Michael Levine-Clark, Allison V. Level, Joan G. Lamborn, and George Machovec
Demand-driven acquisition (DDA) allows libraries to provide users with immediate access to a wide range of titles to be purchased at the point of need. The model is most often used for ebooks, but some libraries use it for print books as well. A “consideration pool” of titles is loaded into the online catalog, allowing users to discover and use books for which the library then pays. For ebooks, a predetermined number of short-term loans (STL) triggers the purchase of the book according to a set of parameters established by the library. For print books, a patron can request that the titles be ordered from a record in the catalog by the library. By 2011 many academic libraries had begun exploring DDA to supplement traditional models such as approval plans and firm ordering.1 DDA has now become a common component of academic library collection development strategies. While there are advantages to an individual library to include DDA as a tool for collection development, DDA for ebooks presents challenges for resource sharing.
Academic library consortia have typically had a strong resource-sharing emphasis; buying a book for a local library in print meant that partner institutions could easily borrow that book for their patrons. The Colorado Alliance of Research Libraries (Colorado Alliance) has in place Prospector, a shared catalog that patrons rely on for unmediated borrowing from other libraries. As member libraries began to acquire more and more ebooks, ever-larger portions of search results in Prospector were taken up by ebooks that could not be used beyond a single library due to restrictions stipulated in ebook licensing agreements.
In May 2012, nine members of the Colorado Alliance began a two-year pilot DDA program with YBP Library Services (YBP) and two major ebook aggregators, E-book Library (EBL) and ebrary. The nine participating libraries are: Auraria Library (serving the University of Colorado Denver, Metropolitan State University, and the Community College of Denver), Colorado College, Colorado Mesa University, Colorado State University (CSU), Regis University, University of Colorado Colorado Springs, University of Denver (DU), University of Northern Colorado, and the University of Wyoming. Prior to the start of the project, CSU and DU had put in place DDA for both print books and ebooks. The Colorado Alliance decision to embark on a pilot project was made to allow all members of the consortium to experiment with DDA, to explore resource sharing for ebooks, and to determine whether DDA could work in a consortial environment. Consortial DDA has had mixed results to date, and the Colorado Alliance project is no exception. This chapter reports on the first two years of the pilot, including the decision by the group to continue into a third year.
LITERATURE REVIEW
Demand-driven acquisition involves the acquisition of ebooks—either ebooks only or ebooks in combination with print books. Ebook usage has been increasing but it is not without problems or concerns. William H. Walters’s article provides, in the form of a bibliographic essay, a detailed description of encounters with the selection, integration, and management of ebooks in academic libraries. He outlines the current state of ebooks in academic libraries, including their availability and non-availability, and presents information about licensing considerations. His concluding list of challenges provides points for discussion and contemplation for those working in collections, acquisitions, or public services in academic libraries.2
Two publications with literature reviews of note are an early article on DDA by Judith Nixon, Robert S. Freeman, and Suzanne Ward, which provides a good beginning discussion and highlights articles on the topic that mostly cover the period 2000–2008,3 and Candice Dahl’s 2012 article, which includes a literature review of early DDA articles and provides points to keep in mind when a library is starting a DDA program.4
The NISO recommended practice for “Demand-Driven Acquisition of Monographs” explains the rationale behind DDA, provides a detailed overview of how various DDA models work, and examines some of the differences between local and consortial DDA.5
Early articles that document a local DDA ebook program at one university include the Karen S. Fischer et al. discussion of the 2009 University of Iowa DDA program. Using ebrary as its vendor, Iowa loaded an initial 12,000 discovery records from 28 publishers into the catalog based on its YBP approval profile. The article provides extensive detail of the initial program and includes data on costs of books by publisher, subject areas purchased, subject analysis of all titles used, title usage, top titles used, and additional tables. Based on eleven months of data, 60,000 records were loaded, 9,387 titles were used (15 percent) and seventy-three titles had over fifty uses.6
William Breitbach and Joy E. Lambert report on the California State University Fullerton library’s DDA program, which was created to increase access to ebooks. The team formed to work on the project selected EBL as the vendor and created a profile directly with the vendor that generated about 25,000 records for the initial load. CSU-Fullerton used librarian-mediated approval for any short-term loan over $25. Three STLs were followed by the purchase of the title on the fourth use. Based on the initial eight months of data, the authors concluded that the program helped them work toward meeting the goal of better access to electronic materials.7
