Count distinct ordernumbers where datediff < 0

[Jodallen123]

Hello everyone,

I have recently started working with a table consisting of date logs, which I find very tricky. I made a similar thread yesterday and got really good help so I had hoped to get some help today aswell. 

My data looks something like this. I am trying to count the distinct ordernumbers where the difference between min "Status created date" for status A and min "Status created date" för status B is negative. 

This is my current measure: 

PL # Revision distinct 5 =

AVERAGEX(

    VALUES(logs'[Ordernumber] ),

    Var kundfbso = CALCULATE (

        MIN ( logs'[Created_date] ),

        FILTER (

            'logs',

            'logs'[status_changed_to] = "A"

                ||'logs'[status_changed_to] = "B"

        ))

    Var kundfb =  CALCULATE (

        MIN ( logs'[Created_date] ),

        'logs'[status_changed_to] = "C"

    )

    Var diff =  DATEDIFF( kundfbso, kundfb, DAY )

    return

    diff

)

When I drag this measure to a table with ordernumber, it gives the correct number of days per order, but I don't know how to count the distinct ordernumbers where the datediff < 0. 

If anyone has any ideas how I can move forward from here I would really appreciate it.

Thanks in advance! 

[Jodallen123]

Creating a second measure that looks like this works: 

# PL Revision correct =

COUNTX (

    FILTER (

        VALUES ( logs'[Ordernumber] ),

        [PL # Revision distinct 5] < 0

    ),

     logs'[Ordernumber]

)

How can I incorporate this into the first measure and get the same output?

Thanks!

[Jodallen123]

I still haven't figured this out so if anybody has any ideas I'm all ears.