David, can you give us a bit of information on why you stopped using
xISBN, and how you are finding the LibraryThing ISBN workset data to
be? I was worried that the LibraryThing data would have too many errors
in it (of the 'false inclusion' sort, rather than 'false exclusion').
Have you found it to be pretty good though?
Also, can you provide more information about how you provide links to
LibraryThing? I may be interested in doing that, but I don't like
providing 'blind' links that may or may not lead to an actual record, I
only like providing links that I know will lead to an actual record. Are
you using the ISBN file for that?
Jonathan
David Pattern wrote:
> Hi Tim
>
> This is using data from the last 12 month period:
>
> 2,055,707 - total keyword searches
>
> The percentage figures in brackets are a % of that total keyword search
> figure...
>
> 62,410 (3.0%) - clicks on "Did you mean?" spelling suggestions
> 33,709 (1.6%) - clicks on "People who borrowed this, also borrowed"
> links
> 4,131 (0.2%) - clicks on xISBN/ThingISBN other edition links*
> 2,058 (0.1%) - clicks on keyword combination suggestions**
> 381 (0.02%) - clicks on "We think you might be interested in"**
>
> * We stopped using xISBN a few months ago and now only use the
> downloadable ISBN list from ThingISBN (thanks Tim!)
>
> ** These last 2 are fairly new. The 1st gives you a suggestion if you
> enter just 1 or 2 keywords, and it suggests up to 4 other keywords that
> you might want to combine with your search (the suggestions are based on
> all previous keyword searches by our users, rather than some uber list
> of "good" keywords chosen by staff). The 2nd gives the user a single
> personalised suggestion when they log into their library account.
>
> The keyword searches used the following indexes...
>
> 1,347,864 (65.6%) - general keyword (default)
> 252,494 (12.3%) - title keyword
> 243,996 (11.9%) - author keyword
> 68,022 (3.3%) - subject keyword
> 42,040 (2.0%) - journal title keyword
>
> In terms of the number of results found...
>
> 439,618 (21.4%) - searches that gave zero results
> 529,848 (25.8%) - searches that gave more than 100 results
> 88,767 (4.3%) - searches that gave more than 1,000 results
> 5,777 (0.3%) - searches that gave more than 10,000 results
>
> Number of keywords used for the search...
>
> 716,539 (34.9%) - 1 keyword
> 741,439 (36.1%) - 2 keywords
> 319,356 (15.5%) - 3 keywords
> 141,490 (6.9%) - 4 keywords
> 67,479 (3.3%) - 5 keywords
> 30,418 (1.5%) - 6 keywords
> 14,779 (0.7%) - 7 keywords
>
>
> What I don't currently measure are:
>
> 1) searches on alphabetical indexes (i.e. "Starts with") -- partly
> because I don't think I've ever seen a student use one of them, and
> mostly because I wouldn't be able to do anything useful with the data
> ;-)
>
> 2) number of item pages viewed (i.e. you've run a search, how many of
> those results do you actually click on to find out more details?)
>
> 3) how many people click on the links to LibraryThing in our OPAC ;-)
>
> Having the number of item pages viewed would give a more accurate %
> figure for the clicks on the borrowing suggestions. So, not quite 3%,
> but 1.6% is still good!
>
> regards
> Dave Pattern
> University of Huddersfield
>
>
>
>
>
>
> Dave Pattern
> Library Systems Manager
> University of Huddersfield
>
> email: d.c.pattern_at_hud.ac.uk
> phone: 01484 472043
>
> -----Original Message-----
> From: Next generation catalogs for libraries
> [mailto:NGC4LIB_at_listserv.nd.edu] On Behalf Of Tim Spalding
> Sent: 04 February 2008 19:59
> To: NGC4LIB_at_listserv.nd.edu
> Subject: [NGC4LIB] Browsing percentages / analytics
>
> Has anyone done analytics on click-throughs for recommendations, tags
> and etc. in the OPAC? I'm no.
>
> Has anyone seen numbers for ecommerce sites? What percentage follow
> the recommendation links on Amazon anyway?
>
> LT is starting to collect some of this data for LTFL. So far-six hours
> of data-it looks like, when book recommendations appear, they are
> clicked on 3% of the time. That seems pretty good to me, considering
> most OPAC use is for specific finding, not for generalized browsing.
>
> Anyone have any info about comparative numbers? Anyone done Google
> Analytics and gotten useful stats out of them?
>
> T
>
> --
> Check out my library at http://www.librarything.com/profile/timspalding
>
>
>
>
>
>
> This transmission is confidential and may be legally privileged. If you receive it in error, please notify us immediately by e-mail and remove it from your system. If the content of this e-mail does not relate to the business of the University of Huddersfield, then we do not endorse it and will accept no liability.
>
>
--
Jonathan Rochkind
Digital Services Software Engineer
The Sheridan Libraries
Johns Hopkins University
410.516.8886
rochkind (at) jhu.edu
Received on Thu Feb 07 2008 - 12:01:18 EST