Two chapters from the book Patron-Driven Acquisitions, edited by David A. Swords, detail DDA programs at the University of Texas and DU. Dennis Dillon covers the work of the University of Texas and EBL to create a DDA profile and load records into the catalog for title selection by patrons. The STL was set to one day with a purchase of the ebook on the third use request. After using it for four years, Texas was very pleased with the program.8
Michael Levine-Clark’s chapter in Patron-Driven Acquisitions details DU’s DDA program, which was set up using YBP and EBL, and is demand-driven for both ebooks and print books. DU previously participated in other ebook pilots, including the Colorado Alliance of Research Libraries proto-DDA netLibrary program. In the current program DU decided that duplication of titles by format was acceptable. DU set the STL to one day and purchase of the title on the fourth use. From the beginning, the DU program was about broadening access to collections: “DDA allows us to build strong and lasting research collections without paying for the books until the point of need. By so doing, we can build broader and deeper collections than we ever could under the traditional acquisitions model.”9
Two articles that provide good information about project workflows and implementation decisions are by Carol Joyner Cramer for the Wake Forest University program10 and Kay Downey for the Kent State University (KSU) program.11 Downey et al. describe KSU’s six-month pilot DDA program and analysis of use of purchased print books relative to usage of ebooks in a DDA program. The article has a good literature review that documents DDA programs at other universities. It addresses these four questions about its DDA model:
1. How does DDA align the library’s collection with current user requirements?
2. Does DDA lead to more active use of the library’s book collection?
3. Is DDA cost effective as an acquisition model?
4. What issues are associated with DDA, and how may these issues be addressed?12
For libraries looking to set up a DDA program, early questions revolve around creating profiles or adapting established book purchase profiles, creating an initial load or pool of records, establishing workflow procedures, and creating open communications between all parties involved in the DDA program.
As DDA projects have become more prevalent among individual libraries, experimentation has occurred at the consortial level as well. Some libraries participate in both their own local programs as well as consortial projects.
Tonia Graves et al. describe the Old Dominion University (ODU) local DDA program and its participation in the Virtual Library of Virginia (VIVA) program (www.vivalib.org/). The ODU plan covers all subjects with no publisher restrictions. It has a higher maximum price limit and requires three STLs before purchase. The VIVA plan is limited to science, technology, engineering, and math subjects from a limited number of publishers. For the VIVA plan, there were initially ten STLs before purchase, but that number was eventually raised to twenty-five.13
The University of Illinois at Urbana-Champaign (UIUC) is a member of two consortial groups, one of which is the Consortium of Academic and Research Libraries in Illinois (CARLI). Lynn Wiley and Elizabeth Clarage present information on a range of print book and ebook DDA projects at UIUC, some of which are local and others consortial. The article shows that sometimes it takes multiple pilots to test out ideas before settling on a model that will work for those involved.14
The most well-known consortial DDA programs include bigger consortia like the Orbis Cascade Alliance and the Ontario Council of University Libraries (OCUL); however, there are useful examples of other recent consortial DDA pilots and programs. George Machovec provides a helpful introduction and overview of consortial DDA programs, placing them in the broader context of ebook development, and discusses the challenges ebooks present for resource sharing.15
In 2011, the Orbis Cascade Alliance introduced an early consortial DDA program for its thirty-seven libraries. In 2010, a team had begun looking at DDA models and vendors. The Orbis Cascade directors mandated that all of the libraries participate in the pilot, which began using EBL and YBP.16 Susan Hinken and Emily McElroy detail the planning for the Orbis Cascade model and discuss the rationale for vendor selection and other recommendations for the pilot.17 Jill Emery and Bonnie Park’s 2012 “serial conversations” article is an interview with nine librarians and vendor representatives involved in the program. The article provides valuable insights from some of those most directly involved.18 One interviewee, Emily McElroy (Oregon Health and Science University) remarked
What this project has demonstrated is the need for ongoing communication of our strategic agenda and what it means to be a member of the [Orbis Cascade] Alliance. It has also demonstrated that even a project as small as this requires an investment of staff time, patience with the process, and that workflow matters.19
Throughout the article, Emery and Parks probe with engaging questions, including asking participants if they would have done something differently from the beginning of the pilot. The three articles together combine to present personal perspectives and detailed specifics of the Orbis Cascade model, which was still going strong in 2014.
