by Jason Price, PhD (Director of Licensing Operations, SCELC Library Consortium)
and Maria Savova (Director of Information Resources and Systems, Claremont Colleges Library)
In a sense, book acquisition strategy has always been driven by access. Before eBooks, the only way to provide library users with immediate access to a large collection of books was to anticipate which ones they would want and purchase them title by title: whether by catalog, slip notification, or customized purchasing profile. The universe of books that a student or faculty member had immediate access to was defined by the size and age of their home institution’s library and the skill of the librarians who built their collections. Speculative purchasing of books that would serve users well in the short and long run was a fundamental library function.
The advent of Internet search and e-commerce, massive book digitization projects, two-day print book delivery, and instantaneous eBook “delivery” brought about by the likes of Google and Amazon has multiplied our users’ universe of immediate book access to a global scale. They easily discover and expect access to the broadest possible range of books, regardless of local ownership. Their concern has shifted from “Does the library have this?” to “How long will it take to get it?” Our users now live with the growing expectation (and under-recognized luxury) of instant delivery in our brave new access-driven world. These changes demand that library content and services quicken and expand, lest they become irrelevant.
On the print side, most libraries have responded in kind, reducing book interlibrary loan waiting periods from a couple of weeks to a couple of days through union catalogs and coordinated courier service. Some have even adjusted their print book acquisitions to ensure rapid access to a broader range of titles, establishing demand-driven print programs and/or cross-library cooperative collection development.
It is within the rapidly evolving world of eBooks, however, that the far greater opportunity to expand book access lies. Internet-based technology has birthed a variety of eBook acquisition modes that differ by orders of magnitude in number of accessible titles per acquisition dollar. As each of these modes emerged, library collection developers made bottom-up or top-down decisions about how much of their library’s budget to dedicate to each mode. These decisions were constrained by historical practice and made without foresight into the nature of future eBook acquisition modes or the (varied) discount levels that would be associated with them.
Some libraries responded quickly, adding subscription and purchase packages — new acquisition modes that significantly expanded access over the title-by-title pre-purchasing approach of the print-based world. Others were (and many still are) much more conservative, avoiding subscription because of its ongoing expense and impermanence, and/or avoiding packages because of the lack of choice that results from their intrinsic inclusion of the unwanted along with the desireable. These aversions are understandable, given unpleasant experiences with ejournal package subscriptions that insidiously command more and more than their share of the budget whilst adding more and more mostly unwanted content.
Then came patron-driven acquisition. After some initial missteps by NetLibrary and a (resulting?) period of significant resistance in the library community, aggregator-based demand-driven acquisition has been widely adopted. It has been popular enough to spur publishers to offer their own low-tech version, known as evidence-based selection. These two acquisition modes offer the best balance of the access and ownership worlds — a likely driver of their burgeoning popularity. Both modes offer temporary access to a large customizable set of eBooks at no charge, with payment due upon significant use or through purchase at the end of a semester or year-long period of unfettered access. Furthermore, they offer libraries a much greater level of choice than any fixed package can.
Although usage-driven acquisition modes have hit a sweet spot on the access versus ownership spectrum, they have a number of drawbacks, not the least of which are reduced predictability, sustainability, and value (i.e., higher cost per accessable title). Given the inherent tradeoffs built into each of these new acquisition modes, libraries have the opportunity — or perhaps the obligation — to define and implement a comprehensive eBook acquisition strategy that uses a combination of acquisition modes to maximize access while ensuring that their users are served as effectively as possible over time.
eBook Acquisition Modes currently in Use in Academic Libraries
(1) Aggregator Subscription Packages cover a wide variety of content in multiple subject areas, from multiple publishers and publication years. These collections provide unlimited simultaneous access to tens of thousands of eBooks at an annual subscription price that is 99% less than the cost of ownership of the included titles. They do not include the most recently published titles, which are embargoed for 6-18 months.
(2) Publisher-direct Subject-specific Package Purchases tend to be discounted by 30-90% depending on the publisher and the age of the content. In addition to this low per-title cost of perpetual ownership, these packages proffer unlimited, simultaneous, DRM-free use. Furthermore, in an open Web search, their contents are much more easily discovered than are their siloed, aggregator-hosted cousins.
It is worth emphasizing here that maximizing access per acquisitions dollar does not always require “just-in-time” acquisition of content. Deep discounts on subscription and publisher packages make these “just-in-case” approaches very attractive as a foundation for the usage-driven “just-in-time” approaches that follow.
(3) Short-Term Loan (STL) — a special case of pay-per-view — is another way to provide discounted access to eBook content. While the discount for STL access (currently ranging from 40-85% off purchase list price depending on publisher and duration) is significantly lower than the discount for subscription access, and STL access is far more ephemeral than subscription or purchase package access, this mode is far more flexible, in that title-by-title choice is driven directly by the user and unused titles are not paid for. Even though titles that are “triggered” for purchase via DDA after one or more STLs ultimately end up costing more than list price, the STL/DDA combination allows libraries to offer immediate access to a larger number of titles with no upfront charge. Since libraries tend to allocate as much to these programs as they can afford or consider appropriate, this approach has the effect of “stretching the dollar” by redistributing the same total expenditure to a larger number of titles.
