New question for the annual Public Library Survey: Virtual Visits -- what does that mean?

Last week I gave a webcast for Infopeople, an organization in California dedicated to the training of library staff throughout the state.  The webcast was entitled Effective "Virtual Visits" Statistics for the Annual Public Library Survey (long, I know, but accurate).  I think the content will be helpful to many libraries.

Many library web staff don't know that a new question was added to the Public Library Survey, the annual survey that each library has to fill out every year noting staffing and budget numbers, circulation, and more.  The new question they added this year is as follows:

Virtual visits to the library (website or catalog):
Virtual visits include a user’s request of the library website or catalog from outside the library building regardless of the number of pages or elements viewed.  This statistic is the equivalent of a session for a library's website. Exclude virtual visits from within the library, from robot or spider crawls, and from page reloads.

From the webinar's page, you can listen to the audio of the webinar, get a copy of my PowerPoint slides to follow along with, and get a handout with some resources on website statistics, tools, and more.  I hope that this will be helpful to people working to meet this new statistical requirement.

The long and short of it is as follows:

• Website traffic: Most libraries are measuring website traffic, and can get visits numbers from there.
• Catalog traffic: ILS vendors, with the exception of open source providers, do not provide any way to measure visits so libraries will have to kludge together some page-tagging to measure hits and (if you're lucky in how you can set it up), maybe even visits.  But this will require work, and perhaps a monetary investment, from each library.
• Extended web presence traffic: Most of us have some presence outside their website and catalog (a MySpace page, a Flickr account, something).  Tracking these will require some page-tagging and more likely than not you'll get hits not visits.  This will also require work, and perhaps a monetary investment, from each library.
• Excluding traffic from within the library is easy for the website with most web analytics software, and harder/impossible for the catalog and the extended web presence.
• Excluding robot or spider crawls can be done with most web analytics software.
• Excluding page reloads is virtually impossible without labor-intensive manual analysis of the server logs.
• Tracking between servers to measure the use of one user for one visit from their hits to the catalog, to the website, to a Flickr page, and back to the catalog is impossible.  Can't be done without creating a whole lot of expensive back-end technology for each and every library. Not gonna happen.

My advice is this: do what you can with what's left of this fiscal year (only a couple of months).  Sampling will be required for most libraries for the catalog and extended web presence numbers.  Document what you do and what you are counting.  Tell the IMLS and the State Library in your state (the folks coordinating this) what you can and can't do, what numbers you're actually counting (e.g. hits vs. visits), and the quantity of time and money resources it's taking to get numbers.  Also tell them what numbers would be meaningful to you.  For me, breaking the catalog out from the website and extended web presence is important and documenting hits vs. visits.  That's my two cents...what's yours?