Break out Multi-Select field into Slicer

[GabeFig]

The (2) images on the left are what the data looks like in the software that is being exported into MS BI.
MS BI is interpreting them as shown in the right image.

The Slicer for both the Category & Feature needs to be split as on the left image but MS BI is creating an entry for every unique selection as shown on the right.
When downloading either of these fields into .csv they appear separated by semi-colon.

[v-ssriganesh]

Hello @GabeFig,

Hope everything’s going great on your end! Just checking in has the issue been resolved, or are you still running into problems? Sharing an update can really help others facing the same thing.

Thank you.

[GabeFig]

Thank you & I apologize for not responding earlier as I was out a couple of days. I did not expect such quick response as this is my 1st time seeking help.

[v-ssriganesh]

Hello @GabeFig,
Thank you for reaching out to the Microsoft Fabric Forum Community.

Power BI treats these fields as text strings when imported, rather than recognizing them as separate selectable items.

To resolve this, you can use Power BI's "Split Column" feature to break the semicolon-separated values into rows. Here’s a general approach:

• Import the .csv file into Power BI and click on the Categories or Features column in the Power Query Editor.
• Go to the Home tab, click Split Column and choose By Delimiter and opt to split into rows (not columns) and Click OK and then Close & Apply to refresh the data model.
• Add slicers for the split Categories and Features columns.

If this information is helpful, please “Accept as solution” and give a "kudos" to assist other community members in resolving similar issues more efficiently.
Thank you.

[GabeFig]

I went through your steps as well as refreshed but the data seems to be the same. I do not see any new columns to use in the Slicer.

Maybe this has to be rebuild rom scratch

[v-ssriganesh]

Hello @GabeFig,
Thanks for the follow-up.

The issue might be due to the field being split into columns instead of rows, or possibly a modeling step was skipped. Let me walk you through the approach again, this time with clarity on creating a slicer-ready field:

• In Power BI, go to Transform Data to open Power Query Editor and Select your Categories column.
• From the top ribbon, click Split Column > By Delimiter and Choose delimiter: ; (semicolon) also note that in the dialog, select Split into Rows, not columns.
• Repeat the same for the Features column (if needed) and Click Close & Apply to update the data model.

• After applying changes, return to your Power BI report canvas and drag the now “flattened” Categories column into a slicer visual.

Best Regards,
Ganesh singamshetty.

[GabeFig]

Thanks Ganesh for the more detailed instruction BUT that seems to be what I did. 
I could not tell there was anything different even after the original slicer was removed and recreated w/ Category field that was split (by rows).

Below are some screenshots of what was done with the existing MS BI.I wonder that this needs to be done completely from scratch to work.

 Both Catefory & Feature were done. The split was done by Row.

I also checked to see if there was a hidden CR or LF or CRLF but there were none.

[v-ssriganesh]

Hello @GabeFig,
Thanks for sharing your detailed screenshots and you’ve done a great job applying the split correctly. I now suspect the issue is not with the split step itself, but with how Power BI handles row-level filtering from slicers in a 1-to-many relationship scenario.

To make the slicers work properly without affecting or duplicating your main data table:

• Duplicate your main table in Power Query call it CategoryMapping and Keep only ID and Categories columns.
• Apply Split Column by Delimiter Into Rows on Categories and Rename the result column to Category. Repeat same steps to create a FeatureMapping table.
• Create Relationships between CategoryMapping[ID] >> MainTable[ID] and FeatureMapping[ID]>> MainTable[ID]
• Then Use CategoryMapping[Category] in the Category slicer and FeatureMapping[Feature] in the Feature slicer

If the issue persists, try to share your .pbix file or sample data so I can better understand the scenario and give you the necessary solution.

Thank you.

[GabeFig]

[v-ssriganesh]

Hi @GabeFig,
Thanks again for your detailed follow-up and for sharing the screenshot.

What you're seeing with the "Business" category behaving differently is due to how Power BI handles split-to-rows transformations on delimited text. When a category like "Application; Business" is split, Power BI creates separate rows one with "Application", and another with "Business". However, "Business" also appears as a standalone value in other rows, which may result in multiple "Business" values in your slicer or inconsistent filtering behavior.

To make sure the "Business" category filters all relevant rows whether it appears alone or with others. We suggest the following steps:

Duplicate your main table and name it CategoryMapping.

• Keep only ID and Categories columns and Split Column by Delimiter >> choose ; and split into rows.
• Trim whitespace from the split column to clean up any extra spaces and rename the new column to Category then Remove duplicates so that "Business" appears only once per ID.

• Create a relationship between CategoryMapping[ID] to MainTable[ID] and Use CategoryMapping[Category] in your slicer.

