2 Replies Latest reply on Aug 28, 2012 8:37 AM by Tracy Rodgers

    Aggregating Distinct Counts in Extract

    John Mogielnicki

      I'm getting into aggregating extracts in order to increase performance of server views, and running into some problems.  I'm hoping someone can help me out.  In my datasource, I have applicant IDs and rebate IDs.  One Applicant ID could have multiple Rebate IDs, and so would be duplicated in the datasource (see attached excel doc for an example).  I often want to count distinct applicant IDs in my calculations, because I want to count unique participants in our program.  My problem is that when I aggregate an extract and roll it up to years, for example, I am unsure of how Tableau is handling these IDs.  In the packaged workbook attached (the data comes from the attached excel doc) I have two views, the first from a non-aggregated extract, the second from an extract that is aggregated to the year level.  As you can see, the first tab is showing 3 unique applicants, which is correct.  The second is showing 2, which is incorrect.  Can anyone tell me why this is happening?  Is there any way to have Tableau aggregate a measure by counting distinct?