Is it better to perform data transformations in Power Query or using DAX calculated columns?

All answers to this question.

The rule of thumb is "as far upstream as possible." Whenever you create a DAX calculated column, that data is processed during the model refresh and stored in the RAM, but it isn't compressed as efficiently as the data loaded through Power Query. Furthermore, calculated columns don't benefit from "Query Folding" (where Power BI pushes the work back to the SQL server). Using Power Query (M) to clean your data ensures that your model remains lean and your refresh times stay fast. Save DAX for measures (aggregations like Sum, Average, etc.) rather than static data transformation.

   Answered 2025-08-20 by Helen Castro

Are you worried about the complexity of M-language? I find the GUI in Power Query can do 90% of the work without you ever having to write a single line of code.

   Answered 2025-08-22 by Paul Higgins

• Paul is right, the UI is very capable. Ronald, one tip: if you do use Power Query, try to put your "Filter" and "Remove Columns" steps at the very beginning. This triggers Query Folding on SQL sources, meaning the database does the heavy lifting and only sends the necessary rows to Power BI. This is much faster than downloading 1 million rows and then filtering them down to 10,000 inside the Power BI application itself.

     Commented 2025-08-24 by Scott Thornton

Calculated columns also make your .pbix file size explode. If you're on the Free or Pro tier with the 1GB limit, you really need to stick to Power Query for transformations.

   Answered 2025-08-25 by Donna Myers

• True that, Donna. I once reduced a file from 800MB to 200MB just by moving five complex calculated columns into the Power Query stage. It’s a massive difference.

     Commented 2025-08-23 by Ronald Fishe