms—offering users the power of Fabric with data they already manage in Power BI.

What is OneLake Integration?

OneLake integration enables users to access Power BI semantic model data directly within Microsoft Fabric environments. The key benefit is being able to work inside Fabric with the data already structured and available in your Power BI reports—maximizing reuse and minimizing duplication.

Key Requirements

• Import Mode Only: Your Power BI semantic model tables must be in Import mode. DirectQuery is not supported for OneLake integration.
• Fabric-Enabled Workspace: The workspace where the report is published must be Fabric-capable. This includes Premium capacity, Fabric capacity, or Fabric Free Trial capacity.

Step-by-Step Setup Guide

1. Create a Power BI report using data from an Azure SQL Database or another source, ensuring the model is in Import mode.
2. Save and publish the report to a Fabric-enabled workspace.
3. Inside Microsoft Fabric, navigate to your workspace and locate the published semantic model.
4. Click the ellipsis (…) on the semantic model, go to Settings, and toggle on OneLake Integration.
5. Refresh the semantic model manually or via schedule. This step is critical to enable the creation of Delta tables in OneLake.

Behind the Scenes: How It Works

When OneLake integration is enabled and the semantic model is refreshed, all Import mode tables are exported as Delta tables (using the Delta-Parquet format) and stored inside Microsoft OneLake. These tables are now accessible across your Fabric environment, such as within Lakehouses or other Fabric experiences.

Connecting a Lakehouse to Semantic Model Data

1. Open the target Lakehouse in your Fabric workspace.
2. Select the ellipsis (…) and create a new shortcut.
3. Choose Microsoft OneLake as the source and locate your semantic model by name.
4. Select the specific Delta tables you wish to link to—such as Production.Product and SalesOrderDetail.
5. Optionally rename the tables and complete the shortcut creation process.

Benefits of OneLake Integration

• Enables reuse of existing Power BI semantic models inside Fabric workflows.
• Supports continuous synchronization—data refreshes in Power BI automatically update Delta tables in OneLake.
• Works across different Fabric experiences such as Lakehouses, improving scalability and flexibility.

Limitations and Best Practices

While this feature offers immense value, Zane cautions users to consider a few important limitations:

• Only supports Import mode semantic models.
• May not be suitable for all use cases—users should evaluate if OneLake integration aligns with their data strategy.
• Microsoft documentation should be reviewed for detailed limitations and requirements.

Recommended Learning Resources

For those unfamiliar with concepts like Delta Parquet, OneLake, or Lakehouses, Zane recommends checking out Pragmatic Works' on-demand training courses, especially the introductory sessions by Manuel covering OneLake and Lakehouse architecture.

Zane concludes the tutorial by emphasizing the power and flexibility that OneLake integration offers. While not a one-size-fits-all solution, it provides a practical bridge between Power BI and Fabric for teams looking to scale their analytics solutions. As always, he encourages viewers to leave comments and explore more training via Pragmatic Works' Learn with the Nerds sessions.

Don't forget to check out the Pragmatic Works' on-demand learning platform for more insightful content and training sessions on  Power BI and other Microsoft applications. Be sure to subscribe to the Pragmatic Works YouTube channel to stay up-to-date on the latest tips and tricks.