(4) Demand-driven Acquisition (DDA) programs typically trigger title-by-title purchase after a usage-based threshold is met. They require sophisticated real-time usage tracking at the title level in order to apply the trigger appropriately, which is available heretofore only on aggregator platforms. Although titles purchased by DDA are rarely available at a significant discount, one recent model offers DDA of a large collection of slightly older eBooks (i.e., with a date profile matching that of subscription collections) at a discount as high as 45% off list price.
While short-term loan is not required for a viable DDA program, the most successful examples employ either affordable STL or a liberal free access policy to ensure that eBooks that are purchased are truly needed by the library’s users. For example, JSTOR’s current model allows for six free chapter views or four free chapter downloads before purchase is initiated. Similarly, Rittenhouse offers three free downloads before the library is required to make a purchase decision.
(5) Evidence-based Selection — Another way of vetting content before paying full price for ownership is the evidence-based mode. It opens up a large pool of titles for unlimited user access, deferring purchasing decisions to a later date. This allows selection to be based on local usage data, among other considerations. As with classic DDA, this mode allows the library to leverage “ownership” dollars to expand current access without requiring sophisticated (yet inflexible) thresholds to make the final purchase “decisions.” Although this mode requires a minimum spend commitment and generally does not include a discount, it is commonly used for providing access to a publisher’s most recent titles, making it a good complement to discounted publisher backfile collections.
(6) Title-by-title Purchasing — this acquisition mode was the only option in the print world and is still preferred by many libraries for eBooks. Although it may include reviewing general lists of newly published eBooks, it is often dominated by profile-based purchasing which is either automatic or notification-based, and usually includes a mixture of both. Although print books profiled or ordered through a bookseller like YBP can be discounted as much as 15 to 20%, title-by-title purchasing of eBooks is rarely discounted more than 5%.
Caveat Emptor — Let the Buyer Beware
Although the optimal ratio of expenditure across these eBook acquisition modes will vary across libraries, we believe that every library’s optimal strategy will incorporate most of them. Too often we have heard librarians express a principle-based judgement against a particular acquisition mode. “My library doesn’t subscribe to eBooks…” “We only buy eBooks title by title…” “We won’t buy eBooks with DRM or simultaneous user restrictions…” Each of these all-or-nothing positions is based on a rational objection to the shortcomings of a particular acquisition mode. Each also belies, however, the rational cost-benefit evaluation that is necessary to maximize access via a well-reasoned comprehensive acquisition strategy.
The optimal mix of acquisition modes will vary depending on the relative value each library places on key factors that differ among them, including:
- Choice of content / Collection quality
- Discount — the cost per accessible title relative to list price
- Ease of use — (Simultaneous, Downloadable, Discoverable)
- Permanence — mandate for current versus future access to the scholarly record
- Predictability of cost — relative predictability or risk
Development of a truly optimized strategy should be based on a thorough evaluation and prioritization of all of these factors as aspects of each library’s institutional value system. Although the recommendations below reflect the authors’ emphasis on maximizing access by minimizing the cost per available title, we encourage libraries to formalize and nurture a broader perspective.
Three Recommended Tactics that Complement Demand-driven Acquisition
(1) Subscribe to both Major eBook Aggregator Subscription Packages — Some libraries continue to hold the line against subscription eBooks, which is understandable given the ballooning portion of the average library budget that is dedicated to subscription-based products. The difference between the annual percentage price increase of subscriptions and a library’s annual budget increase percentage creates an “inflation gap.” Left unchecked, it reduces the discretionary proportion of the library budget, typically used for book purchases. In fact, in librarians’ minds, “more subscriptions” often equals “fewer books.” The logical result is a regular cancellation program and a growing aversion to new subscriptions. Thus it is not without a great deal of respect that we disagree with libraries that adhere to a “no subscription” eBooks policy. eBook access must not fall victim to the tyranny of the inflation gap just because eBook packages are the new kids on the subscription block. Instead, their value should be assessed relative to other subscription-based resources and eBook acquisition modes. In our opinion, their value proposition is simply too great to be refused.
At a 99+% discount off purchase prices for one year of access, it would take between 100 and 300 years of subscription expenditure to gain ownership of the same collection of eBooks. This makes the relative value of these subscriptions 10-30 times greater than a primary source database subscription (given that the standard primary source purchase cost is ten times the cost of subscription).
Subscription collections include unlimited simultaneous user access: full price purchase of those same eBooks on the same platform are subject to simultaneous or maximum use restrictions.
Like DDA, subscription collections expand access, allowing usage statistics to be used to identify titles for perpetual access purchase (as desired), and they allow for this at a tiny fraction of the cost of short-term loan.
