tracking users with google analytics after they leave my domain to make a purchase and come back

Question

I would like to track where users originally came from when they make a purchase on my site so I know which keywords are more profitable and which websites are best for advertising.

an example is a user is on my site with my google analytics tracking code which has details of where they came from, and then decides to upgrade. they leave my domain to go to my biller (2checkout) complete the purchase and return to my thank you page.

I have transaction code and analytics code on my thank you page and the transactions are showing up with the correct product/amounts in GA however there is no other data and in my reports the referring url is always my biller or a credit card companies authorisation page.

i can manually connect which customer is which by saving their referring data when they first come to the site and then matching it up after they make a sale, but I would like it to show up in my google adwords / analytics account where it is easier to manipulate the data and see trends.

if anyone can help me with this annoying issue I would be vbery greatful, but I fear I may end up living off reports I create and then matching them up with adwords manually :/

Answer 1

One thing you can do is have a click event trigger a custom variable. When the user clicks on whatever link that takes them to your biller, have the custom variable trigger with the information you want to carry over (like the current page URL, some campaign name, whatever). Specify the custom variable's scope as Session or Visit so that it get associated with the thank you page.

http://code.google.com/apis/analytics/docs/tracking/gaTrackingCustomVariables.html

An alternative is to do campaign tracking:

http://www.google.com/support/analytics/bin/answer.py?hl=en&answer=55540

That is more or less the same principle as the first suggestion, but with using specified URL parameters. Depending on how your pages are actually coded, you may need to push a virtual page view with the campaign code(s):

http://www.google.com/support/analytics/bin/answer.py?hl=en&answer=55521