Hi. Sorry that my first post is a question, and a waffly one at that! I've been reading through the articles on cubes/MDX for a little while so I'm OK with the basics but would like to be pointed in the right direction with this one.

The problem in short: How to compare cube data to flat data in a Visual Studio report.

The problem in long: I'm fairly new to cubes and have so far done little more than define a few simple cubes with a number of dimensions and a selection of calculated measures.

This has been fine up to now but a requirement has come up to link in extra data which is pre-aggregated and will need to be reported alongside the cube data (e.g. simple comparisons on cube data vs. pre-aggregated data).

The sort of dimensions used within the cube are Gender, Age and Residence. The sort of measures used are CarOwner (1=true, 0=false) and AnnualSalary. The reports from this data look at things like the percentage of car owners by age.

My problem is that the external data is pre-aggregated and pre-calculated e.g. percentage of car owners at a national level by male, female and total. I don't have any raw data beyond just the final percentages.

Ideally I would like to produce a report that shows something along the lines of:
Residence | Gender | %CarOwnerLocally | %CarOwnerNationally | %Diff
Birmingham | M | 70% | 66% | 4%
Birmingham | F | 68% | 65% | 3%
Manchester | M | 64% | 66% | -2%
Manchester | F | 60% | 65% | -5%

I'm open to any suggestions with this but the more flexible and easier it is to maintain the better. For example, if I take Gender out of the report so that the cube data shows for all people then the relevant national data for all genders would also be displayed. Or if I were to add in Age the report would look for any matching Age aggregated national data and if found display it.

I know this isn't the clearest explanation so if you need any more details please ask and I will try to provide them.