Forum Discussion

ytc-reports's avatar
ytc-reports
Icon for Helper I rankHelper I
9 years ago
Solved

How To Filter Based on Sum Column

Hi. I'm new to Power BI, so sorry if this is a stuipd question with an easy answer.

 

I have a table with the following relevant columns: User ID, Quantity, and Warehouse Document Number

 

When I put User ID and Quantity in the Values area of my report, but not Warehouse Document Number, it sums up all the Quantities for all the rows with a matching User ID, as expected.

 

Now I want to filter my report based on that *summed up value*. For example if User ID Sam has a summed up qty of 45 and User ID Josh has a summed up value of 12, I might want a filter that doesn't show any User IDs with quantity below 15, so it takes out Josh and leaves Sam.

 

But when I put Quantity in the Report Level Filters, rather than filtering based on the summed up quantity value for each User ID, it filters out *each individual row* that's below 15.

 

So how do I filter based on the Quantity at the User ID level?

  • To Get the Solution, I have utilized the  GROUP BY  feature of the query editor.

     

    Follow the screenshots for the solution.GROUP BYCONDITIONAL COLUMNFILTER THE COLUMN

  • Sean's avatar
    Sean
    9 years ago

    ytc-reports There are several other way to do this...

     

    1) Once you create your table visualization with User and Quantity

     

    go to Visaul Level Filters and select Show items when the value: is greater than 15

     

    This would work the same way whether you just use the

    Quantity Column or create a

    Quantity Measure = SUM ( Table[Quantity]) or

    Quantiy Measure per User = CALCULATE ( SUM('Table'[Quantity]), ALLEXCEPT('Table', 'Table'[User]) )

     

    2) You can create another Measure based on the last one above

     

    Quantity per User > 15 = CALCULATE ( [Quantiy Measure per User], FILTER ('Table', [Quantiy Measure per User] > 15) )

     

    then when you create a table visualization with just User and this Measure

    you'll get the result you want with no need to use the Visual Level Filters

     

    It's up to you which way fit best in your case! :smileyhappy:

     

5 Replies

  • Just for example, let's say I had the following list of rows:

    UserWarehouse NumberQuantity
    SamABC14
    SamBCD14
    JoshCDE10
    JoshDEF3

     

    When I take out the Warehouse Document Number, it looks like this:

    User IDQuantity
    Sam28
    Josh13

     

    I want to add a filter that filters out the *second table* based on Quantity below 15, so that it would only show the Sam row (and any other User with a sum > 15). But at the moment, when I filter based on Quantity below 15, it would filter out *each individual row* from the top table, leaving me with nothing.

    • BhaveshPatel's avatar
      BhaveshPatel
      Icon for Super User rankSuper User

      To Get the Solution, I have utilized the  GROUP BY  feature of the query editor.

       

      Follow the screenshots for the solution.GROUP BYCONDITIONAL COLUMNFILTER THE COLUMN

    • Sean's avatar
      Sean
      Icon for Community Champion rankCommunity Champion

      ytc-reports There are several other way to do this...

       

      1) Once you create your table visualization with User and Quantity

       

      go to Visaul Level Filters and select Show items when the value: is greater than 15

       

      This would work the same way whether you just use the

      Quantity Column or create a

      Quantity Measure = SUM ( Table[Quantity]) or

      Quantiy Measure per User = CALCULATE ( SUM('Table'[Quantity]), ALLEXCEPT('Table', 'Table'[User]) )

       

      2) You can create another Measure based on the last one above

       

      Quantity per User > 15 = CALCULATE ( [Quantiy Measure per User], FILTER ('Table', [Quantiy Measure per User] > 15) )

       

      then when you create a table visualization with just User and this Measure

      you'll get the result you want with no need to use the Visual Level Filters

       

      It's up to you which way fit best in your case! :smileyhappy:

       

      • ytc-reports's avatar
        ytc-reports
        Icon for Helper I rankHelper I

        It turns out the 'visual level filters' is all I needed. Thank you all for your help.