Forum Discussion

sqlguru448's avatar
sqlguru448
Helper III
5 years ago
Solved

Ranking measure with Table variable

Hello Folks,    I'm trying to create measure for Top 10 customers based on sales amount in Power BI which is connected Live to the Tabular model.   Customer is Dimension and Sales amount is in fa...
Need Help
  • MFelix's avatar
    MFelix
    5 years ago

    Hi sqlguru448 ,

     

    Be aware that making a summarizaition of a full table will bring for sure performance issues, in your case you are using the FACT table that I assume have a huge number of data so be aware of that.

     

    Try the following code:

     

    Measure = 
var temp_table =
        FILTER (
            SUMMARIZE (
                ALLSELECTED ( Customer[Name] );
                Customer[Name];
                "Total_Sales"; [Sales Amount]
            );
            [Sales Amount] > 0
        )

var Ranking = RANKX(temp_table; [Sales Amount])


RETURN
    Ranking

     

     

    Then filter out all rankings above 10.

sqlguru448's avatar
sqlguru448
Helper III
5 years ago

Hi vivran22 

 

Below is the sample data screenshot. The ranking should be from 1 to 10 based on highest sales amount. but my above dax gives me Rank 1 in all rows.

Ashish_Mathur's avatar
Ashish_Mathur
Super User
5 years ago

Hi,

Try this measure

=rankx(all(data[Customer]),[Sales])

I have assumed that sales is an explicit measure.

Hope this helps.

  • sqlguru448's avatar
    sqlguru448
    Helper III
    5 years ago

    Ashish_Mathur 

    This has performance issue when other dimesions are added to the visual, that is why I am trying to get top 10 first within variable table and then add other dimensions to it. Thank you.

    • sqlguru448's avatar
      sqlguru448
      Helper III
      5 years ago
      Any thoughts from the DAX Experts please?
    • v-lili6-msft's avatar
      v-lili6-msft
      Community Support
      5 years ago

      HI  sqlguru448 

      Try to use this formula to create a measure

      Measure 2 = 
VAR  _RANK= RANKX (
        ALLSELECTED ( Customer[Name] ),
        CALCULATE ( SUM ( 'Fact'[Sales Amount] ) ))
RETURN
IF ( _RANK <= 10,  _RANK)

       

      if not your case, please share some simple sample data and your expected output.

       

      Regards,

      Lin

      • sqlguru448's avatar
        sqlguru448
        Helper III
        5 years ago

        Hi

        v-lili6-msft   The above DAX gives performance issues as posted in my earlier comments as my real data is in millions, I am adding other dimension columns to the table which is causing the issue. Below is the expected data result for rank, apart from this I have to add other columns from different dimensions.

         

          

        I am trying to fix it by writing below DAX, this works fine in SSAS dax query but not in Power BI desktop.

        The power bi desktop gives rank as all 1's

        Measure =
        var _top10 =
        TOPN(10, FILTER(SUMMARIZE(all('Fact'), Customer[Name],"Total_Sales", SUM('Fact'[Sales Amount])), [Sales Amount] >0 ), [Sales Amount])
         
        RETURN
        COUNTROWS(FILTER(_top10, [Total_Sales] >= EARLIER([Total_Sales]))
        )
         
        Thanks