Please start any new threads on our new
site at https://forums.sqlteam.com. We've got lots of great SQL Server
experts to answer whatever question you can come up with.
Author |
Topic |
rmason
Starting Member
12 Posts |
Posted - 2006-02-09 : 04:38:11
|
Hi,I want to incorporate a new [Advertising] dimension into our data warehouse within an e-commerce domain but am unsure how to then assign member keys to the various facts. For example if we run a TV ad campaign for a month in several regions and I want to measure affect on user registrations, all the documentation and white papers say to add the advertising dimension key to the registration fact... but how do I decide what facts??Any help or direction on this would be greatly appreciated.Rob. |
|
mwjdavidson
Aged Yak Warrior
735 Posts |
Posted - 2006-02-09 : 06:36:39
|
Pay a consultancy to do it for you (I know a good one) . Alright mate - how's it going?Mark |
|
|
rmason
Starting Member
12 Posts |
Posted - 2006-02-09 : 07:12:27
|
Not bad.Hey if you have a solution for my question please then do tell… I can see the easy way would be to just assign the advertisement id to any facts that occur during the ad campaign in the correct geographic area but that seems pretty simplistic to me, i.e. what if you have multiple campaigns occurring in the same area at the same time over different medium, e.g. TV and Radio… how do you decide what key to assign?From what I have read, you should treat this seperately from your standard Promotions dimention as they are attributed entirely differently, e.g. Advertisement Dimension properties would include things such as Advertising Channel, Advertising Location Name, Advertising Location Owner. Whereas Promotions are more along the lines of Promotion Type (Discount, 2 for 1's, etc)... |
|
|
mwjdavidson
Aged Yak Warrior
735 Posts |
Posted - 2006-02-09 : 11:46:16
|
Yeah, I only meant to draw the comparison in as much as you have to assign the key at POS (or point of registration in your case!)Mark |
|
|
rmason
Starting Member
12 Posts |
Posted - 2006-02-09 : 11:55:27
|
Thanks for the comment.I agree and of course this is one of my recommendations for improving the site moving forwards but unfortunately it doesn't help with the analysis of historical data.Hmmmm |
|
|
|
|
|