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

Enhance your career with this limited time 50% discount on Fabric and Power BI exams. Ends August 31st. Request your voucher.

Reply
Databasic
Regular Visitor

Model relationships breaking unexpectedly when visualizing Planning table columns in Power BI

Hi

We are going to change ERP system and I have to alter my data so we can upload it in the right format to our ERP system.

Here’s my current setup:

  • All relevant columns are formatted as Text via Power Query.

  • I use a central Name table, which contains unique rows per 4-digit-code.

  • This code links both to Planning and RM_Film via bidirectional one-to-many relationships (Name is on the "one" side).

  • There are no blanks or mismatches—codes are clean.

Databasic_2-1752217056279.png

Additionally:

  • I created a small Fixed table with static values, joined to Name using an Index column.

  • I use calculated columns like this:

GEWEI =

VAR Unit = RELATED('Name'[MEINS])
RETURN
SWITCH(
        Unit,
        "KG", "Kg",
 "M2", "G",
         BLANK()
    )

I’m trying to build a table visual containing selected columns from Name, RM_Film, and Planning, structured to match our ERP import format. It needs to be dynamic so we can run multiple simular reports when we test our ERP.

What’s baffling:

  • Initially, making the relationships bidirectional solved the problem and let me build the table as expected.
  • Later, after adding more columns (incl. calculated columns), the same error returned: “relationship not found” when pulling anything from Planning.
  • I rolled back by removing those columns—but the error persisted.
  • I even recreated the entire file and again reached a point where the visual worked—until I added a Planning column and it broke, although the relationships hadn’t changed.
  • Everything else seems stable. The same issue doesn't happen when I build a separate visual with just Planning and Name.

What it should look like
CLASS_01DWERKVERSGKTGRMMTPOSMTVFPTRAGRLADGRPRCTRSTAWNMTVERHERKLHERKREKGRPEKWSL
FPFILMBE401YYNORMF111GC-AAV481159001BE B10A002
FPPAPBE401YYNORMF111GC-AAV480255151DE B10A002

Reason why I don't use Power Query in Excel:
  • Our database server runs Excel 2010, which lacks modern Power Query features.

  • I prefer to visualize and manage relationships transparently in Power BI.

  • It worked yesterday—and now it doesn’t. I’d love insight into what could be causing this inconsistency.

Any help or ideas would be amazing. Thanks!

1 ACCEPTED SOLUTION
rajendraongole1
Super User
Super User

Hi @Databasic  - Only keep Name → Planning and Name → RM_Film relationships as single-directional (one-to-many, from Name to Planning/RM_Film). That’s usually safest.

 

When both are set to bidirectional, Power BI may struggle to resolve filter propagation, especially when you try to pull data into a single visual that includes columns from all three tables errors like , ambiguous filter paths. Avoid bidirectional joins unless truly needed.

suggest you, make them single-direction from Name → Planning and Name → RM_Film.Use that table for your ERP export visuals/reports — reliable and isolated from model quirks.If needed, use slicers from Name or Fixed tables to make dynamic test versions.

Bidirectional relationships and ambiguity in DAX - SQLBI

 

Hope this helps.

 

 





Did I answer your question? Mark my post as a solution!

Proud to be a Super User!





View solution in original post

3 REPLIES 3
v-pnaroju-msft
Community Support
Community Support

Thankyou, @rajendraongole1, for your response.

Hi Databasic,

We appreciate your question on the Microsoft Fabric Community Forum.

Based on my understanding of the issue, please find attached a screenshot and a sample PBIX file that may assist in resolving the matter:

vpnarojumsft_0-1752507006487.png

We hope the information provided helps to resolve the issue.
Should you have any further queries, kindly feel free to contact the Microsoft Fabric community.

Thank you.

rajendraongole1
Super User
Super User

Hi @Databasic  - Only keep Name → Planning and Name → RM_Film relationships as single-directional (one-to-many, from Name to Planning/RM_Film). That’s usually safest.

 

When both are set to bidirectional, Power BI may struggle to resolve filter propagation, especially when you try to pull data into a single visual that includes columns from all three tables errors like , ambiguous filter paths. Avoid bidirectional joins unless truly needed.

suggest you, make them single-direction from Name → Planning and Name → RM_Film.Use that table for your ERP export visuals/reports — reliable and isolated from model quirks.If needed, use slicers from Name or Fixed tables to make dynamic test versions.

Bidirectional relationships and ambiguity in DAX - SQLBI

 

Hope this helps.

 

 





Did I answer your question? Mark my post as a solution!

Proud to be a Super User!





This web page gave me the right understanding of relations to troubleshoot. I had to dive into the data and was able to pull out some duplicates to remove the indirect many to many relation. Resolved, thanks a lot 💪

Helpful resources

Announcements
August Power BI Update Carousel

Power BI Monthly Update - August 2025

Check out the August 2025 Power BI update to learn about new features.

August 2025 community update carousel

Fabric Community Update - August 2025

Find out what's new and trending in the Fabric community.