Boss: “Our ecommerce conversions in Google Analytics are less than the actual pace of orders.”

Me: “Nothing has changed in the tracking code or setup. It must be our goals setup which you have to have a Ph.D. to understand, plus whatever mood Google’s algorithms are in today.”

He’s not mad at me. We’re both just confused why Google AdWords, Analytics, and Tag Manager have to be so damn hard to get right. I’ve never been able to do it right. And most data is thrown out because people browse websites while logged into their Google Account, which makes their clickstream disappear and become unattributable because of understandable privacy policies. I don’t want my data tracked when I’m logged in either!

So now we have had to hire specialists at several thousand dollars per month to figure this out.

Hi everyone! I'm in need of some help regarding the approach to my bachelor thesis.

The practical stuff is basically clickstream/task and usability analysis on an existing platform and creating mockup improvements for some processes. I was thinking about using a spider to generate a tree (or another datastructure) regarding all the different tasks available and then trying to optimise said tree, thus automatically optimising the processes within. I'm having however issues imagining how this optimisation might be generalised for more than this one platform.

Basically, I'm a bit lost and grasping for any pointers in any direction regarding these ideas.