Forum Discussion
Single master PBI report for multiple clients
- 11 months ago
Hi mbravonxp ,
Power Query parameters are inherently static and are evaluated only during dataset refresh, which means they do not update dynamically at runtime based on the logged-in client. This explains why you are not observing the expected dynamic behavior. In an Embedded environment, the recommended method is to define a parameter for the container path in Power Query and have its value provided at runtime by your embedding application. When generating the embed token, you can use the EffectiveIdentity or parameter binding feature to supply the appropriate container path. This ensures that the master report selects the correct container for each client without requiring report duplication or manual parameter changes. If you later move to a semantic layer, you can implement Row Level Security, but dynamic container switching per client is only achievable through dynamic binding in Embedded, not directly via Power Query or DAX.
Thank you,Tejaswi.
Hi mbravonxp ,
This scenario can be accomplished using a single master report by leveraging dynamic parameters and row-level security (RLS) alongside Power BI Embedded. Rather than managing multiple report copies, you can create one report and control data access at runtime. Since each client’s data is stored in separate Azure Data Lake containers, parameters in Power Query can be configured for the container path. When accessed via your embedded application, you can pass the appropriate parameter for the logged-in client, ensuring data isolation.
Additionally, RLS can be set up so only data relevant to the authenticated client is accessible. With an Embedded licence, the EffectiveIdentity feature allows you to securely pass client context from your application to the report.
This approach keeps the report structure unified, streamlines modifications, and maintains data privacy. As your client base expands, implementing a semantic layer, such as Synapse or Dataflows, can further enhance scalability.
Thank you,
Tejaswi.