Freelance Web Consultant at architxt.net
27 November 2008 13:11pm
I've just had e-commerce tracking implemented and are exploring metrics under the E-commerce tab.
One thing I can't seem to find are site stats about each transaction. It would be great to know the origin of each transaction: whether it has come from an adword, an organic search, a referring site, etc... as well as the path throught the site before and after purchase.
Is this kind of analysis possible in GA?
Analyst at CxFocus
02 December 2008 18:27pm
That's a more complex question than you might think. It's sort of a "no and yes" situation.
Google have a very firm position on privacy and you will find that your Google Analytics terms and conditions specifically forbid tracking personally identifiable information.
At transaction level, you are very close to such information. So I wonder if this is why the e-commerce 'transactions' reports seem much less well-endowed than most other reports. You can see what products were associated with each sale if you drill down and that's about it. But you know that anyway, because you got the order!
However you can get plenty of insight into the subject by coming at it from different starting points. For example, you can start with the products sold as the basis of the research and then look at where the sales of that product came from. Since we're talking about products, not people, Google Analytics will deliver loads of actionable information. Start at the e-commerce 'product overview' report and then drill down into a product. The default dimension is then the product SKU, which is normally pretty dull -- but you can use the dimension drop down menu to change that to show source, keyword, landing page etc. That lot should provide tons of useful information.
The 'drill down into the product report' approach is a bit tedious, so you might want to consider using a filter to copy the keyword data to the end of the product name. You will then end up with several entries for each products, one for each key phrase. The filter for doing that was devised by ROIRevolution and can be found at http://cli.gs/roirevolutionaddkwtoproduct along with loads of other useful stuff like this.
In the same way you could start with any of the traffic source reports, such as the keywords report and then drill down into keywords (or source, or medium, or...) and then use the dimension report to look at the transaction numbers.
Every site is different, so it's hard to generalise about which reports are going to yield the most significant discoveries. I would recommend exploring some of the ideas above as a start. If you can't learn anything which helps you make improvements from those approaches then you will still probably gain a more specific idea about which missing piece of the puzzle you need to drive the change.
Once you know exactly what you need, somebody may be able to point you in the right direction. Loads more is possible in GA once you start digging in to it, and it's getting better all the time.
Free market research on digital marketing
Daily Pulse: award winning newsletter
It takes 30 seconds to register