If the issue persists try to share your sample data: How to provide sample data in the Power BI Forum - Microsoft Fabric Community

Best Regards,
Ganesh singamshetty.

[GabeFig]

Software 

The CategoryMapping & FeatureMapping look great but again if 'application' is selected it will show only those that have just 'application'.

If the item has 'application; cloud' then that will not show.

If 'application' & 'cloud' are chosen then only those that have either 'application' or 'cloud' but not those that have 'application; cloud'.

The problem seems to be the Manage Relationship' I've tried several if not all variations.

It seems best to use the Categories field cause if the Records ID# is used it only returns (1) as w/ the Feature

[v-ssriganesh]

Hi @GabeFig,
Thanks again for the follow-up and the screenshot.

To better assist you, could you please provide sample data that fully reflects the issue? Kindly share the data in a usable format (such as Excel or a simple table), not as a screenshot. The sample should:

• Clearly show the structure you're working with
• Include enough data to replicate the issue
• Exclude any sensitive or unrelated information
• Include the expected result/output based on that data

If you're unsure how to provide data in the forum, please refer to this helpful guide:
How to provide sample data in the Power BI Forum - Microsoft Fabric Community

Once we have that, I’ll be happy to help you resolve it.

Best regards,
Ganesh Singamshetty

[GabeFig]

Record ID#	Software/Service	Categories	Features
705	Zoominfo	Unknown	Unknown
498	Zoom Outlook add-in	Unknown	AI-Enabled
559	Zoom	Cloud	AI-Enabled;Communication
707	Zkipster	Unknown	Unknown
997	World Trademark Review	IP Management;Legal & Market Research	Unknown
998	World Intellectual Property Review	IP Management;Legal & Market Research	Unknown
843	Working Advantage	Unknown	Unknown
589	Work Management System	Application;Cloud	Unknown
459	WordRake	Unknown	Unknown
277	Word	Application;Business	Unknown
999	Women's Wear Daily	Legal & Market Research	Unknown
557	Windows OS	Application;Commercial	Unknown
452	WinDJView	Unknown	Unknown
995	Win Remote	Business	Unknown
1003	Washington Business Journal	Legal & Market Research	Unknown
851	Walkingspree App	Application	Unknown
1006	VitalLaw	Legal & Market Research;Legal Education	Unknown
948	Vault Verify	Automation	Unknown
961	Valimail	Regulatory & Compliance	Unknown
590	Vacation Reporting Tool	Unknown	Unknown
910	V Drive File Share	Collaboration & Communication	Unknown
423	USDHS Check Requests	Application;Cloud	Unknown
1007	USCIS Policy Manual	Legal & Market Research	Unknown
945	U Drive File Share	Collaboration & Communication	Unknown
954	Tritech	Collaboration & Communication	Unknown
884	Trexo Global	IP Management	Unknown
980	Travelling Coaches	Legal Education	Unknown
1014	Trademark Navigator	Legal & Market Research	Unknown
580	Ticket Tracker	Application;Cloud	Unknown
988	Threat Manager	Regulatory & Compliance	Unknown
1002	The Washington Post	Legal & Market Research	Unknown
1036	The Kluwer IP Law Manual	IP Management;Legal & Market Research	Unknown
588	Task Management System	Application;Cloud	Unknown
909	T Drive File Share	Collaboration & Communication;Project Management	Unknown
601	Synchronizer Player	Unknown	Unknown
717	Symphony Web Services Admin	Unknown	Unknown
798	Swoogo	Unknown	Unknown
1063	Supreme Court Brief	Legal & Market Research	Unknown
759	Supabase	Unknown	Unknown
708	SunIP	Unknown	IP Management;Legal Analysis & Strategy

[v-ssriganesh · Accepted solution]

Hello @GabeFig,
Thanks for sharing the sample data.

I have reproduced your scenario using the sample data you shared and implemented the required filtering logic based on multi-valued Categories and Features columns (with delimiters like ;).

Output:

• When “Application” is selected from the Categories slicer:
  You will see only rows where “Application” is one of the multiple categories.
  
  
  
• When “AI-Enabled” is selected from the Features slicer:
  You will see only rows where “AI-Enabled” is one of the feature values.
  

For your reference, I’m attaching the .pbix file.

Best regards,
Ganesh Singamshetty

[GabeFig]

Ganesh

  Thank you so very much!

  I believe I have got the BI showing as requested.

  Your patience is above and beyond!

 Gabe

[Greg_Deckler]

@GabeFig Perhaps try creating a group column (right-click the column and choose "New group"). Group your values into your desired buckets and then use this new column in your slicer.