Data modelling issue

[#1]

HI,

Need help on data modelling.

We have 2 fact tables and some common dimension tables

Product dimension table will contain the product information like Product Family, Product and so on.

Sales Fact will contain the sales at Product leveland Sales Forecast table will contain the Forecast at Product Family Level.

Since two fact tables are in different granulrity, we couldn't concatenate the fact tables. Please suggest how to handle this.

[MayilVahanan]

Hi,

     You shouldn't concatenate with tables? if so, use

     Noconcatenate concept

The NoConcatenate prefix forces two tables with identical field sets to be treated as two separate internal tables.

Ex:

Load A,B from file1.csv;

noconcatenate load A,B from file2.csv;

Edit:

Can you try this and post the screen shot for table viewer?

Hope it helps

[#3]

If we keep the facts seperately then it will create the loop.

[Anonymous]

Hm, instead of having Sales and Forecast as separate fields.. can you perhaps combine them to a field, Amount, and then have a second field called type which is eiter Actuals or Forecast. Then you can combine the data into one table and just attach it to your calendar on month. Just off the top of my head here

[Oleg_Troyansky]

I'd recommend concatenating the tables into a single fact table.

In order to preserve the relation to Product Group, you should move it from Products into Sales and drop it from Products. This way, both fact tables are associated to the Product Group.

cheers,

Oleg