The value of these collections increases in the context of a demand-driven program because they provide a critical mass of potentially relevant eBooks whose cost is NOT based on usage. Given roughly 170,000 unique titles available to a library that subscribes to both EBSCO’s eBook Academic and ebrary’s Academic Complete, and a non-overlapping 42,500 eBook demand-driven acquisition profile (for example), unlimited usage of an average of four out of every five eBooks or these aggregator platforms is covered at no additional charge to the library.
It is perhaps the case that a majority of libraries do subscribe to one of these two collections. To our knowledge, however, it is relatively rare for libraries to maintain subscriptions to both. Some cite the significant 30% overlap between the packages as a deterrent. Yet the annual cost of the collections is low enough, relative to the cost of purchasing even a small subset of the unique eBooks in the second collection via (STL &) DDA, to effectively justify subscription to both despite this overlap.
(2) Use DDA to Fill in the Gaps between Discounted Acquisition Options — We recommend targeted purchase of discounted publisher packages in subject areas of interest for eBooks that are not available via subscription — like front list titles for example — or eBooks that are considered of “ownership value” for long-term preservation. Along with subscription, these should comprise the foundation of every academic library’s eBook “collection.” DDA programs with STL are most effective at filling in the gaps and providing access to additional titles that libraries might not want to retain in perpetuity. At the same time, the built-in purchase trigger ensures that well used content will not disappear from the collection.
Over the past year-and-a-half, however, many publishers have more than doubled their short-term loan prices, and some have even completely disqualified their most recent titles from STL availability, presumably to compensate for the lower number of books being purchased as DDA has been more widely adopted. Libraries that have not adjusted their profiles to compensate are exhausting their annual DDA budgets on STLs well before the fiscal year is out. Others have downsized their programs to cover fewer titles, thereby decreasing the value of STL/DDA programs to libraries and their users. If this trend culminates in short-term loan becoming a thing of the past, it is likely to lead to much more focused DDA profiles, designed to keep the end user engaged in the process, rather than to expand the local library’s breadth of access. In the face of these changes, publishers, aggregators, and libraries continue to struggle together to develop a better solution aimed at expanding access in a sustainable way.
(3) Minimize eBook Pre-purchase Plans — In most cases, the range of new eBook acquisition modes (1-5) should allow libraries to minimize speculative “just-in-case” purchasing of individual eBook titles at full price. In the context of demand-driven acquisition, eBook pre-purchase plans (more commonly — and colloquially — known as “eBook approval plans”) require careful scrutiny. Why would a library buy its eBooks speculatively when it can make those eBooks (and more!) available for no additional charge, and avoid purchase until the point of significant use? In a world uncomplicated by short-term loans which significantly inflate the ultimate cost of purchase, the answer is clear — no library would! Instead, eBook “approval” plans should profile a larger set of the best eBooks that the library would purchase — if it could afford them all. These profiles then establish the eBooks that will be available for DDA, counting on a predictably-sized subset of them being triggered for purchase, and allowing users to select the most useful from among a set of the best possible. As such, eBook pre-purchase plans make little sense when they can be replaced by well-crafted DDA profiles.
In the presence of increasingly costly short-term loans, however, minimizing eBook pre-purchase plans is a bit more complicated. To the extent that a library can be certain that an eBook will be triggered, it makes sense to pre-purchase it to avoid having to pay for short-term loans and full purchase price for the same eBook — for example, course adopted eBooks. If publishers were to extend the same discounts to “approval” eBooks as they do for print, they would give libraries an additional incentive to pre-purchase speculatively. In a way, e-approval plans “pre-package” content by certain criteria, determined by the profile, most commonly LC class, similarly to the way publishers pre-package subject content, and thus it makes sense for them to be discounted more.
Where Will We Go From Here?
This article has provided a snapshot of currently available eBook acquisition modes and recommended a few tactics for employing them as part of a much-needed library-specific overall eBook acquisition strategy. The cauldron and crucible of current conditions jointly tended by aggregator, publisher, and library alike will undoubtedly “bubble up” more options and (hopefully) will consume some of the dross. As we proceed in search of a sustainable future for eBook publishing in the electronic world, key questions arise: what portion of each library’s budget should be spent on books, and to what extent is it the library’s responsibility to maintain it?
Booksellers, publishers, and aggregators have observed a significant decrease in book sales to academic libraries over the past few years. Have libraries greatly reduced their book spending as they have implemented STL and DDA? Or has that spending just shifted away from ownership and toward access? It seems likely that both have occurred.
To the extent that libraries have reduced book spending via implementation of demand-driven acquisition, they have reduced the perceived sustainability of academic book publishing. It also seems clear, however, that as libraries shift book spending toward access and away from 100% ownership, the payoff from publication will shift from a typical three-year profit and loss calculation to one where income from book access (and purchase!) continues on a longer time horizon. If libraries ensure that any money that is saved via DDA and STL in a given year is immediately reinvested in books (rather than other content types), that will go a long way toward supporting the path to ongoing book publishing sustainability that libraries, publishers, and aggregators are seeking together. In order to do so, libraries will need to actively defend their book budgets against the ongoing impact of the subscription-based inflation gap.