by Peter Spitzform (Collection Development Librarian, University of Vermont) <Peter.firstname.lastname@example.org>
When we started our Order-on-Demand pilot project at the very end of 2007, we did so because large swaths of our book collections were going unused. A then-recent study of circulation data showed that fully 40% of our books had not been checked out for years after they began sitting on our shelves. While our study predated the economic crash by a year or so, times were nevertheless getting tougher, and we felt that it was only a matter of time before university administrators began to request more accountability for how we are spending our money.
Our grim circulation statistics were hardly representative of our being an outlier. In fact, the University of Vermont turned out to have the same rate of (non-) circulation as virtually every other circulation study to be found in the literature, beginning with the famous Pittsburgh Study from 1973 conducted by Allen Kent.1
As the reality of our circulation study sank in, we considered creating a print-on-demand trial. While our central printing office on campus did not own an Espresso Book Machine, they nevertheless possessed advanced, sophisticated equipment, and they were willing to join us in an experiment to print and bind books from electronic files when (and only when) patrons indicated their need for these titles. It didn’t take long, however, to determine that publishers were not making their front-list titles available electronically, and we already owned the older titles they were providing the files for.
Next, we considered what would happen if we were to provide access to book titles we did not yet own, but would commit to purchase at the point of patron need for them. Following an interesting discussion with our rep from YBP, who took our ideas seriously, we were able to institute a demonstration project wherein we loaded the MARC records for books from three large, academic presses (Wiley, Palgrave Macmillan, and Oxford), link these records to order forms within our OPAC and allow patrons to order the books we did not yet own; we made a commitment to get the books in their hands within three working days from when they placed the order, if their need was urgent (as noted on the online form). We were told that, as of November 2007, we were the only library in the United States that had instituted a version of what we now know as Patron-Driven Acquisition (PDA) for print books, though our term of choice is Order-on-Demand. We very much wanted to test the hypothesis that books specifically wanted (and ordered) by patrons might have a better rate of circulation than books we obtained in the traditional way (primarily from our shelf-ready books received on our profile with YBP).
After three full years in operation, I believe the evidence suggests that this project has been a success, though there is not unanimous agreement about that in our library. Some librarians feel that our collections role should remain unchanged from years or even decades ago, wherein the library should purchase any and all books that fall within our “approval” profile with YBP in case there should ever be a need for these books by our patrons. I maintain that while there are students who are undoubtedly disappointed that a book that looks promising is not yet on our shelves, it is more common for patrons to simply order the book when they need it, and pick it up when they are notified that it is awaiting them at the circulation desk. Here are some data about our Order-On-Demand program that might allow you to draw your own conclusions.
In fiscal 2009, 505 books were ordered on demand from the records in our OPAC for books we did not yet own. If we had not done this pilot study, but instead continued to receive books from these three publishers based on our approval plan with YBP, we would have received 1,759 books. That means that we did NOT receive 1,254 books that we otherwise would have. Given this, we spent $33,149.64 for the books our patrons ordered on demand, while we would have spent $142,231.12 under the old method. So, in FY2009, we saved $109,081.48 and roughly 120 linear feet of shelf space. The books that were ordered on demand circulated an average of 1.18 times since the start of the program, while books received on our approval plan over the same period circulated an average of .67 times. (This circulation data was collected in February and March of 2010.)
In fiscal 2010, we ordered 590 books on demand, while we would have received 1,502 books under our approval plan. By not receiving the 912 other books we would have received on our approval plan, we saved roughly 90 linear feet of shelf space. In that year, we spent $37,971.86 for books ordered on demand, though we would have spent $88,604.60 if those books had shown up on our approval plan, saving us $50,632.79. Meanwhile, the books ordered on demand circulated an average of 2 times each since the inception of the program, while books received on our approval plan over the same period circulated an average of 1.18 times. (This circulation data was collected in March of 2011.)
With books ordered on demand circulating about 70% more than books bought “just in case,” this model of collection development makes intuitive, spatial, and financial sense.
And clearly, the meteoric rise of PDA for eBooks takes our Order-On-Demand trial much further. Rather than having to impose a delay for desired material to arrive for patrons, eBooks are instantly viewable and readable. Thus, the primary objection some librarians had to our pilot study — that the books should already be waiting on the shelves for patrons when they need them — vanishes. eBooks via the MARC records that link to them from the OPAC are provided with every bit of discoverability that print books have, with the highly desirable bonus of paying for books only when they’re needed and used. We have just instituted a Patron-Driven Acquisition program for eBooks in conjunction with YBP and EBL. It has only been active for six weeks so far, so we do not yet have sufficient data to draw any conclusions. We are using the same three publishers we used for our print trial, substituting eBook records whenever they exist.
To review so far: Books ordered on demand show superior circulation rates; they provide savings in money, savings in shelving space; books are at least as discoverable as those received in the usual fashion, and provided at the point of need.
I’m not seeing a big downside here.
