Hi, I’m Rob and I work on the Vice-Provost Teaching and Learning Service Team as a Web Designer/Analyst. I try to make the internet a better place. Specifically, I conduct research into our users' needs (usually students and prospective students), gather feedback received online, and work with editors across campus to improve the digital experience of our audiences.
In order to design a better homepage for our university, it’s helpful to analyze the usage of our current homepage. There is a lot of data captured by Google Analytics, but today we’ll focus on a single measure: where do users go to from the homepage? I’ll pull out five thoughts from looking at the data.
PAWS is a big deal
Over a third of users who go to another page on our site after the homepage, go to PAWS. This has additional significance given that the link to PAWS on the homepage is relatively subtle and the acronym would have little meaning to those outside of our community. Clearly, a significant amount of traffic to our homepage is from an internal audience.
|Page||Pageviews||% of total|
|Arts and Science||1318||0.69%|
Traffic inside and outside of Saskatoon is different
The data can be segmented in a million different ways. A useful one though is to look at traffic geographically. The patterns change significantly.
|Page||% of total|
|Page||% of total|
In both instances, PAWS is by far the most common next page on a visitor’s path. After that though, there is a divergence. In Saskatoon we see greater interest in search and navigation tools (search page and the A-Z index). The job posting page is the most often visited page after PAWS and search. Outside of Saskatoon, we see a greater interest in our prospective student sites: explore.usask.ca and grad.usask.ca.
Users have a few top tasks
Some patterns are consistent, and the majority of users are doing one of the following:
- Logging in to PAWS (36%)
- Using the search tool (19%)
- Apply to study here (14%)
- Find a job (5%)
- Read a news story (3%)
- Access the library (2%)
The above tasks represent 80% of the interaction with the homepage.
Elements of our homepage are unevenly used
Below is a heatmap of what users clicked on for a period of two days in May. Note the lack of even distribution. Of interest are the UI elements that allow a person to scroll through news stories. It seems users are often clicking on the arrows to see the next story, but more rarely clicking on the stories themselves.
Search is varied
Below are the top ten search terms used on the homepage in the past month.
|Search Term||Total Unique Searches|
It is harder to see patterns for two reasons:
- it is most often used when a person is looking for something specific
- people tend to be economical with their word choice and favour nouns to verbs
These make it difficult to discern intent, audience, and larger trends. For example, ‘Parking’ is the tenth most popular term. Who searched for this and what did they want to do? It could have been an incoming student who wants a parking permit for next year, a faculty member paying a parking ticket, or a community member attending the graduation powwow and wanting to know where to park.
That being said, we do see a few terms that have links on the homepage, but are tucked away in menus. For example, only 29 users (of nearly 200,000) clicked on the Parking menu item beneath the ‘Visit’ tab on the homepage, but 102 used the search box to type ‘parking’. I’ll offer three explanations for this occurrence, though of course there are more possibilities:
- The label ‘visit’ is problematic and doesn’t represent the intent of the users who preferred search.
- The users who searched have previously found success with the tool and find it a more efficient way of getting parking information than either exploring or interacting with the menus.
- The number of menu items and sub-items makes free typing a faster and more reliable way of getting to this information.
We also see a large number of internal terms such as concur, dynamic schedule, vpn, webex, uview, and so on. Again, this illustrates the significant internal traffic of students, staff, and faculty members starting from the homepage when performing day to day work at the university.
All data was from the usask google analytics account from June 4th, 2016 to July 2nd, 2016, accessed on July 3rd, 2016. Next page paths were derived by combining data from: ‘next page path’ metric from ‘Page’ dimension, where ‘Page’ was usask.ca/index.php; and ‘Total events’ metric from ‘Event Label’ dimension where ‘Page’ was usask.ca/index.php. The latter metric was used to capture traffic to presences that do not use the central google analytics account.