Forum Discussion

sshook's avatar
sshook
Frequent Visitor
2 years ago
Solved

Divide rows by specific total row

I've looked through the boards and have seen solutions that came close to what I am trying but dont work exactly. 

I am looking to create a % Share metric based on dividing the sales of each product, by a row title "Total Products". Total Products is not a sum of the products listed below but includes every possible product in that category, some of which we dont pull data for. 

 

Essentially I need a measure that is [Product] / [Total Products] but both are in the same column in the same table. I have been able to use Filter to return the $ sales of "Total Products" but have not been able to replicate that value across the rest of the rows. 

 

 

 

  • sshook Assuming you have a production dimension with all the products, you can write the following measures to get the % and also the sum of all products at the total:

     

    Base Measure = SUM ( Table[Sales] )
    
    % Share = DIVIDE ( [Base Measure], CALCULATE ( [Base Measure], ALL ( ProductTable ) ) )
    
    $ Sales for Visual = 
    IF ( HASONEVALUE ( ProductTable[Product] ), [Base Measure], CALCULATE ( [Base Measure], ALL ( ProductTable ) ) )

     

3 Replies

  • sshook Assuming you have a production dimension with all the products, you can write the following measures to get the % and also the sum of all products at the total:

     

    Base Measure = SUM ( Table[Sales] )
    
    % Share = DIVIDE ( [Base Measure], CALCULATE ( [Base Measure], ALL ( ProductTable ) ) )
    
    $ Sales for Visual = 
    IF ( HASONEVALUE ( ProductTable[Product] ), [Base Measure], CALCULATE ( [Base Measure], ALL ( ProductTable ) ) )

     

  • sshook's avatar
    sshook
    Frequent Visitor

    parry2k The problem is I dont have a table that is that includes all products. We pull data from a syndicated database and the data pulls are too large if we include the entire product universe available. 

  • sshook in PQ, you can always create a product  dimension table, reference your data table, remove all columns except product related columns, and then keep a distinct row of it. It will create a dimension table for you.