GROUPBY with filter conditions on columns

[Anonymous]

Hello,

Is it possible to use GROUPBY with some sort of conditions on the columns?  Something along the lines of 

GROUPBY(

    'Employee Detail',

    'Employee Detail'[Cost Ctr],

    "Count of Primary", COUNTAX(CURRENTGROUP(), 'Employee Detail'[Employee Category] = "Primary Department"),

    "Count of Residual", COUNTAX(CURRENTGROUP(), 'Employee Detail'[Employee Category] = "Residual Employees")

    )

The first column (Primary Department) works but the 2nd column (Residual Employees) just repeats the same values as the 1st column.  I have maybe a dozen of these columns I want to add with conditions.  I know we cannot use CALCULATE with GROUPBY so maybe this is just the wrong function and I should stick with SUMMARIZE or SUMMARIZECOLUMNS instead?  What is best practice for grouping data like this with conditions?  Thanks!

-Scott

[johnt75]

I think you'd need to use ADDCOLUMNS ... SUMMARIZE, like

Summary Table =
ADDCOLUMNS (
    SUMMARIZE ( 'Employee Detail', 'Employee Detail'[Cost Ctr] ),
    "Count of Primary",
        CALCULATE (
            COUNTROWS ( 'Employee Detail' ),
            'Employee Detail'[Employee Category] = "Primary Department"
        ),
    "Count of Residual",
        CALCULATE (
            COUNTROWS ( 'Employee Detail' ),
            'Employee Detail'[Employee Category] = "Residual Employees"
        )
)

[amitchandak]

@Anonymous , I think you can do some calculations using Earlier

Earlier: https://youtu.be/CVW6YwvHHi8

If this does not help
Can you share sample data and sample output in table format? Or a sample pbix after removing sensitive data.