Decomposition Tree Showing (Blank) Values Despite Data Filtering

[nymoslike]

Hello everyone,

I'm currently facing an issue with the Decomposition Tree visual in Power BI, where I'm seeing "Blank" values in various levels of the hierarchy, despite applying several data cleaning and transformation steps.

Scenario: I've created a Decomposition Tree using sales data with the following structure:

• Levels: Region, Product Category, Product, Sales Representative.

• Measure: Sales Amount to analyze the data.

I split the data into multiple levels (Level 1, Level 2, Level 3, Level 4) based on the product and sales structure, and ensured that I filtered out all null and empty values using Power Query transformations.

Despite this, the Decomposition Tree is still displaying "(Blank)" values at several levels, which makes the visualization less useful and cluttered. I've made sure to remove nulls and even added conditional columns to replace blanks with default values like "Unknown", but the blanks persist.

What I've Tried: I have also looked into several existing solutions and tried implementing them, but unfortunately, they have not worked for my case. Here are some of the references I've used:

• Dealing with blank/null values in Dynamic Decomposition Trees: I tried the solutions discussed here, but I might have applied something incorrectly as the blanks still persist.

• Need to hide blanks in decomposition tree: This thread discusses a similar issue, but unfortunately, no solution was found.

• I also tried the approach mentioned in this video using ISINSCOPE. This method seemed promising, but it only worked in a Matrix visual and not for my Decomposition Tree.

Attached Example Data: I've attached some sample data to reproduce the issue. Some Sales Rep and other values are missing to simulate blank fields, yet these blanks still appear in the visualization even when I explicitly filtered them out in Power Query.

My Goal: I want to ensure that the Decomposition Tree only displays levels and branches where valid values exist, and completely hides any branches that would otherwise show "(Blank)".

Has anyone faced similar issues or have suggestions on how to prevent blanks from appearing in the Decomposition Tree? Any guidance would be greatly appreciated.

Thank you in advance!

Attachments:

• Example CSV data with blank fields for reproduction:
  Order ID;Customer Name;Product;Category;Sales Amount;Order Date;Region;Sales Rep
  101;John Doe;Laptop;Electronics;1200;15.01.2023;East;Alice Smith
  102;Jane Smith;Office Chair;Furniture;300;10.02.2023;West;Bob Brown
  103;Alan Brown;Smartphone;Electronics;800;25.01.2023;North;Carol White
  104;Emily Davis;Standing Desk;Furniture;450;05.03.2023;South;Alice Smith
  105;Michael Scott;Printer;Office;150;20.01.2023;East;Bob Brown
  106;Sarah Connor;Monitor;Electronics;250;18.02.2023;West;Carol White
  107;David Smith;Coffee Machine;Kitchen;100;15.03.2023;East;Alice Smith
  108;Laura Lee;Tablet;Electronics;600;08.03.2023;North;
  109;George Miller;Bookcase;Furniture;200;29.01.2023;South;Carol White
  110;Rachel Green;Whiteboard;Office;75;12.02.2023;East;Alice Smith
  111;Anna Bell;Keyboard;Electronics;100;22.03.2023;West;
  112;Chris Rock;Mouse;Electronics;50;05.02.2023;North;
  113;Patricia Ray;Desk;Furniture;400;18.01.2023;East;
  114;Robert Paul;Headphones;Electronics;150;24.02.2023;West;
  115;Kate Winslet;Lamp;Furniture;80;02.03.2023;South;Carol White
  116;Peter Pan;Router;Electronics;120;31.01.2023;North;Alice Smith
  117;Luke Skywalker;Notebook;Office;25;13.03.2023;East;Bob Brown
  118;Leia Organa;Webcam;Electronics;70;09.02.2023;West;
  119;Clark Kent;Chair;Furniture;150;22.01.2023;North;Alice Smith
  120;Bruce Wayne;Monitor Stand;Office;40;19.03.2023;East;Bob Brown
  121;Diana Prince;Graphics Tablet;Electronics;200;25.02.2023;South;Carol White
  122;Steve Rogers;Speaker;Electronics;180;11.01.2023;West;Alice Smith
  123;Tony Stark;Desk Lamp;Furniture;60;04.03.2023;East;Bob Brown
  124;Natasha Romanoff;Wireless Charger;Electronics;30;28.01.2023;North;Carol White
  125;Bruce Banner;Coffee Table;Furniture;250;16.02.2023;South;Alice Smith
  126;Thor Odinson;Paper Shredder;Office;110;10.03.2023;West;Bob Brown
  127;Loki Laufeyson;Standing Lamp;Furniture;90;17.01.2023;North;
  128;Sam Wilson;Laptop Stand;Office;45;26.02.2023;East;Alice Smith
  129;Bucky Barnes;Projector;Electronics;500;29.01.2023;South;
  130;Wanda Maximoff;Office Drawer;Furniture;220;11.03.2023;West;Carol White
  131;Vision;Surge Protector;Electronics;35;15.02.2023;North;Alice Smith
  132;Stephen Strange;Whiteboard Marker;Office;15;13.01.2023;East;Bob Brown
  133;Scott Lang;Chair Mat;Furniture;70;07.03.2023;West;Carol White
  134;Hope Pym;USB Hub;Electronics;25;28.02.2023;South;Alice Smith
  135;T'Challa;Wireless Mouse;Electronics;50;23.01.2023;North;
  136;Shuri;Pen Holder;Office;10;18.03.2023;East;Carol White
  137;Okoye;Stapler;Office;15;06.02.2023;West;Alice Smith
  138;M'Baku;;Electronics;300;27.01.2023;South;
  139;Nakia;;Electronics;20;14.03.2023;North;
  140;Peter Parker;Notebook Stand;Office;35;17.02.2023;East;Alice Smith
  141;Miles Morales;Desk Organizer;Furniture;50;12.01.2023;West;Bob Brown
  142;Gwen Stacy;Wall Clock;Furniture;45;21.03.2023;South;Carol White
  143;Eddie Brock;;Electronics;85;19.02.2023;North;Alice Smith
  144;Cletus Kasady;Desk Pad;Office;30;14.01.2023;East;Bob Brown
  145;Norman Osborn;;Electronics;25;12.03.2023;West;Carol White
  146;Harry Osborn;File Cabinet;Furniture;300;03.02.2023;South;Alice Smith
  147;Otto Octavius;Printer Paper;Office;20;26.01.2023;North;Bob Brown
  148;Flint Marko;Trash Can;Furniture;35;09.03.2023;East;Carol White
  149;Quentin Beck;;Electronics;40;20.02.2023;West;
  150;Adrian Toomes;Cork Board;Office;25;21.01.2023;North;Bob Brown
  
  

[Bibiano_Geraldo]

Hi, You have two ways: Replace null values or take off null values:

1- Replace null values (i recomend): In Power Query, in your Product column replace empty values with any String you want:

1.

2. Replace null with any string you want:

Closse and apply, now you decomposition should look like this:

Note: Do the same for other columns.

2- Take Off Null values:

Now your decomposition should look like this, but note, taking off null values affect your sales amount value:

[nymoslike]

Hello Bibiano_Geraldo,

Thank you for your detailed response and the step-by-step images — I really appreciate it!

Unfortunately, applying your methods to my dataset causes me to lose crucial data. My goal is to create a hierarchical visualization that includes all data, with null or empty fields representing "leaf nodes" that should not be filtered out.

The issue is that the Decomposition Tree seems to treat these nulls as values to eliminate or replace, but they are a valid part of my hierarchy, marking the end of a branch. I’ve read that the Decomposition Tree is more suited for metric analysis, which may explain the challenge.

Do you have any other suggestions for representing this kind of hierarchy? I’ve also found that most native Power BI visuals or marketplace alternatives either don't work well for my needs or require paid options.

Thanks again for your help — any further ideas would be greatly appreciated!