v23 #3 Developing a Model for Long-Term Management of Demand-Driven Acquisitions

by | Jul 14, 2011 | 0 comments

Download PDF

by Michael Levine-Clark  (Collections Librarian, Associate Professor, University of Denver)  <michael.levine-clark@du.edu>

Anyone who has been to a library conference in the last several years has been to a talk or two on patron-driven acquisition (PDA).  Academic libraries have reported — with mixed results — about PDA pilots of varying sizes.  But all of these pilots have been add-ons to standard collection building practices.  I want demand-driven acquisition (DDA) — my preferred term for this concept — to be the primary means for my library to build collections, and I think that it should become the main method for collection building at most academic libraries.  In order to move to a DDA model on a broad scale, libraries need to develop new ways of thinking about our collections and services, publishers need to make eBooks available at the time of publication and provide a local print-on-demand (POD) option, and approval vendors need to develop tools to help us manage the complex workflow involved in this new process.

At the University of Denver (DU), we have been interested in DDA for quite some time.  From 2001 to 2005, we participated in a consortial DDA program with NetLibrary through the Colorado Alliance of Research Libraries.  Though this model had some flaws, the concept has always seemed sound, so in 2010 we moved pretty aggressively into DDA for eBooks again with Ebook Library (EBL).  We also automatically purchase interlibrary loan requests that meet certain criteria relating to publisher, price, and date of publication.  But these projects have represented fairly small portions of our overall collecting strategy.  I have long wanted to expand to include the widest range of monographs possible into a DDA program.

Ideally, we would be able to develop a DDA plan that would deliver all of our scholarly monographs as eBooks with a local POD option.  This would allow us to provide our users with a much wider range of choices, would keep us from buying material that will never be used, and would make it possible for us to provide instant (or almost-instant in the case of POD) access to any title.  Because most scholarly monographs are not made available as eBooks in a timely manner, and even fewer are available for local POD, this is not possible now.  Instead we need to develop a hybrid model that incorporates print and electronic books, and we need to do so using the existing approval structure.  We have been working with YBP to develop a plan that includes our existing EBL DDA plan, eBooks from other aggregators and publishers, and print books from our current slip notification plan.  When this model is fully implemented, we will be able to provide access to a wide range of books and subjects on demand.1

The more I have thought about DDA, the more I realize that what originally seemed to be a fairly easy concept is actually tremendously complex.  If widely adopted by academic libraries — as seems likely to be the case — DDA will force us to reconsider how we define the library collection; will allow us to rethink traditional library functions; and will necessitate development of new tools and services to manage the complex workflow involved in the process.

Library collections have traditionally been defined by ownership and have recently been defined as well by access.  Under a DDA model, collections are expanded to include everything that can be acquired in a reasonably short time.  The collection is the pool of titles available for potential lease or purchase — and its size is bounded only by the library’s budget.  Unlike a traditional purchased collection or even a collection based on long-term (annual) lease, a DDA collection is fluid.  This sort of collection is based on immediate access rather than long-term stewardship.  For academic libraries, that is a radical shift, and one that makes many librarians uncomfortable.  If we embrace DDA, then we should also embrace — or at least accept — the notion that our collections are not permanent.  We should be comfortable with the notion of purchase or lease at the point of need, even if that need is decades after the publication date — which should be possible because eBooks should not go out of print.

Library functions have evolved over centuries where the primary mission was to preserve physical collections.  We have long been the only place to get particular books because after we bought them they went out of print.  The only way for a user at another library to gain access to those books was to use the owning library’s copy.  Interlibrary loan (ILL) evolved as a relatively efficient means of getting these books to their users, and it is a service that libraries are rightfully very proud of.  While ILL is incredibly efficient for transporting books from one library to another, it makes no sense at all for eBooks.  With DDA, libraries have already begun leasing books for a day or a week rather than purchasing them to keep forever.  We should replace ILL whenever possible with a DDA model that will allow us all to go to a single location to gain immediate access to any eBook, often at a cheaper cost than borrowing a print book from another library.  As POD becomes a viable option, it can be added to this function.  ILL should become a means of borrowing only materials that are not available — either as POD or eBook — on demand.

While DDA allows us to reconsider our collections and the ILL function, it also requires us to invent an entire array of tools and services to manage it.  As our collections expand to include a large pool of titles under consideration for demand-driven acquisition, we need to develop a management structure to allow us to keep the consideration pool at the optimum size, to keep it filled with the right mix of titles, and to remove titles that no longer make sense for the collection.  This process is too complicated to be managed by the library, so we will need vendors to help us develop these services.

