This time we’re going bigger than ever. Fabric, Power BI, SQL, AI and more. We're covering it all. You won't want to miss it.
Learn moreDid you hear? There's a new SQL AI Developer certification (DP-800). Start preparing now and be one of the first to get certified. Register now
I am trying to find an efficient way to filter a data set using powerquery. I need to filter out all rows representing a company only when there is no data in another column. But I need to retain all the rows of a company when there is at least one row of data in that column. So for the below table, I need to filter out all rows of Company B as it does not have any data in the "Descr" column. But I need to retain all rows of Company A (including those where the Descr column is empty) and Company C.
I am currently using a two-step process. The first step involves finding companies like A and C by filtering the Descr column for non-null values and then creating a company list (after removing the duplicates). Then I apply inner join as seen below to filter out Company B.
Is there a straightforward way to perform this filter?
Thank you.
Solved! Go to Solution.
NewStep=Table.Combine(Table.Group(YourTable,"Company Name",{"n",each if List.NonNullCount([Descr])=0 the #table(0,{}) else _)})[n])
Thank you. It did work. But it does take time for a large data set over 100K rows.
NewStep=Table.Combine(Table.Group(YourTable,"Company Name",{"n",each if List.NonNullCount([Descr])=0 the #table(0,{}) else _)})[n])
Check out the April 2026 Power BI update to learn about new features.
Sign up to receive a private message when registration opens and key events begin.
If you have recently started exploring Fabric, we'd love to hear how it's going. Your feedback can help with product improvements.