I am a novice at MDD and am in the process of designing a departmental mart
which will be implemented with Sybase AS-IQ. I have a particular design
question for which I am seeking help. (BTW, I am using Kimball's DW Toolkit
book as my design reference).

CASE:

I am working in a credit card company that has N-million customers.

For my basic model, I have the following tables in mind:

- Customer dimension
- Month dimension
- Product dimension
- Monthly Snapshot Fact

The problem I have is that I want to address the following type of query:
"Find the 36 month purchase total (or any other sum-able fact) for customers
who had a credit score of 600 12 months ago"

I am struggling with how to deal with the credit score attribute (which can
also be a fact but that's not relevant) for the customer.

I don't think any of the slowly changing attribute solutions described in
Kimball work nor can I create a "dimensional outrigger" because I am
interested in finding the set of customers first then their transactions (as
opposed to the query "Find all purchase totals of people who had a credit
score of 600 in a given month", which can be addressed by having a "credit
attributes" dimensional-outrigger mapped directly into the fact table).

Therefore, I am left with the (perhaps obvious) solution of having a 1:M
from customer to customer-monthly status with the latter structured with a
Customer FK and a Month FK.

Based on Kimball's book, this last approach seems like something I should
avoid but I don't have any other ideas.

I'd appreciate input from the gurus out there.

Thanks in advance.