Forum Discussion

subhendude's avatar
subhendude
Microsoft Employee
3 years ago
Solved

Filtering matrix rows by a column value

I've a dataset which has few test cases running in two locations.

I'm able to pivot it using the matrix table like below.

However I would like to show it using a matrix table with the rows where the test cases didn't run for either of the location like below

Can anyone help how to show this in a matrix table?

 

Thanks!

  • Hi subhendude 
    You can achieve the desired result by adding 3 measures:
    Firstly counters for Paris and London test cases :

    LondonCount = CALCULATE(COUNTROWS('Table'),FILTER('Table','Table'[Location]="London"))
    ParisCount = CALCULATE(COUNTROWS('Table'),FILTER('Table','Table'[Location]="Paris"))
    At this stage we will get your "pivoted" table:

    The next step is to create a "flag" measure which you can use as a filter:

    TestCount = if([LondonCount]=0 || [ParisCount]=0,1,0)
    Now you can put it on the visual filter and filter by its values is 1:

    If you need red formatting for blank cells, you can base it on the first measures too.

     

    Result:

    Link to a sample file 

     

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


2 Replies

  • Hi subhendude 
    You can achieve the desired result by adding 3 measures:
    Firstly counters for Paris and London test cases :

    LondonCount = CALCULATE(COUNTROWS('Table'),FILTER('Table','Table'[Location]="London"))
    ParisCount = CALCULATE(COUNTROWS('Table'),FILTER('Table','Table'[Location]="Paris"))
    At this stage we will get your "pivoted" table:

    The next step is to create a "flag" measure which you can use as a filter:

    TestCount = if([LondonCount]=0 || [ParisCount]=0,1,0)
    Now you can put it on the visual filter and filter by its values is 1:

    If you need red formatting for blank cells, you can base it on the first measures too.

     

    Result:

    Link to a sample file 

     

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