In my role as web developer at the Smithsonian Libraries, I recently attended the LITA National Forum in Columbus, OH. At the conference, I participated in an 8-hour pre-conference session on website analytics and how to use them to understand and improve the usability of a website. Since this is Open Access Week, I thought a summary of this session might be interesting to share.
The pre-conference was conducted by Nina McHale (@ninermac) and Tabitha (Tabby) Farney (@sharebrarian) who have given presentations in the past and who are rumored to be working on a LITA publication on website analytics. Experts? I’d say so!
The sessions were broken into two main chunks, the first of which was an introduction to the concepts of web analytics and the second being specific techniques to convert your organizations goals into measurable values via web analytics.
One of the most interesting things that I learned is that the Smithsonian Libraries is similar to, yet different from, many other organizations out there. This is because we are primarily a research library, but we are also open to the public (by appointment only, of course!). Our website is set up to serve these two major groups of users, researchers and the public. This caused some interesting discussions in the session as the question was asked: perhaps you need to build two separate websites?
Well, that’s unlikely at this point, but we can still build our site to tailor itself to these two main groups of users. But what about the behavior of our users? Researchers are likely to use the site in vastly different ways than, for example, a teacher coming to our site to see our online exhibition of the building of the Panama Canal. How do we measure whether our site is successful in getting these two groups of people to the content they want.
In fact, how do we know our visitors are finding what they are looking for?
And for that matter, how do we even know what people are searching for?
The culprit: A few of our images were picked up by StumbleUpon and rapidly became very popular.
After eight hours and lots of discussion, exercises and many questions asked (and some answered) we learned that web analytics is not a silver bullet to solve these problems. However, we did learn that we can take the goals of our library organization and convert them into goals for the website. From there the website goals are “translated” into indicators that we watch for in the analytics reports.
With some analysis and understanding of what the reports mean, we can divine whether our visitors are finding what they are looking for on our site. If not, we can look for ways to make our site better, easier to navigate, easier to read, or simply clearer in how we present information.
The point being, our website is useless if our patrons don’t find it useful. Sounds almost too simple doesn’t it? Therefore as we build out our new site, please keep in mind that it’s always going to be a work in progress but using web analytics, we will continue to improve it over time.
Be First to Comment