Power BI Matrix Row Calculations

[Anonymous]

Here is my source data:

Here is the matrix I am trying to create in Power BI:

I am running into a few issues such as performing the Count/Sale calculation and assigning different formatting to each individual row.  I can get the Count and Sale columns formatted but I am lost at how to calculat the Count/Sale row.

[v-yingjl]

Hi @Anonymous ,

To achieve this, need to create a new table like this:

Use union() to combine two tables:

test = UNION('Table','Avg table')

Create a calculated column in 'test' table:

result = 
VAR _sale =
    CALCULATE (
        SUM ( test[Value] ),
        ALLEXCEPT ( 'test', test[Year], test[Region] )
    )
VAR _count =
    CALCULATE (
        SUM ( 'test'[Value] ),
        ALLEXCEPT ( test, test[Year], test[Region] ),
        'test'[Measure] = "Count"
    )
RETURN
    IF (
        'test'[Measure] = "Count"
            || 'test'[Measure] = "Sale",
        'test'[Value],
        _sale / _count
    )

Use a Matrix visual to show the result:

Attached my sample file that hopes to help you, please check and try it: Power BI Matrix Row Calculations.pbix

Best Regards,
Yingjie Li

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

[Anonymous]

I tried using this solution.  One issue in my matrix is I need to show Sales and Count as a SUM, but the Result as an Average.  How could I do this?

[v-yingjl · Accepted solution]

Hi @Anonymous ,

Create this measure to show the result:

avg = 
VAR _sale =
    CALCULATE (
        SUM ( test[Value] ),
        ALLEXCEPT ( 'test', test[Year], test[Region] )
    )
VAR _count =
    CALCULATE (
        SUM ( 'test'[Value] ),
        ALLEXCEPT ( test, test[Year], test[Region] ),
        'test'[Measure] = "Count"
    )
RETURN
    IF (
        ISINSCOPE ( 'test'[Measure] ),
        IF (
            SELECTEDVALUE ( 'test'[Measure] ) = "Count"
                || SELECTEDVALUE ( 'test'[Measure] ) = "Sale",
            SUM ( 'test'[Value] ),
            _sale / _count
        ),
        SUM ( test[Value] ) + _sale / _count
    )

Best Regards,
Yingjie Li

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

[lbendlin]

Don't use these aggregation functions. Instead create separate measures as I suggested.  Then you can control the behavior both inside the visual and in the totals (via HASONEVALUE() etc)

[amitchandak]

@Anonymous , refer if this can help

https://www.kasperonbi.com/use-isinscope-to-get-the-right-hierarchy-level-in-dax/

https://www.sqlbi.com/articles/clever-hierarchy-handling-in-dax/

[lbendlin]

when you add your measures to the Values area of the matrix visual you can select the option "show on rows". That will give you the desired outcome even with different data types for your measures.