One Semantic Model for Multiple Reports

[ndias]

Hi all,

Question:
Is there a supported Fabric / Power BI pattern to deploy one Master Semantic Model to multiple customer workspaces with different parameter values without maintaining N .pbix files?

I have multiple customers, each with its own ID for queries but identical report logic (same KQL, same model, same visuals). Everytime I need to update a query, I have to update every file (all reports live in the same workspace).

To avoid manually editing each model, I've been trying to create a Master Semantic Model, use Parameters to set the ID, and deploy it to individual workspaces (one workspace per customer).

What happened: when I connect the semantic model to the report, it connects, obviously, as DirectQuery, and I cannot change parameters. This make the Master SM useless.

I've tried to use deployment pipelines to deploy the SM to customer workspaces but it didn't work, the result is the same: I cannot set the parameter.

Any official guidance or recommended architecture would be appreciated.

Thanks!

[v-saisrao-msft]

Hi @ndias,

Checking in to see if your issue has been resolved. let us know if you still need any assistance.

Thank you.

[v-saisrao-msft]

Hi @ndias,

Have you had a chance to review the solution we shared earlier? If the issue persists, feel free to reply so we can help further.

Thank you.

[GeraldGEmerick]

@ndias The standard pattern for this is to publish to one central workspace and use Row Level Security (RLS).

[ndias]

Hi @GeraldGEmerick thanks for your reply. 

I should mention that the reports still differ, different logos for different customers. RLS is not going to help me with this. I still need N individual reports/dashboards.

[GeraldGEmerick]

@ndias RLS could be used for this, even logos. You specify the URL for each logo and RLS control it. The measure uses MAX to grab the URL (only 1 available for each customer). 

The other option you have is parameter binding with DirectQuery, however the lmitations around that not supporting RLS are likely going to prevent you from being successful with what you are trying to accomplish. Dynamic M query parameters in Power BI Desktop - Power BI | Microsoft Learn

[ndias]

hi @GeraldGEmerick , thanks for answering! I believe that will work for a dashboard published in one Fabric/Power BI workspace to which all viewers have access. In my case each customer has its own workspace and App (and thus I have several security groups). Or am I missing something here?

[v-saisrao-msft · Accepted solution]

Hi @ndias ,

Thank you @GeraldGEmerick , for your insights,RLS is most effective when all customers use a shared semantic model or report within a single workspace, allowing data to be filtered for each user. Because your customers each have their own workspace and app, you’re using a workspace-per-tenant isolation model. In this setup, a single master model can’t adjust parameters for each workspace from the report side, so some duplication is unavoidable. The recommended approach is to maintain one master PBIX file and automate its deployment to each workspace, setting the parameter for each customer as needed.

Row-level security (RLS) with Power BI - Microsoft Fabric | Microsoft Learn

Row-level security (RLS) guidance in Power BI Desktop - Power BI | Microsoft Learn

Develop scalable multitenancy applications with Power BI embedding - Power BI | Microsoft Learn

Thank you.