Forum Discussion

ilooper's avatar
ilooper
Frequent Visitor
4 years ago
Solved

Creating a measure - divide filtered sum by count on different table

Hello,

 

This is probably an easy one.  I have site level information on the revenue side.  I do not on the cost side.  What I'm trying to do is write a measure that helps me allocate cost to each site (facility) by utilization.  In this case it would be how many patients we performed in that given month.

 

How I do it in my excel model that works really well is create a ratio of patients done in a month for that market by site. example 

 

Houston market performed a total of 100 patients in month of Feb 22

site 1 did 50 patients = .5

site 2 did 25 patients = .25

site 3 did 25 patients = .25

site 4 did 0 patients = 0.0

 

Then I mutiply the total cost for the market by the ratio.  If total cost is 10,000

site 1 = 5,000

site 2 = 2,500

site 3 = 2,500

site 4 = 0

 

I'm not sure how to create this ratio or how I would mutiply it by the cost for the market.  So I can see what is my cost by site.  Later I will want to figure out my margin by subtracting the site level revenue by the cost by revenue.  Any points in the right direction would be helpfull.

 

  • Anonymous's avatar
    Anonymous
    4 years ago

    Hi ilooper ,

     

    Is there a relationship between Cost and Patients tables?

    My Sample:

    Patients:

    Cost:

    In my sample there is no active relationship between two tables.

    Rate =
    VAR _Total =
        CALCULATE (
            SUM ( Patients[Amount] ),
            ALLEXCEPT ( Patients, Patients[Year/Month] )
        )
    VAR _Amount =
        SUM ( Patients[Amount] )
    VAR _RATE =
        DIVIDE ( _Amount, _Total )
    RETURN
        _RATE
    Cost =
    VAR _Cost_CurrentMonth =
        CALCULATE (
            SUM ( Cost[Cost] ),
            FILTER ( Cost, Cost[Year/Month] = MAX ( Patients[Year/Month] ) )
        )
    RETURN
        _Cost_CurrentMonth * [Rate]

    Result is as below.

    If there is a relationship between two tables, like [Year/Month] to [Year/Month]. It can make Cost code easier.

    Cost =
    VAR _Cost_CurrentMonth =
        CALCULATE (
            SUM ( Cost[Cost] )
        )
    RETURN
        _Cost_CurrentMonth * [Rate]

     

    Best Regards,
    Rico Zhou

     

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

2 Replies

  • Anonymous's avatar
    Anonymous
    Not applicable

    Hi ilooper ,

     

    Is there a relationship between Cost and Patients tables?

    My Sample:

    Patients:

    Cost:

    In my sample there is no active relationship between two tables.

    Rate =
    VAR _Total =
        CALCULATE (
            SUM ( Patients[Amount] ),
            ALLEXCEPT ( Patients, Patients[Year/Month] )
        )
    VAR _Amount =
        SUM ( Patients[Amount] )
    VAR _RATE =
        DIVIDE ( _Amount, _Total )
    RETURN
        _RATE
    Cost =
    VAR _Cost_CurrentMonth =
        CALCULATE (
            SUM ( Cost[Cost] ),
            FILTER ( Cost, Cost[Year/Month] = MAX ( Patients[Year/Month] ) )
        )
    RETURN
        _Cost_CurrentMonth * [Rate]

    Result is as below.

    If there is a relationship between two tables, like [Year/Month] to [Year/Month]. It can make Cost code easier.

    Cost =
    VAR _Cost_CurrentMonth =
        CALCULATE (
            SUM ( Cost[Cost] )
        )
    RETURN
        _Cost_CurrentMonth * [Rate]

     

    Best Regards,
    Rico Zhou

     

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

  • You can do this the hard way (via crossjoin or M:Mrelationship), or you could cheat and merge these tables in PowerQuery to "augment" the header level table with the detail level weightings. What's your preference?