RE: Data on circulation of books

[<Toby.GREEN@oecd.org>]

Joe, In an attempt to help you answer your last question on mass 
digitization:

We've been loading all our books into Google Books since Oct 
2005. They appear on Google roughly 10 weeks post-publication (it 
takes Google this long to process them) and we've got around 300 
books loaded now. According to the metrics reported by Google, 
we're getting some 200,000 page views a month and this figure is 
steadily increasing. Visitors are offered links to various online 
bookshops, one of which is our own. There has been no measurable 
change in the number of books sold via the various online 
bookshops listed, nor via our own. If the extra exposure drove 
demand for printed copies in libraries, one might expect print 
sales to libraries to increase over time (since they don't buy 
all our books as soon as they're published!). This doesn't seem 
to be happening: our printed book sales (via online or offline 
channels) are gently declining - like-for-like down by around 2% 
in 2006 in line with our trend since 2000.

I have to put this observation into context. All our books are 
also available via our e-library, SourceOECD. This is available 
to users via around 800 institutional libraries, the majority 
universities. Of these libraries, around 25% also buy printed 
copies of all our books. Access is on an all-you-can-eat basis, 
so there is no limitation on access or downloading. Like-for-like 
downloads grew by 42% last year (cf 38% growth in 2004 and 20% in 
2005), so growth was much faster last year than the previous two 
years. What was driving this extra demand?

I think there are three factors behind this jump in downloads: 1. 
the market as a whole for e-books seemed to 'gel' in 2006 with 
users accepting them more and librarians being more skilled at 
offering them via OPACs 2. Extra online exposure via systems like 
Google Books 3. Our new offline promotion efforts (we send 
targeted users at subscribing institutions a magazine that, among 
other things, lists all recent books with links to them in 
SourceOECD).

Of course, this doesn't answer your question directly, since you 
are wondering if this exposure might drive demand for the print 
copies on library shelves. I have no data on this. But the 
observations above would suggest that digitization is increasing 
use of scholarly books quite significantly, but not driving use 
of print editions if access to an e-book is easily available.

I hope this is helpful.

Toby Green
Head of Dissemination and Marketing
OECD Publishing
Public Affairs and Communications Directorate
http://www.oecd.org/Bookshop
http://www.SourceOECD.org  - our award-winning e-library
http://www.oecd.org/OECDdirect  - our new title alerting service

-----Original Message-----
[mailto:owner-liblicense-l@lists.yale.edu] On Behalf Of Joseph J. Esposito
Sent: 13 February, 2007 12:41 AM
To: liblicense-l@lists.yale.edu
Subject: Data on circulation of books

Perhaps the members of this mailgroup can help me with some 
questions about the circulation of books in academic libraries.

A distinguished academic librarian told me that "most books never 
circulate."  Allowing for rhetorical exuberance, I was wondering 
what the facts are behind "most" and "never."  Is it that "many 
books circulate only rarely," or "some books never circulate, but 
a larger group circulates only rarely," or "almost all books 
circulate, but a sizable portion circulates rarely,"--or some 
other qualified formulation?

The questions that come to mind are these, defining "books" in 
the ordinary sense (e.g., no other media types and hardcopy only; 
and not including books that are not designed to circulate such 
as reference books):

*What percentage of books never circulate at all?  Does this 
percentage vary by the size of a library?

*What percentage circulate rarely (assuming that there is a 
library convention for what "rarely" means in this context)? This 
would be the so-called Long Tail of a collection.

*Are even those books that never or rarely circulate findable in 
an electronic card catalogue, which is searchable by various 
means, or is the lack of circulation a function of inadequate 
"finding" tools?

*If the full text instead of just metadata of rarely circulating 
books were exposed to search engines of various sorts, what is 
the expectation for the increase in circulation?  In this context 
presumably online viewing would count as a form of circulation.

What's driving these questions is what the practical effect of 
mass digitization will have on materials use.  My hypothesis is 
that for books there will be a discernible but modest increase in 
use.  Any information that could help me prove or disprove this 
hypothesis would be welcome.

Joe Esposito