topic Average difference between two MIN dates (Summarizing Measures with Averages) in Desktop

Average difference between two MIN dates (Summarizing Measures with Averages)

SebL — Fri, 11 May 2018 21:15:42 GMT

Hi Guys,

First post.

I'm trying to create a measure that calculates the average difference between dates. To make things more complicated, these dates fields come from tables that contain multiple rows for the same "user" and I have to first find the MIN of those dates. On top of that, the measure must determine which dates are available and pick the two to subtract.

Measure 1 = Min(Table1[Date1])
Measure 2 = Min(Table2[Date2])
Measure 3 = Min(Table3[Date3])

Measure 4 = IF(ISBLANK([Measure 1]),[Measure 2]-[Measure 3],[Measure 1]-[Measure 3])

When I drop Measure 4 into a table to see the value for each user, the values are correct. However, if I remove the User ID and replace it with the Year the user signed up, the values for Measure 4 can't be shown as user averages for those years. The years are not a date used in the measures and reside on the same table as the User ID.

How can I summarise a measure of this kind as an average?

Maybe I'm taking the wrong approach to this problem.

Re: Average difference between two MIN dates (Summarizing Measures with Averages)

OwenAuger — Sat, 12 May 2018 07:34:22 GMT

Hi SebL

From your description, you are wanting to calculate the average of the expression in [Measure 4] over Users in the current filter context. You can use AVERAGEX to do this.

Something like this should do the trick (I'm guessing the table containing the User ID column table is called Users):

Measure 4 Average over Users =
AVERAGEX (
    VALUES ( Users[User ID] ),
    IF (
        ISBLANK ( [Measure 1] ),
        [Measure 2] - [Measure 3],
        [Measure 1] - [Measure 3]
    )
)

Or this should work as well:

Measure 4 Average over Users v2 =
AVERAGEX (
    Users,
    IF (
        ISBLANK ( [Measure 1] ),
        [Measure 2] - [Measure 3],
        [Measure 1] - [Measure 3]
    )
)

Regards,

Owen

Re: Average difference between two MIN dates (Summarizing Measures with Averages)

SebL — Mon, 14 May 2018 15:49:04 GMT

Thanks, this worked perfectly.

I'm wondering if I could add a follow-up question.

I now have the average number of days per user being displayed over time. How would I go about creating a measure that displays the percentage of those users that are under a certain threshold of days?

Re: Average difference between two MIN dates (Summarizing Measures with Averages)

OwenAuger — Wed, 16 May 2018 09:27:57 GMT

That's good news :)

I'm assuming you still have [Measure 4] which can be evaluated per User.

To calculate % users under a certain threshold, you can do something like this:

Users under threshold % =
VAR UsersUnderThreshold =
    COUNTROWS ( FILTER ( VALUES ( Users[User ID] ), [Measure 4] < Threshold ) )
VAR TotalUsers =
    DISTINCTCOUNT ( Users[User ID] )
RETURN
    DIVIDE ( UsersUnderThreshold, TotalUsers )

Does that work for you?

Re: Average difference between two MIN dates (Summarizing Measures with Averages)

SebL — Wed, 16 May 2018 17:38:53 GMT

That worked perfectly.

Thanks for the help!