1 Reply Latest reply on Feb 5, 2019 3:53 PM by Esther Aller

    Adding filter leads to "Not Available"

    Josh Ciszek

      My unknown friends...


      I have a map worksheet that shows crime rates and the count of customers by zip code.  In order to get the color palette for customer count to change based on the selected program, I had to re-create my original (OLD) so that the client data was the primary data source (since swapping primary and secondary on the fly does not seem to be possible).


      Everything was going fine until i tried to re-add a filter from the secondary (crime) source (Date Occurred) to the NEW worksheet.  When I did this everything faded out and was no longer accessible.


      The OLD worksheet does not have this problem, but the primary and secondary data sources are wrong.  Why is this happening in the NEW worksheet?

        • 1. Re: Adding filter leads to "Not Available"
          Esther Aller

          Hi Josh,


          Data blending has many limitations, including the behavior you're seeing in your new worksheet. AKA the "Cannot blend the secondary data source because one or more fields use an unsupported aggregation" error that you see when you mouse over AGG(Crime Count) on the last Marks card.


          Because data blending is an ad-hoc LEFT join that aggregates data from the secondary data source, Tableau cannot always perform aggregations that require the full domain. For example, COUNTD(). See Troubleshoot Data Blending - Tableau for more information on this and other errors that may occur while using data blending.


          You can still use the field with COUNTD() from the secondary data source if you remove all dimensions from the secondary data source (i.e. remove the [Date Occurred] filter).


          Better yet, you can use SUM(Number of Records) instead of COUNTD([DR Number]) because that data source appears to only have one record per [Dr Number].


          Typically I recommend joining data rather than blending it. If your data sources are from different databases, you can use a cross-database join. See Join Your Data: Combine tables from different databases for instructions.


          Hope this answers your question!

          1 of 1 people found this helpful