“since i was trying to get the average per month I decided to go for the following formula: =Sum(FactInternetSales[OrderQuantity]) / Max(DimDate[MonthNumberOfYear])”

A formula such as =CALCULATE(SUM(FactInternetSales[OrderQuantity])/COUNTROWS(DISTINCT(DimDate[EnglishMonthName])),ALL(DimDate[EnglishMonthName]))

should provide the correct values for month average. Oddly, when I tried the formula with the Adventure WorksDW2008 database, it computed a value for August 2004, which has no data. So for this year only, the month average was computed over 8 months instead of 7 months.

“One major issue i find here is that the columndimensions are grouped by measure, not making it easy to compare measures. The OLAP pivottable has this by default:”

That’s odd. When I created the calculated measures, they appeared next to the standard aggregate for the year (like you show in the OLAP figure). What’s not clear is why the year field appears at the bottom of the column hierarchy in your case. I didn’t have this problem.

“Next up was creating a way to select the current month and last month of last year. I know that in CTP3 time intelligence functions will be available so i will be using them when CTP3 is released.”

I’m looking forward to those time intelligence functions. One possible way to do the selections you want is to create slicers for month and year. Since either or both of these fields could end up in the pivot table, to use the fields also as slicers requires duplicating the fields in the Gemini DimDate table i.e. creating calculated columns for month and year.

“I hope a lot of work is spent on user friendliness since Excel is the GUI the end user must be able to create the comparisons.”

I couldn’t agree more. While the issue of time intelligence might be solved, there are other issues that would challenge even seasoned Excel users.