Forum Discussion
Using multiple models within a report
In reporting tools like Excel or SSRS, creator can connect to multiple models (via Excel Pivot Tables or SSRS Datasets/Query Designer) and refine/filter the data they need from the model. This is rather intuitive for the report creators.
What is the best way to do this with Power BI?
Connecting to existing Shared Semantic Models (via “Power BI semantic models” data source) users can’t further refine/filter the data from the model (except via the Filter pane). In addition, if a report needs to connect to multiple Shared Semantic Models (via a local, composite model) this seems to require modelling skills for creators as opposed to connect, filter and drag-n-drop.
Basically, our self-service creators (business users) will need to create their own KPI reports with data from multiple Shared Semantic Models. ATM this seems more intuitive via Excel than Power BI.
Appreciate any thoughts or advice.
3 Replies
Hi ScottKC ,
For your question, you can ask someone with modeling experience to create a Power BI report first. After joining the required multiple tables, then create relationships between tables. Then place the visuals you need to use. Use DAX to create measures to populate visuals. Finally, set up RLS and publish to a collaborative workspace in Power BI Service. Add your users to the workspace as viewer roles. The data will be filtered based on the user.
Thanks for your suggestion, however it isn't really a suitable user solution when looking to move business users from Excel to PBI.
