Forum Discussion

manish2k's avatar
manish2k
Regular Visitor
9 years ago
Solved

Calculating percentage based on multiple top N parameters

HI ,

 

I have a table with supplier name , their category and their spend . I need to create a visual table which will show me the fragmentation of supplier spend across these categories . The output I am looking for in the visual is something like this :

 

CategoryTop 3 Suppliers contribution to overall category spend

Top 5 Suppliers

contribution to overall category spend

Top 10 Suppliers

contribution to overall category spend

A75%85%100%
B55%65%70%
C20%25%30%


  

Any ideas on how to go about this ?

 

  • Hi manish2k

     

    Try creating a calculated table that summarises the data using this code

     

    Summary Table = SUMMARIZE(
                        'Table1',
                        Table1[Supplier Name],
                        'Table1'[Category] ,
                        "Total Spend",SUM('Table1'[Spend]) 
                    ) 

    Then add the following calculated column to this newly created table

     

    Supplier Category Rank = 
    	CALCULATE(
    		COUNTROWS('Summary Table'),
    		FILTER('Summary Table',
    			'Summary Table'[Category] = EARLIER('Summary Table'[Category]) 
    			&& 'Summary Table'[Total Spend] >= EARLIER('Summary Table'[Total Spend]
    				)
    				)
    				)+0

    Finally create the following measure

     

    Top 3 Suppliers contribution to overall cat spend = 

    VAR TopNValue = 3
    VAR Top3Spend = CALCULATE(
    SUM('Summary Table'[Total Spend]),
    FILTER(
    ALLEXCEPT('Summary Table','Summary Table'[Category]),
    'Summary Table'[Supplier Category Rank]<=TopNValue)
    )
    VAR TotalSpend = CALCULATE(SUM('Summary Table'[Total Spend]),ALLEXCEPT('Summary Table','Summary Table'[Category]))
    RETURN DIVIDE(Top3Spend,TotalSpend)

    and format the measure to be a percentage.  I think this might be close :)

     

    If it's close, just clone the final measure 2 times for the other TopN values

     

3 Replies

  • Phil_Seamark's avatar
    Phil_Seamark
    Icon for Microsoft Employee rankMicrosoft Employee

    Hi manish2k

     

    Try creating a calculated table that summarises the data using this code

     

    Summary Table = SUMMARIZE(
                        'Table1',
                        Table1[Supplier Name],
                        'Table1'[Category] ,
                        "Total Spend",SUM('Table1'[Spend]) 
                    ) 

    Then add the following calculated column to this newly created table

     

    Supplier Category Rank = 
    	CALCULATE(
    		COUNTROWS('Summary Table'),
    		FILTER('Summary Table',
    			'Summary Table'[Category] = EARLIER('Summary Table'[Category]) 
    			&& 'Summary Table'[Total Spend] >= EARLIER('Summary Table'[Total Spend]
    				)
    				)
    				)+0

    Finally create the following measure

     

    Top 3 Suppliers contribution to overall cat spend = 

    VAR TopNValue = 3
    VAR Top3Spend = CALCULATE(
    SUM('Summary Table'[Total Spend]),
    FILTER(
    ALLEXCEPT('Summary Table','Summary Table'[Category]),
    'Summary Table'[Supplier Category Rank]<=TopNValue)
    )
    VAR TotalSpend = CALCULATE(SUM('Summary Table'[Total Spend]),ALLEXCEPT('Summary Table','Summary Table'[Category]))
    RETURN DIVIDE(Top3Spend,TotalSpend)

    and format the measure to be a percentage.  I think this might be close :)

     

    If it's close, just clone the final measure 2 times for the other TopN values

     

    • manish2k's avatar
      manish2k
      Regular Visitor

      Awesome ! This worked ! Thanks a million !

    • manish2k's avatar
      manish2k
      Regular Visitor

      Phil_Seamark I have one change to ask , what if the main summary table gets one more filer . Let say a department . How will i be able to incorporate it into the below code?