Web Visitor Fact Table
2 posters
Page 1 of 1
Web Visitor Fact Table
I am designing a star schema for bank that wants to capture information for web channel analysis. So far I have a web visitor star at web page level. I can have web visitors that can sign up online for, let's say, a checking account and are now known i.e. go from a web visitor to now a known customer. Currently i have a web visitor dimension, my question is should I also have a customer dimension (seperate than web visitor) or one dimension that handles both? Or should I have a altogether seperate fact table: One for visitors who haven't signed up and second for visitors who have signed up. The complexity there is how I link the two fact tables?
The grain currently is one record per web page visit. If a customer is browsing the internet and goes to a bank website and decides to sign up for a checking account, how does account dimension fit into this? Should I have a default record for account dimension until the visitor signed up for a checking account?
The grain currently is one record per web page visit. If a customer is browsing the internet and goes to a bank website and decides to sign up for a checking account, how does account dimension fit into this? Should I have a default record for account dimension until the visitor signed up for a checking account?
dw_user- Posts : 8
Join date : 2011-05-11
Re: Web Visitor Fact Table
Assuming this web site has some form of log-in so the customer can authenticate themselves, it would seem to follow that any fact table collecting clickstream data from the site would already have a customer and account dimension associated with it. Assuming you also already have a visitor dimension (i.e. the cookie), then record what information you know for each click. Have an 'unknown customer' and 'unknown account' row in their respective dimension and use them when you don't know. Also consider building a session aggregate of the clickstream data.
Similar topics
» 'Routing' the grain of the fact table to multpile members of multiple dimensions causes the fact table to 'explode'
» Integrating new fact table which has one to many relationship with the main fact table in existing star schema
» How to create fact table with measures derived from comparing two fact table rows
» Healthcare Hospital Stay fact table - transaction fact table - a row for In and a row for Out?
» Design a snapshot Fact table in data mart from compliance-enable fact table in enterprise data warehouse
» Integrating new fact table which has one to many relationship with the main fact table in existing star schema
» How to create fact table with measures derived from comparing two fact table rows
» Healthcare Hospital Stay fact table - transaction fact table - a row for In and a row for Out?
» Design a snapshot Fact table in data mart from compliance-enable fact table in enterprise data warehouse
Page 1 of 1
Permissions in this forum:
You cannot reply to topics in this forum