Tuesday, 4 September 2007

Web usage of library services

I've been doing some work over the last few months, on a very sporadic basis, on benchmarking the usage of online library services across Greater Manchester. We compare a lot of library service provision across the region, which seems to work very well.

So the questions I've been tackling are, in no particular order:

  1. What technologies are used for measuring website use?

  2. What terms do those technologies use?

  3. What methodologies do those technologies use?

  4. What, if any, measurement goes on in libraries rather than via the Council's ICT department?

This all comes from the statement we librarians are likely to make when faced with declining visit and issue statistics: we have more stuff online, so people don't actually need to come to the library as often. What we don't seem to do very well is measure how much use people are making of the online equivalents for our services - renewals, reservations - and the services we buy in - dictionaries, encyclopedias, British Standards, business information...

Just as every Red Hot Chilli Peppers song is basically the same, the many and various web performance management tools are inherently similar. The terms are different, the algorithms behind the methodologies are trade secrets (and likely to stay that way), and we seem to rely on our ICT providers to crunch numbers for us. This makes our position simultaneously easy to deal with - everyone's figures are probably measured in a broadly comparable way - and complex at the same time - there's too much "probably" involved, and not enough "definitely". OPACs, again, are much of a muchness and are increasingly good at providing this sort of data for us: ours can provide stats for online as opposed to in-person renewal, which is the sort of activity we should be measuring.

In an ideal world, I'd stick a Google Analytics tag in every library's website so that the same (free) standard tool could be used. I'd also encourage MLA to negotiate with our online resource providers for a single statistical monitoring package: since MLA negotiates our deals for us, and sets the performance targets that rely on measuring usage, they ought to be in a position to do that sort of thing.

I'm making an initial presentation on Thursday afternoon (after a morning stint on Enquire that I can't really weasel out of). Hopefully, I'll get to write the work up properly after that, and I'd definitely like to expand it a bit... How does your library measure online usage?