2 Replies Latest reply on Jan 23, 2012 5:58 PM by Andrew Fisher

    Blending data sources using date

    Andrew Fisher

      Hi all,

       

      I've been pulling my hair out on this one for some time and can't work out why it doesn't work. At the moment I have data in two sheets in Excel (but will move to mySQL once I have this worked out properly).

       

      Sheet one contains a set of orders that come in over a month and are one order per row. I have changed the dates incoming to excel so that they are ONLY dates (no time information).

       

      Obviously as there's many orders for any given day.

       

      Sheet two contains traffic related data from the website (keeping it simple visits only at this stage) on a per day basis so there is only one row per date. Again this date has been simply specified as the date only.

       

      In tableau both fields have been set to date only not date time.

       

      What I'm trying to do is perform a data blend without using a join. Obviously as soon as I use a join I end up in a situation where I'm duplicating or removing data depending on whether it's left or right.

       

      Ideally what I'd like to be able to do is show the traffic numbers per day (or rolled up per month, quarter etc) alongside the total number of orders per day. From there I can start doing things like periodic conversion rates etc.

       

      Every time I attempt to do this even on highly simplistic data Tableau tells me that I can't blend the data sources because there's no linking fields used in the view. This is even after I've created a relationship using date between the two data sources and I'm using date as one of the measures I'm reporting on.

       

      Is this even possible in Tableau as I can't find any documentation on it? It seems to me that this is a pretty standard use case - using time series data that is recorded at different rates and plotting it against each other at the lowest level of resolution you can attain (in my case daily data).

       

      Cheers

      Andrew