CODEWITHSUNDEEP
sql-serverolapbusiness-intelligencecubes
moleboy
moleboy

Reputation: 874

SQL Server BI: SIngle cube, multiple fact tables

I'm new to creating cubes, so please be patient.

Here's an example of my data I have multiple companies, each company has multiple stores. Sales are each linked to a particular company, with a particular store on a particular date. ex:5 sales took place for Company A, Store 1, on 5/19/2011 Returns are linked to a particular company on a particular date. ex: 3 returns took place for Company A on 3/11/2012 Sometimes my users want to see a list of stores, the date, and how many returns took place, and how many sales. Sometimes they want to see a list of companies, the specific stores, and the number of sales.

I have a table that stores COMPANY - DATE - STORE- SALES - RETURNS I end up having the value for returns repeated for each store under a particular COMPANY - DATE pair. so if I'm writing a query, and I want to find out returns, I just do a select distinct company, date, returns from mytable

but I am not sure how to this into a cube (using SS BI and Visual Studio). (I've only made a couple of cubes so far)

Thanks! (also, feel free to point me at appropriate references)

Upvotes: 0

Views: 998

Answers (1)

brian
brian

Reputation: 3695

It sounds like Company is an attribute of the Store and should be in the Store dimension rather than the fact table. There may have to be a transformation on returns to convert the Company to a store.

Am I missing anything?

Upvotes: 1

Related Questions