Join us at FabCon Atlanta from March 16 - 20, 2026, for the ultimate Fabric, Power BI, AI and SQL community-led event. Save $200 with code FABCOMM.
Register now!Calling all Data Engineers! Fabric Data Engineer (Exam DP-700) live sessions are back! Starting October 16th. Sign up.
Hello everyone,
I have been using Power Query to transform a single table held in SQL into a dimensional model. It works as follows:
1. Single table in SQL server which holds the entire dataset. ~ 2.5 million rows.
2. One query in Power Query which connects to this table, performs steps to remove unnecessary columns, de-duplicate the data and add an index column. This forms a dimension.
3. One query in Power Query which connects to this table, does a merge to the newly created dimension table on the fields held within the dimension to go and get the index columm. The merge makes a perfect match each time.
4. In the model view the two tables are joined by the index column in the relationship.
When refreshing the dataset the numbers being shown keep changing; sometimes they are right, sometimes they are wrong. It appears the index column is not being passed correctly over from the dimension to the fact in the merge when it refreshes. In Power Query everything looks fine. Interestingly It is only ever the index that is wrong. If I merge back any of the dimensional columns back into the fact, and keep the original columns insde the fact they all match. It is only ever the index that is incorrect.
Refresh 1:
Refresh 2:
Refresh 3:
Has anyone ever encountered this before and any tips for why it is happening / how to stop it happening? The above method seems pretty standard way of creating a dimensional model from a single table so unsure why this would be happening.
Any help is much appreicated.
Thanks,
Alex
Solved! Go to Solution.
I found adding an explicit sort step in Power Query after doing the de-duplication fixed the issue.
I found adding an explicit sort step in Power Query after doing the de-duplication fixed the issue.
Hello - can you please post your m-queries?
Join the Fabric FabCon Global Hackathon—running virtually through Nov 3. Open to all skill levels. $10,000 in prizes!
Check out the October 2025 Power BI update to learn about new features.