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 nowJuly 7 - July 17 | Round 2 of the Power BI Dataviz World Championships. Don't miss your chance! Learn more
Is there a way to remove duplicates without breaking query folding? I need the folding (for incremental refresh), but I have a ton of duplicates (I go from 3 0378 271 rows to 10 472 after removing duplicates)...
Use a Group By transform instead - that will fold on SQL data sources.
OPTIONAL - If you don't want an aggregate column added to the rowset (so you want it to behave functionally exactly like a remove duplicates), then edit the M that gets generated to remove the aggregate. For example, turn this:
= Table.Group(Source, {"Grouping Column 1","Grouping Column 2"}, {{"Count", each 1, type Int64}})
into this:
= Table.Group(Source, {"Grouping Column 1","Grouping Column 2"}, {})
Yeah, exactly.
Hi @grggmrtn ,
This can happen if the date ranges are set NOT up to as follows: [Date] >= RangeStart and [Date] < RangeEnd. Is your upper bound set to be equal to or less than or just less than?
Refer to this thread:
Best Regards,
Liang
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.
Hi @V-lianl-msft thanks for the reply...
Your answer would make sense if I had actually set ranges, but I haven't gotten to setting the ranges yet, because query folding on my dataset breaks. The final step in my query (remove duplicates) breaks the folding.
So my question is, how can I remove the duplicates without breaking the folding?
I could use this as well, did you ever find out how to do this?
Table.Distinct() breaks the fold. How do we tell the servers to return a distinct list after all our previous steps?
Never mind. Table.Distinct() only breaks the fold if you use it to remove duplicates from a subset of the tables columns, instead of the whole table. Using Table.Distinct() on the entire table does not break the query folding.
Join us in Barcelona for FabCon and SQLCon, the Fabric, Power BI, SQL, and AI community event. Save €200 with code FABCMTY200.
Join Data Days 2026: 60 days of free live/on-demand sessions, challenges, study groups, and certification opportunities.
| User | Count |
|---|---|
| 27 | |
| 23 | |
| 18 | |
| 18 | |
| 15 |
| User | Count |
|---|---|
| 62 | |
| 44 | |
| 42 | |
| 37 | |
| 37 |