Get certified for free when you join Fabric Data Days 2026 and dive into Fabric, Power BI, SQL, AI, and other essential data skills.
Join nowTry your skills in the Power BI Dataviz World Championship! Round one ends June 26. Join now
I'm working with a dataset in a data warehouse. It is a very large dataset but I suspect in most cases the users will only be querying newer data. I think a Direct Query solution will likely work, but I wondered what the best place is to put in my date filters would be. Is it better to insert these filters at the Power Query level or in Power BI? Or does it matter since the queries are only actually sent when a visualization is refreshed?
Theoretically, then, I could insert a report-level filter for dates and have that set as default but allow report viewers to change it if for some reason they want to query older data? In most cases that should keep performance within reasonable bounds, right? If that works, it seems like a nice compromise.
Solved! Go to Solution.
Doesn't really matter in Direct Query but you can use Power Query as a sort of safety net. Whatever the users filter in the UI will be added to the Power Query filters.
Doesn't really matter in Direct Query but you can use Power Query as a sort of safety net. Whatever the users filter in the UI will be added to the Power Query filters.
| User | Count |
|---|---|
| 24 | |
| 22 | |
| 22 | |
| 18 | |
| 13 |
| User | Count |
|---|---|
| 67 | |
| 56 | |
| 45 | |
| 42 | |
| 30 |