The Ontario Council of University Libraries (OCUL) includes twenty-one publicly funded universities in Ontario, Canada. OCUL has a strong commitment to consortial collection development programs as demonstrated by the consortially created Scholars Portal, which is its digital library ebook platform.20 At the 2011 Electronic Resources and Libraries (ER&L) Conference, Kate Davis et al. reported on OCUL’s sixteen-school, $150,000, nine-day DDA experiment, which used a model similar to that used in the Colorado Alliance netLibrary program. The money flowed quickly, and after just a few days 467 books were triggered or purchased.21 Davis et al.’s 2012 article provides more details about the program and specific perspectives from two institutions. It addresses the development of their DDA model, creation of the title list, feedback from several schools, and plans for the future. According to Davis, there was support for additional trials.22
Lorraine Huddy talks about the CTW Library Consortium (Connecticut College, Trinity College, and Wesleyan University) pilot, which began as part of a 2008 Mellon Foundation grant for collaborative collection development. The schools added DDA to the pilot and used MyiLibrary. The CTW ended up exploring other DDA options and planned to implement a new pilot with YBP and EBL.23
The California State University system has experimented with three different ebook trials. The University Chancellor’s fund provided money to fund two pilot DDA programs, one for 2011 and one for 2012, with twenty-three campuses included. California State University has a history of other consortial collection development programs, so DDA was a natural extension. The California State University pilots used different vendors, and as the second pilot was concluding, it was decided to continue with a third pilot. As of April 2013, the ebook committee recommendation was, “to put consortial PDA on hold until the [California State University] examined the possibility of ebook package subscriptions, specifically from ebrary, EBSCO, and Elsevier.”24 The California State University article also includes a literature review that briefly discusses the Triangle Research Library Network (TRLN) summit meeting about ebooks, the Orbis Cascade DDA program, and the OCUL program.