This consideration pool will consist of all titles available for purchase or lease through a DDA model.  In DU’s case, it will consist not just of the EBL titles already there, but also of eBooks on other publisher and aggregator platforms and the print titles that will be added over time.  Since the goal of DDA is to provide our users with timely access to the widest array of titles possible, the pool should be kept as full as our budget will bear.  This means that we will constantly be adding and removing titles based on a wide range of criteria.

Filling the consideration pool is relatively easy.  We already use YBP profiling to add EBL titles to this pool.  We will soon add eBooks from other vendors and publishers, as well as titles based on our slip notifications — all also based on YBP profiling.  While we are using our existing profile now, it may make sense to write a less restrictive profile for DDA, adding more books in subjects that we have traditionally been unable to afford.  For instance, though DU does not have an architecture program, we do have some demand for books on architecture, so we purchase a handful of titles every year.  With an approval plan, our aim has always been to narrow the number of titles in this area down as much as possible.  With a DDA plan, it might be more logical to expand it as widely as possible under the assumption that we will still have relatively low demand for architecture so we should only buy the few titles specifically needed by our users.

A much more complicated aspect of DDA management is long-term maintenance of the titles available in the consideration pool.  At some point, we will need to remove titles for a number of reasons — age, content, replacement with a newer edition, and financial risk.  Some titles will be removed because they are no longer current or because the content no longer matches collecting needs.  More titles will be removed to keep the pool to the optimal size relative to budget; this is much more complicated than populating the pool in the first place.

Libraries will need to craft rules for how long particular types of material remain in the pool.  At DU, we assume that we will keep some subjects around longer than others.  For instance, we already know that books in the humanities are used much longer than those in subjects like business and engineering; it stands to reason that we should leave books in these sorts of subjects in the pool for potential purchase longer.  We also assume that some titles, or series, or publishers will be deemed to be core.  In other words, they will be seen as important enough to keep in the pool forever — even if they are never purchased.  These titles will have to be identified and tracked in some way to ensure that we provide access in the long term.  Other titles will be removed regularly — after a certain length of time in the pool perhaps — but only if they have not been used.  In a model that allows some use before purchase, we do not want to frustrate our users by removing titles that have recently been used.  In some cases we may want to remove titles temporarily.  For instance, if spending outstrips the budget, some portion of the pool might be blocked until the next fiscal year.  All of this is incredibly complicated and requires someone –— most likely our approval vendors — to develop the services to manage the pool.  In order to adopt DDA models on a large scale, we need vendors to develop these services.

Approval vendors are the logical developers and managers of these services.  They already have experience developing rules to get books into our collections based on a variety of subject and non-subject criteria.  They should expand those rules to include methods for managing DDA from multiple eBook vendors, and should develop new rules and services to help us keep the optimum number and mix of titles in our consideration pools.

Demand-Driven Acquisition is the hot topic of the moment, but so far it is too difficult to make it the primary means of building collections.  Publishers are not providing enough titles as eBooks (and certainly are not providing enough for local POD).  And there are no structures in place to allow libraries to easily shift from prospective to demand-driven collection building.  We need those shifts to occur because DDA can allow us to provide broader collections to our users and can make it possible to more efficiently access any book not owned in the local library.  Thoughtful management of a DDA plan will involve many of the same sorts of decisions always made in managing traditional collections — with the emphasis now placed on the pool of titles available for potential purchase: which titles to make available, how long to keep those books available, and how and when to weed titles.  The key difference between traditional collection building and demand-driven collection building is that after the point of initial selection into the consideration pool we will have to manage the size of this collection based on budgetary constraints.  While librarians should set these rules based on local collecting needs, we should employ vendors — just as we do now for our approval plans — to manage the complex sets of decisions we have made about our collections.


1.  For detailed discussion of the DU DDA plan, see Michael Levine-Clark, “Developing a Multi-Format Demand-Driven Acquisition Model,” Collection Management 35, nos. 3-4 (2010): 201-207, and Michael Levine-Clark, “Building a Demand-Driven Collection: The University of Denver Experience,” in David Swords, ed. Patron-Driven Acquisitions: the Revolution, the Evolution, and the Future. Munich: K.G. Saur, 2011 (forthcoming).


Sign-up Today!

Join our mailing list to receive free daily updates.

You have Successfully Subscribed!

Pin It on Pinterest