1 Reply Latest reply on May 10, 2017 12:23 PM by Elliott Stam

    Plan Vs Actual - Data Blending

    drew guyer

      Hey Everyone!

       

      I am fairly new to Tableau and had a question on the Data Blending.  Currently, I am trying to blend Salesforce Data(Actual - Primary Data Source) with an Access DB(Plan - Secondary Data Source).  Each dataset has the following dimensions that I have set-up a data relationship for: Fiscal Quarter, Region, Product.

       

      When I build an Actual Vs Plan table on the Region by Fiscal Quarter, the data is correct.  When I build an identical table with Product by Fiscal Quarter, the data is also correct.


      However, when I attempt to build an identical table with: Region, Product by Fiscal Quarter.  The plan data does not seem to be pulling through correctly.  I am imagine it is something with the way tableau is aggregating the blended data but not sure.  Can anyone help??

        • 1. Re: Plan Vs Actual - Data Blending
          Elliott Stam

          Hi Drew,

           

          A couple things I would look for:

           

          1) On your data pane, when you select your secondary data source, do you see orange 'linked chains' for each of the dimensions you are blending on? If one of these links is not connected, connect it

          2) Do all of your values match up properly in the underlying data you are blending?

          • for example, Larry in data source 1 will not match to larry in data source 2 because these names are not exact matches.
          • spelling and/or case sensitivity is a big deal for blending, so make sure the values (not just the field names) are matching

          3) Do both sources of data go down to the same level of granularity? If you have product-level information in one, but only region-level information in the other, what you're trying to do won't work

          • Data source A has region-level; Data source B has product-level. Data source A can only contribute information at the region-level because it's missing the more granular product dimension, however Data source B may be able to aggregate its products based on the regions those products are sold in. You can always aggregate UP to higher dimensions, but you cannot drill down to a granularity that doesn't exist for a data source.

          4) Are you using Tableau v10? You could potentially dodge a lot of data blending bullets by doing a cross-database join instead of a blend. In my opinion, joining is almost always preferable to blending and it leaves more doors open to you later on down the road (in terms of formulas and flexibility) that would be otherwise closed if you went with blending.

           

          Hope that helps,

           

           

          Elliott Stam - InterWorks