A couple of weeks ago, I submitted my annual Web Presence Usage Report for the past fiscal year. I can't provide a link to it (it's a Web-original document), because access to our staff site is IP protected. Yes, we keep big secrets here at the University at Albany! But I'd like to share a handful of interesting findings. In a previous blog post, I've mentioned a couple of things. I'll repeat them here and add a bunch more. We've got no lack of data.

• On-campus use rose, mainly because we've got a new Information Commons page that is a browser default on our public workstations. Off-campus use dropped off, because late in the year I removed my Internet Tutorials site and hosted it elsewhere. This site had accounted for a third of the total use of our site. Next year, we'll have a better idea of the ramifications of this move.
• Less than half of multi-page visits to our site - 42% - started with the main screen
• Roughly half the visits to our site (minus visits to our main screen and Information Commons page) were single page visits. This means that half of our users get to specific pages within our site by using bookmarks, links in e-mail messages, and links from other sites.
• Approximately 2 million of our total of 3.7 million visits were "no referrer" visits, meaning that our URLs were typed into a browser or clicked as direct links from e-mail messages or browser bookmarks. This figure is so high that I almost wonder at its accuracy. If it's in any way accurate, this suggests that many users have taken purposeful action to access previously-identified resources on the site.
• 770,000 of our 3.7 million visits originated at search engines.
• Our bibliographer-created Subject Pages and Guides had generally low use. There were a few exceptions, most likely because individual bibliographers have highlighted their pages in instruction sessions or otherwise marketed them. But given that we're in a research university setting, the numbers are low. Our extensive online Reference Collection, which presumably might be popular because of its general interest topics, fared even worse.
• Use of some of the subsections of our site has dropped.
• Our Databases & Indexes application was, as usual, the most popular feature on our site. Users connected to databases from this application about 530,000 times. Considering that the total number of visits to our catalog - which also links to these databases - was 375,000, this is a very telling finding. You need to use the catalog to get to our e-journals. But when it comes to databases, users have made their choice. More user-friendly interface, anyone?
• Speaking of our catalog, it continues to be used far more on campus than off campus. We had a bad robot spam problem this year, so it was difficult to figure out the exact numbers. Last year, 93% of catalog use came from on campus. Our Web site had about three times the use from off campus as on campus.
• Robot visits were overwhelming this year, generating nearly 4 million hits on our site. This interfered with my ability to estimate off-campus use. For example, our database of university photographs was hit over 700,000 times by robots. How many actual visits this translates into, I couldn't say, because my Web analytics tool (Urchin) doesn't calculate this. My robots.txt file can do only so much good.
• Usage is rising for electronic reserves (250,000 visits) and off-campus access to licensed resources (630,000 visits through EZproxy).
  

So what does all this tell me? I've already weighed in on the implications for our traditional notions that most visitors begin with our main screen and navigate into the site from there. These results show clearly that there are many more avenues for finding and accessing our pages.

What these stats are mostly telling me is that some parts of our site are getting stale. We're presenting our material in pretty much the same way as we always have. Our recent main screen redesign (in which I was not involved) was mostly a rehash of what we already had. Our pages don't change much, so there isn't a lot of motivation to keep visiting them unless there's something you really need to know. Might it be the case that certain pages on a library's site will be intrinsically low-use pages? Sure. Pages about library hours, circulation policies, and the like, won't attract visitors in droves. Yet they're essential to have on a library's site. I'm not concerned about this.

What I am concerned about are pages related to topical research. We're a research library, and these pages should be highlights. But much hard work - and internal importance - is given over to pages that are not, in many cases, seeing significant use.

I think we can have a lot of pride in our high-use single-use pages. We must be doing something right here. We should take a close look at these pages and figure out why they work.

Things are happening with our digital collections and tutorials. The Databases & Indexes application, off-campus access to our licensed collections, and electronic reserves, all are seeing robust and growing use. The site of our Interactive Media Center, which contains all kinds of useful tutorials, is also doing well, especially with off-campus users. Our Instruction & Tutorials site, containing many course-assigned tutorials, is also seeing increased use.

And so: licensed resources (especially from off campus), e-reserves, and tutorials are doing well.

Static lists of topical links are not doing nearly as well. I've made some suggestions about enhancements to try, and I'm intending to inspire someone to experiment. (I can just hear some of you: "Inspire, eh? How, exactly, does she intend to do that? Pity the poor bibliographer!") If these experiments work out, maybe they'll interest other staff.

Here's another thought: maybe these pages aren't needed any more. Maybe many students and faculty are finding resources on their own. We should do some assessment.

One of our branch libraries has just launched a blog for its site. Since its usage stats last year were virtually unchanged from the year before, this is an interesting and heartening move.

Usage stats: fascinating and important.