I just want to check what customers are coming through using the concat rather than a count. Makes it easier for proving the concept.
Here I am trying to have the customers who transacted in the previous month listed against the dimension of month. The problem I'm having is that the max(Month_D) here goes to the 'overall max', so I'm getting Mar 2013 -1 = Feb 2013. So I'm getting 'those customers from February who transacted in the month of the dimension'.
Re: Re: Count where NOT in a dimension but is in the previous dimension
Thanks, unfortunately the Above function won't work with customers when I want to count distinct or to include/exclude based on their behaviour in the past/future. I understand the problem with the dimension, but this was exactly what I was trying to overcome with Set Analysis.
Re: Count where NOT in a dimension but is in the previous dimension
Unfortunately it is impossible to overcome with set analysis. If this is something that is important then you can always work around this by using a for loop in your script (and consolidate the last months customers into a new field), that way you can create a few artificial fields that will not be excluded by your use of the month as the dimension.