Skip to main content
cancel
Showing results for 
Search instead for 
Did you mean: 

Join us at the 2025 Microsoft Fabric Community Conference. March 31 - April 2, Las Vegas, Nevada. Use code FABINSIDER for $400 discount. Register now

Reply
kvidales
Frequent Visitor

Does a new column in a Common Data Model (CDM ) folder reflect automatically in a Power BI dataflow?

Hello Community!

 

I’m using a Power BI dataflow that was configured from a Common Data Model (CDM) folder in Azure Data Lake. Recently, I was informed that a new a new column will be added to an existing table in the CDM witch is called in my Dataflow, but I’m unsure if this new column will automatically be reflected in the dataflow or if additional steps are needed (e.g., recreating the dataflow).

 

Since dataflows created from CDM folders cannot be edited directly in the Power BI service, has anyone faced a similar case and could share how they handled it? Do I need to recreate the dataflow, or is there a process to update entities automatically?

 

Any shared experience or official references would be greatly appreciated

2 REPLIES 2
v-kongfanf-msft
Community Support
Community Support

Hi @kvidales ,

 

When you add a new column to an existing table in the CMD folder, it is not automatically reflected in the Power BI data flow. You need to take additional steps to update the data flow to include the new column.


From my understanding, when new columns are added to a data source, they are not automatically updated to the Power BI data flow even after a successful refresh. To include the new columns, you may need to delete the existing table and rejoin it in the Power Query editor.

 

Best Regards,
Adamk Kong

 

If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.

Hello, and thank you for your comment!

Actually, that’s an important point: dataflows created from CDM cannot be edited directly in the Power BI service. While you can consume them, editing their structure isn't supported.

Given this limitation, do you think I’ll need to drop the existing dataflow and recreate it for the new column to be reflected? We haven’t added the column yet, but I’m trying to anticipate or estimate the impact. Since we have several semantic models connected to this dataflow, dropping and recreating it would mean updating the dataflow ID in all of them, which could become a significant task.

Looking forward to hearing your thoughts!

Helpful resources

Announcements