Re: Usage-based pricing (was ebooks in libraries a thorny problem)

[dillon@mail.utexas.edu]

We obtain usage information on print publishers from the OPAC.

For E-books, we do not use vendors that keep usage information 
from their customers.  Without useful statistics, we would have 
no way to evaluate what kind of job we were doing, what kind of 
job the vendor was doing, and whether or not we were spending our 
money wisely.  We primarily use e-book vendors that provide 
e-books from multiple publishers.  This year, our usage-based 
e-books are coming in way under budget.  We've seen this trend 
before.  When readers are able to pick their own titles, usage 
tends to cluster around the best titles (80/20 rule) and since we 
have an automatic purchase plan under which we pay for usage up 
to a certain point, and then we automatically purchase the best 
books -- as the years roll by, most of our usage is occurring in 
books we have previously purchased.

Of course, as a library that has long embraced embraced 
usage-based pricing, we have learned that you do have to limit 
the amount of content/$$ you are willing to put risk, or your 
budget could get out of control.  We try to control our risk by 
limiting our usage-based pricing to no more than 100,000 e-book 
titles at a time.

--Dennis

Dennis Dillon
Associate Director for Research Services
University of Texas Libraries, University of Texas at Austin


----------------------
Dennis,

Are you able to obtain usage stats for publishers who do not
provide those stats?  If so, I would be interested in having you
share the process with which you accomplish this.

With thanks,

Susan

Susan Raidy-Klein
Associate Director, Collection Services
Simmons College
Boston, MA 02115
susan.klein@simmons.edu

-----Original Message-----
From: owner-liblicense-l@lists.yale.edu
[mailto:owner-liblicense-l@lists.yale.edu] On Behalf Of
dillon@mail.utexas.edu
Sent: Thursday, April 15, 2010 6:03 PM
To: liblicense-l@lists.yale.edu
Subject: Re: Usage-based pricing (was ebooks in libraries a thorny problem)

- Show quoted text -

The Univ of Texas has been accessing e-books on a patron-driven
pay-per-view model (along with with an automatic purchase of
frequently used titles) for several years.

Budgeting has been a snap.  We have ebook usage data for our
campus going back to the late 1990's and extrapolating this data
(X number books in our profile set, with an expected usage rate
of Y, with avg cost-per-use of Z equals - the budget).

Buying books on spec in the traditional manner is simply too
risky a behavior for our library to unthinkingly continue doing
in this type of economy.

We run performance numbers (circulations) for every publisher in
both print and ebook formats, and then eliminate automatic
purchasing plans for any high priced or low performing publishers
- and move these low performers/high-priced publishers to strict
title-by-title firm order purchasing, which is performed by
subject specialists working within a firm budget.

We are not going back to wholesale speculative book purchasing
anytime soon - too many of our books purchased on spec are simply
not read in the first few years, or ten years, or sometimes ever.
We cannot afford these kinds of opportunity costs when there are
other needs that our readers are clamoring for.

Library purchasing of books requested via inter-library-loan, the
loading of MARC records into OPACs and then paying for our users
to instantly read the ebooks they are interested in, designing
plans to let users' select print-on-demand, current paper
imprints, or even out-of-print titles by interacting with the
OPAC or other discovery tool- all of these kinds of efforts
insure that every book we purchase or rent will find a reader.

We believe that there is ample room in the market for fewer books
to be purchased on speculation, and for more publisher revenue to
be generated by usage-based pricing, patron-driven selection, and
print-on-demand options.  Moving to usage-based pricing and
patron-driven selection means publishers and librarians have to
rethink some paradigms and be more in tune with their readers,
but that is not necessarily a bad thing.

Dennis Dillon
Associate Director for Research Services
University of Texas Libraries, University of Texas at Austin