How do I combine Group_By and Lag?

[calcstuff]

Employee	Sales	Lag	LagGroup
Billy	1
Billy	2	1	1
Billy	3	2	2
Bobby	5	3
Bobby	10	5	5
Bobby	15	10	10

I know it's fairly easy to generate the Lag column shown above, based off of the Sales column. Instead, can I first group by Employee and then add the custom LagGroup column, as shown above? If you notice, Bobby's first LagGroup is blank, as the intent is to group by employee, evaluate lags, and not pull in lag values from other employees. What power query or what DAX do I need to do? This is all done in a dataflow if that matters.

[v-yinliw-msft · Accepted solution]

Hi @calcstuff ,

You can try this method:

In the Power Query, you can add a index column:

Then new a column:

LagGroup =
VAR _index =
    CALCULATE (
        MAX ( 'Table'[Index] ),
        FILTER (
            'Table',
            [Employee] = EARLIER ( 'Table'[Employee] )
                && [Index] < EARLIER ( 'Table'[Index] )
        )
    )
RETURN
    CALCULATE ( MAX ( 'Table'[Sales] ), FILTER ( 'Table', [Index] = _index ) )

The result is:

Hope these help you.

Here is my PBIX file.

Best Regards,

Community Support Team _Yinliw

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