The UVM Libraries rely on two off-site storage facilities to hold our older and less (or non) circulating materials in order to make room for more volumes on the libraries’ shelves. One is a huge underground space on the far side of campus with half devoted to journals, the other half to books, all on compact shelving. The other is a rented commercial warehouse space in a neighboring town. Both sites are at capacity. Renting a third off-site storage space would be cost-prohibitive and, frankly, irresponsible. In the past, we had no alternative to acquiring print books, which take up room on physical shelving. Electronic books may well help libraries manage their collections less expensively, and acquiring only those specific titles that patrons want, rather than purchasing all those that we think patrons might someday need, will certainly reduce the footprint of the print collections.
My vision for the library of the not-sodistant- future takes all these positive aspects of PDA into account, and maximizes them to the point where the patron is actually presented with more access to materials than in the old model of libraries-as-repositories of books and other knowledge objects. It might look something like this:
Book publishers get beyond their anxiety about eBooks, and provide simultaneous publication of their books in both print and electronic formats. This will do away with the current problem of many publishers clinging to a model where the print edition is released, followed (maybe) by the electronic version at some indeterminate point in the future. Once print and electronic editions of books are published simultaneously, libraries (and the scholarly book-buying public) will be able to structure their buying in a rational, predictable manner, rather than the largely hit-and-miss fashion now that results from libraries desiring to purchase more eBooks within their normal workflow, but being forced to purchase the print because of uncertainty when the electronic format will be issued. At that point, libraries will load the MARC records for any and all books that fall within their academic scope and mission, and patrons will encounter these in their normal course of conducting research. But instead of noting the call number and trudging into the stacks to find the book they need, they simply click on the link in the catalog and access the entire book on the computer (or tablet, or smart phone) they are using, and start reading. Despite the slight variations in how different eBook vendors set up their rules, most will incorporate some form of a free preview of an eBook, a loan period of some length, the first, say, two to ten clicks on that book, followed by an automatically-triggered purchase of the book on its Nth click (N being determined by the library.) For the initial loan periods, libraries will pay some fraction of the list price of the book, followed by a payment for the full list price of the book when a purchase is triggered. All vendors will create a rule (based on their agreement with publishers) that a purchased eBook can be opened a certain number of times in a given period, after which the library must purchase an additional copy of the book.
Because it may take a very long time for all publishers to be able to issue eBooks (either simultaneously with the print editions they publish, or on their own — it is currently not economically feasible for very small presses to offer eBooks), libraries will continue to receive books in print for the foreseeable future, alongside the numerous eBooks they make available. Consequently, it will be in virtually all libraries’ interests to work with book vendors, who can supply whichever format of books a library prefers, and consolidate the invoicing for both purchases and loan charges for eBooks. Libraries will not have to juggle invoices either from numerous book publishers or vendors — at least no more so than they do today.
Several thorny issues will need to be resolved over time in order for this new method of collection development to be fully functional. One is when — or even whether — to list their books in WorldCat. WorldCat has become an ever-stronger tool for discovery, especially when patrons utilize it via Google or Google Scholar searching. However, current limitations placed on eBooks by vendors (at the insistence of publishers) preclude eBooks from being lent to other libraries via ILL in the normal sense of lending an entire book. OCLC holdings, then, have at least partly evolved away from their traditional use as a tool for ILL, and more toward their function as a discovery tool. Publishers worry that the lending of eBooks will diminish their income to unsustainable levels, and generally allow only parts of eBooks to be lent to other institutions as a result. Until ILL rules for eBooks settle on a new norm, there will be confusion about WorldCat holdings, and what they have traditionally represented to the library community. If libraries were to load their “holdings” of MARC records for books they do not yet own into OCLC, there would no doubt be ILL requests for those books. Should libraries be purchasing books when other libraries request them for ILL purposes? Or should libraries list only those books that have been fully purchased on their PDA program to be listed in WorldCat? And if libraries are prohibited from lending entire eBooks, but can only lend chapters, what will the long-term effects of this be on traditional interlibrary loan? In my opinion, librarians and the publishing community should be working together to address these concerns. While ILL should not be a primary driver of libraries’ participation in OCLC holdings, it is nevertheless a time-honored tradition in much of the world to loan materials between libraries and their patrons, and eBooks should not be cast in the role of killing ILL.
As eBooks proliferate, and especially as libraries move toward patron-driven acquisition of eBooks, another issue that arises is that of reading devices. eBooks can certainly be read online easily enough. The question of portable reading devices raises the problems of economic disparity. Patrons who have laptops, tablets, smart phones, etc. can easily download eBooks to them and use them away from the library. While such devices are relatively inexpensive, given their power, they nevertheless can be prohibitive to a number of patrons. Many libraries lend devices to patrons, but doing so can be a hardship for libraries, or at least a complex problem. Keeping such devices working and clean of viruses and other infections can be challenging, and such simple events as rainstorms and gravity can take their toll on them.
None of these problems are even remotely insurmountable, and as standards and norms slowly begin to evolve, eBooks will become ubiquitous. We should welcome such evolution and begin to embrace the fact that patrons do a better job at selecting what they need for their research, and that what they select will tend to be used by others, as well. We need not feel threatened by the fact that the books we choose to purchase “just in case” are not used nearly as much as those selected by our patrons when given the opportunity to choose them “just in time.”
1. Kent, Allen. Use of library materials: the University of Pittsburgh study. New York: M. Dekker, c1979.