I'm working with a SSAS 2000 cube that's built off of a fact table that
has a dimension for Sales Type (Budget, Actual, etc...) and measures
that are defined independently of sales type (e.g. measures such as
'Qty', not 'Actual Qty' or 'Budgeted Qty'). So a user would select
Sales Type = Budget and Measure = Quantity to see Budgeted Qty. This
structure was chosen as we have many different sales types, and wish to
avoid having a huge list of measures, all being basically the same,
just differentiated by type of transaction.

The issue I'm working through now is how to best store/display common
Variance values - Budgeted Qty vs. Actual Qty, for example. I can do
this at the reporting layer, but would like to find a way for users to
be able to manipulate this within the cube itself. Any ideas or
thoughts on options for handling in the cube?

Thanks,

dabluhm wrote:
Realized the subject line is misleading - I'm looking for options that
allow selection of a logical variance value basis 2 members of the same
Dimension (Sales Type in this case) and a measure. Apologies for any
confusion.

How about defining a calculated member on the Sales Type dimension?

For example,

IIF(<Condotion>, Actual - Budget, Budget - Actual)

Ohjoo

"dabluhm" <dabluhm (AT) gmail (DOT) com> wrote