The University of California (UC) System established a pilot consortial DDA program with YBP and ebrary, which focused on university press titles and targeted subjects in the arts, humanities, and social sciences. The pilot, including nine campuses, was set up with three STLs and purchase on the fourth use for most publishers, and four STLs and purchase on the fifth for other publishers. Additional pilot parameters included a $250 price limit and a pool of books with publication years of 2010–2014. About 1,600 titles were in the pool as of mid-2014. The project is still in process.25
In 2011, the University of Florida (UF) and Florida State University (FSU) worked together to develop a shared DDA program. Florida’s State University System (SUS) has an infrastructure that supports collaborative collection development through the Council of State University Libraries (CSUL). The UF and FSU librarians worked with Ingram Coutts to develop a profile and launched a joint plan that included the 900 publishers available from MyiLibrary. Initially, 3,265 records were loaded; by January 2013, the record count was up to about 8,000. Initial funding was $60,000 from each school. In the second year the two law schools joined the program, with each contributing $20,000 for books in the Library of Congress K class.26
The three big state universities in Arizona joined together to support the Arizona Universities Library Consortium (AULC), a cooperative demand-driven model. Arizona State University (ASU) began looking at demand-driven models in 2009 and established a local DDA program. In 2011 the University of Arizona (UA) started a local plan similar to ASU’s plan. In 2012 the AULC began using the Ingram Content Group for a shared DDA plan. To fund the project, ASU and UA each contributed 40 percent of the cost and Northern Arizona University contributed 20 percent. At the commencement of the project, 2,211 records were loaded and a book with three uses triggered a purchase.27
As shown in the literature, experimentation with the demand-driven model has occurred in individual libraries as well as at the consortial level, with some libraries participating in both local and consortial projects. Consortial DDA projects range in size from the CTW Library Consortium of three small libraries to the Orbis Cascade Alliance’s thirty-seven. Sources of funding also vary. In the case of the CTW consortium, grant money was used to seed the project. Funding from the chancellor was used for the California State University system project. In other cases, the participating institutions made financial contributions. Many of these are opt-in projects, but Orbis Cascade mandated participation for all members.
Consortial projects use different vendor-aggregator combinations—Ingram with MyiLibrary, or YBP with EBL or ebrary. Consortia also use different approaches to create pools of titles to be considered. The University of Florida and Florida State University relied on a shared profile, which is more workable for a pair of institutions than a larger group. Other consortia, such as Orbis Cascade, base selection on publisher lists. The group of publishers affects the subject content offered. The University of California system limited the pool to university press titles with a focus on the arts, humanities, and social sciences.
CONSORTIAL EBOOK PROGRAM
The Colorado Alliance of Research Libraries (www.coalliance.org), a nonprofit 501(c)(3) library consortium, began in 1974 and was formally incorporated in 1981. The original mission of the consortium was cooperative purchasing of expensive print products to be shared among the members. By 2014, the program had grown to over 250 group contracts representing over $13 million. The Colorado Alliance currently includes thirteen academic libraries and one public library and focuses on collaboration and reducing costs. The Prospector union catalog (http://prospectorhome.coalliance.org) allows for discovery and unmediated borrowing across all Colorado Alliance libraries as well as an additional thirty non-member institutions.
Since the mid-1990s the Colorado Alliance has licensed e-resources for its members. The Colorado Alliance first began licensing ebooks as a consortium in 1999 with netLibrary, the first major aggregator of ebooks. Early in its corporate history, netLibrary introduced the first demand-driven acquisition model, called patron-driven acquisition, and its first consortial contract was with the Colorado Alliance. The Colorado Alliance tested the netLibrary system in March 1999, and after its April 1999 official launch became the first customer—and therefore the first consortium—to participate in a demand-driven acquisition program for ebooks.28
The original netLibrary patron-driven acquisitions program was designed so that a book would be purchased after the second viewing; the first use was free. There was no leased or short-term loan access, no free preview period except for the table of contents, no triggers for printing or copying, and no downloading because ebook readers did not exist then. This first attempt at a demand-driven program, although crude by today’s standards, set the stage for what have become the sophisticated DDA programs in use today. The early patron-driven program sold one copy of the ebook to the consortium once a purchase was triggered at list price (eventually changed to 155 percent of list price—a multiplier of 1.55). The software only supported one copy per reader (i.e., one simultaneous user). If simultaneous uses were warranted, then additional copies of the book had to be purchased. In reality, additional copies were never purchased because one copy sufficed for a small consortium.
The Colorado Alliance successfully ran the program from 1999 to 2005 but it eventually ceased for several reasons. First, in 2005 publishers begin to institute a multiplier for consortia. By requiring that the Colorado Alliance purchase more than one copy, the new contractual requirements essentially doubled the cost of the program. Although multipliers are now common for consortial ebook programs, modern ebook platforms have mitigating technology such as free preview periods. Second, with the crude technology of requiring a purchase on the second viewing even if the viewing is only for seconds, the cost of the program spiraled out of control, which was similar to what occurred with OCUL a few years later. Finally, by 2005, new players in the ebook aggregator marketplace were releasing much better user interfaces that made the netLibrary interface somewhat obsolete.
