Reporting on statistics
Writing about web page http://writetoreply.org/actually/2009/10/28/thinking-about-user-tracking-on-writetoreply/
I just asked on the UKCoRR list about Google Analytics, after forwarding a link to Tony Hirst's blog, as recommended by Andy McGregor of the JISC.
The replies got me thinking about how we use the statistics that we get from GA. Some repo managers are writing regular monthly reports for managers, as blogged by the CADAIR team: http://welshrepositorynetwork.blogspot.com/2009/10/statistics.html
I look at the stats at least once a month, in order to write to our "top content" authors. I use that e-mail as a way of promoting WRAP to the authors, especially those who might not be aware of WRAP or that their article is in the repository. (Deposited by co-authors or administrators on their behalf.) It has resulted in raised awareness, some goodwill and conversations about WRAP but has never led directly to further deposits - yet. I have copied the heads of department in to some of these e-mails, when I know the author is already comfortable and happy with WRAP, although I've no idea whether they pay any attention to the e-mails!
What do I say to our top content authors? Here's a template, which I don't often have to vary much....
I'm writing to inform you that your paper in WRAP: (REFERENCE)
Is the most popular paper in WRAP in the last month. I'm keeping our highly read authors informed of what I can about the visitors to their content. I should point out that it is actually the record that is being visited/read, rather than the full text itself. There have been NUMBER pageviews of the record describing your paper from DATE to DATE. All visitors came from a search engine, the vast majority from Google. Most looked at the record and went away again, but some explored the subject area in WRAP.
There is a great variety of keywords that have led visitors to your paper, including the following: (LIST KEYWORDS)
Visitors came to your record from NUMBER different networks, so it is not all Warwick people looking at your work. Noticeably academic/educational networks that your visitors came from include:
The vast majority (NUMBER) of visits were from within the UK, but your paper's record had visits from....PLACES.
There have been no great peaks and troughs of activity: visits come every day and remain at or under NUMBER per day.
I did a quick Google search for PAPER'S TITLE and your paper's WRAP record is Xth in the results list.
Whilst looking at the stats I might spot something interesting, which I would usually blog about here and write to people in the library who I think ought to know: managers and subject librarians, or even our internal e-mail newsletter to all staff.
I know that our library management group are interested in big numbers, like how many pageviews there have been since we went live, from how many hundreds of countries/territories, etc. They want to illustrate the success story that we're gaining in visitors every week as we grow in content ever more rapidly! In compiling such a news piece, I might look at our growth chart on ROAR as well, or at the number of items we hold for a particular department, to provide further background information about the interesting pattern.
I also send out a "newsletter" once a term, by e-mail to people who are interested in hearing more about WRAP. I know that they're interested because I introduced an "I would like to hear more" tick box onto the deposit form and they ticked it!
Otherwise, statistics might make their way into my presentations to departments or articles that I write to raise awareness of WRAP, or onto our web pages about the repository. They are something to say when we talk about WRAP and it's important to be able to give the detail and context that they provide, to keep people interested in our work.