Model 2 Post - Let's Get Analytical
Hi Everyone,
This has been hands down my favorite module of the course so far. I'm coming out of this week energized and excited to see what's next.
In this module we focused on Web Analytics, Google Analytics and Google Adwords. We looked at some of the important metics to use in Web Analytics in terms of both KPIs and Task Completion Rates. We then took a deeper look at Google Analytics and some of the main functionality it provides for users.
I think that I was better prepared for this assignment as it closer aligns with both my previous experience and my interests. I previously worked in Ecommerce, so going through this Assignment for the Google Merchandise Store was a fun throw-back experience for me. I never used Google Analytics before and it is extremely intuitive. It was very easy to follow and provided clear insights.
I had fun digging into the different areas and learning about customer segments, product sales, task completion rates and conversion. I think that it's nice to be able to easily read the data and understand what further questions you have. From there you can create a strategy of where to focus.
The only thing that wasn't the most intuitive, at least in my case, was the Behavior Flow looking at first and second interactions. I did not find this beneficial for tracking how customers move through the site. I would have enjoyed a feature like Hotjar.
I'm not sure if anyone is familiar with it or has used it before but its essentially heat maps to see where users click and linger on a site. I think it's a really creative and intuitive way to not only track users but see if there are any opportunities for improvement in the site.
It's really interesting to me the concept of being able to track a user starting a search in one place and picking it up somewhere else. In a different location, on a different device. Probably some version of this already exists that I'm not aware of. I don't use Sync on my Google account because I don't really wanted to be tracked in this way but it's definitely something interesting to think about.
Let me know if you have any insights!
References:
Dossetto, F. (2019). Map clicks: Using them to optimize your website. Available at: https://www.hotjar.com/blog/click-maps/. Accessed on October 3, 2020.
D2L Materials. Lecture Notes. (2020).

Hi Julia, Great that you have some prior experience in E-Commerce! I found it to be very interesting as well, and I'm also looking forward to what is next!
ReplyDeleteI believe you are referring to the different methods that customers can get to a website? I found assessing the behavior segment interesting, primarily the Click Path.
Hi Julia, great to hear that other people are enjoying this course as much as I am, I am finding that the modules in the course are some of the most applicable to what I want to do career wise. I have never used hotjar before, but I am going to look into it, it looks super cool.
ReplyDeleteHi Julia! I also missed more functionality for behavioral analysis in the free GA version. Although I found the flow diagram useful to generally describe how visitors access and exit the site, page-oriented analysis and some predictive modeling on visitor behavior could help the analyst understand how visitors navigate the site. That approach would be useful to find attractors of behavior and detect locations/events in a website that increase the likelihood of visitors exiting or failing to complete a conversion. In addition to Hotjar's click heatmaps, I think it would be helpful to incorporate association rules (e.g. Apriori) and sequence mining (e.g. GSP algorithm) for the kind of behavioral modeling I refer to. Navigating a website has conceptual similarities to the type of questions made in itemset mining. Now I am wondering whether/how a Markov chain "with memory" (actually, just MC with varying length or high order) could be used for this question. Bing Liu's Web Data Mining book has a nice intro to these questions, more from a theoretical than a use case standpoint though...
ReplyDeleteHi Julia. I believe you're quite ahead of the pack in terms of titling your blog posts! I don't have any previous experience in ecommerce so this whole module was quite new to me. I agree with you though, Google Analytics seemed fairly intuitive and easy pick-up. Not to say I became an expert throughout the module but I at least felt comfortable navigating across GA and interacting with different reports. I like what you had to say on cross-traffic. I find myself always looking up stuff on my phone only to finish what I started on my laptop. I didn't think about that when I was going through this module but it would be quite interesting to see some statistics/reporting on cross traffic. I imagine task/goal completion might look a little better if cross-traffic was accurately captured!
ReplyDeleteHi Julia!
ReplyDeleteI wanted to drop in to comment on how I also disable Google Tracking and other apps because I am not a fan of others tracking my behavior. Ever since taking the cybersecurity related courses in the MIS program, I am more conscious of my online activity and the GA assignment made me even more conscious of what is going on.
I was really appreciative that web analytics does its best to anonymize their date by focusing on actions, but being able to segment data on gender and age revealed there are other data pits connected to users that GA is collecting or already collected. It leads me to question whether having all that information is worth it at the risk of revealing identities of users.
Best,
Dustin
Hi Julia,
ReplyDeleteI really like your idea of cross-device traffic. I think we are similar in the sense that we use multiple devices to solve a problem. I use chrome on my desktop and firefox on my phone and they are not connected at all. I don't do this out of any particular reason, it's just a rut I fell into. I wonder if I used chrome on both my Laptop and on my phone if google would be able to produce this metric about me. Certainly interesting to think about like you said.
Thanks again!