Library website analytics

I have just been looking at the presentation [ppt] given by Tony Hirst and Hassan Sheikh to Internet Librarian International last October.

They report the results of their analysis of traffic to the home page of the Open University Library using Google Analytics.

Here are some of the things that I noticed:

  • The most clicked links on the home page were: Databases (20%), eJournals (19%), Library Catalogue (11%).
  • Slightly over half the traffic came directly to the website, about 40% from referring sites, and just over 5% from search engines.
  • The most popular pages (aside from the home page) are Journals, Databases, and eResources (Online collections by subject).
  • There is data on referrals to particular resources, by source website (e.g. from a learning management system) and by network location.
  • The presentation concludes with some questions for further investigation, including thinking about slicing data by referring course (from within the learning management system) and by database.

A couple of things jumped out at me. First, I wondered whether the low number of clicks from the home page through to the catalogue as compared to licensed materials would be repeated in other institutions. And second, I was interested in the large number of referrals from other internal OU systems (e.g. the learning management system aka VLE). Again, I wondered whether this pattern was repeated in other university environments.

For related work, see Tony Hirst's blog entries beginning with this one.

Update: Check out the comments for a note by Tony Hirst about ways in which the OU may not be typical. He also talks a little about measuring traffic from the VLE to the library.

Comments: 1

Jan 20, 2009
Tony Hirst

Lorcan

It's maybe worth my pointing out a couple of reasons why OU Library results may differ from other HEI libraries':

- the OU is a distance learning organisation, and as such we have no students on campus using the OPAC. Indeed, the Library isn't a huge one and a lot of the stock is quite dated, so I'm not sure how heavily it's used even by OU academics?
- the OPAC lives on the URL voyager.open.ac.uk, which isn't included in the stats we reported (they focus on http://library.open.ac.uk/*); many people know the URL, or link to it directly, and the catalogue runs on many of the terminals in the library building;
- being a distance learning organisation, the majority (if not all) of our courses have an electronic presence/homepage for students taking the course. The OU Library has been quite active in getting resource links into Moodle VLE pages for courses that have a Moodle presence; typically, these links to library resources drive traffic back to the Library site (NOT the OPAC).

I have another slide deck which started to explore how we could interpret Google Analytics reports segmented by traffic arriving on library.open from both the VLE in general, and from a specific course in particular, so we could start to get a feel for what sort of traffic was coming to the Library from the VLE as a whole, as well as how students from a particular course were using the Library site when they clicked through to it from the online course materials.

With the OU Library tasked with providing an increasing number of elibrary services to courses/students, knowing how Library website usage is being driven by direct traffic is important for future development, I think?

(Some of this usage data, and insights arising from it, may in turn be of value to course teams, e.g. so they can see how they students were using Library resources...)

tony

PS I don't think i can publicly post the course->library slides, but if you mail me, I can let you have a copy...