After the collapse of the consortial Colorado Alliance netLibrary contract in 2005, individual member libraries moved into other ebook contracts to varying degrees. DU immediately began subscribing to ebrary’s Academic Complete collection, and over the next few years added additional subscriptions to ebooks through Books 24x7, Knovel, and Safari. By 2011, DU had additional contracts in place for publisher packages of ebooks from Duke University Press, Oxford University Press, and Springer, and had begun its own large DDA program with EBL. CSU explored DDA through an initial publisher-based pilot with EBL in 2010. In 2011, the program changed to a YBP profiled program that now includes EBL and ebrary. CSU also has ebooks through selected subscriptions and publisher packages. The University of Colorado system began a DDA program with MyiLibrary around the same time. Other libraries in the consortium had varying degrees of experience with ebooks post-netLibrary, but most had relatively little experience with DDA.
While the early primitive patron-driven ebook program pioneered by netLibrary and the Colorado Alliance of Research Libraries introduced the concept of DDA, in 2001 EBL began developing the model we now understand as DDA, with a free browsing period and short-term loans that ultimately lead to a purchase.29 Other aggregators, including ebrary and EBSCO, have embraced this model. With the overwhelming success of this new DDA model, a number of consortia began exploring DDA again around 2009.
Individual libraries and consortia have been piloting a variety of different consortial ebook models with the intent of obtaining the most efficient use of funds and leveraging the new technology that publishers and aggregators have developed. Dozens of consortial DDA ebook programs are extant in North America at the present, some of which are referenced in the Literature Review section. Two are highlighted below.
One of the earliest and largest consortial ebook initiatives in the United States was developed by the Orbis Cascade Alliance in the Pacific Northwest, also referenced in the Literature Review section.30 Orbis Cascade, in its current configuration, was created in 2003 by the merger of two regional consortia, Orbis (Oregon) and Cascade (Washington), and has been one of the most successful library consortia in the United States. In 2009–2010 the consortium went through a detailed process to develop an RFI for a consortial ebook program, which was ratified in August 2010 with the selection of EBL as the key vendor in the demand-driven pilot using the “multiplier model,” in which the consortium shares ownership of titles when they are purchased at some multiple of list price. Several characteristics made it an interesting program being watched by many. First, the program mandated that all Orbis Cascade members join the program. Second, costs were tiered by library size/type so that smaller libraries (e.g., community colleges) paid a lower participation fee while larger academic institutions paid substantially more. The program proved very successful and now has approximately $1 million per year in expenditures (https://www.orbiscascade.org/ebooks/). As an early adopter of a large consortial ebook DDA program, Orbis Cascade discovered that it could throttle the rate of spending depending on the number of short-term loans before a purchase was triggered. With its use of the multiplier, Orbis Cascade established the model being followed by many other library groups, including the Colorado Alliance.
Novanet developed another early consortial DDA program in 2011 that uses a different model (www.novanet.ns.ca/novanet-ebook-dda-pilot-project/). Founded in Nova Scotia in 1988, Novanet includes eleven post-secondary institutions. Unlike Orbis Cascade, Novanet purchases only one copy of an ebook at a time (i.e., a multiplier of one), but when a threshold of use is reached in a particular contract year, additional copies are purchased. The advantage of this is that only one copy of a title is purchased at a time and additional copies are purchased as needed, keeping overall purchase costs down. The program moved from a pilot to a permanent program in February 2014.
As purchases of ebooks increased among Colorado Alliance libraries, and as CSU and DU began to experiment with DDA plans, the challenges of resource sharing in a new environment in which licensing terms restricted access to the purchasing institution became clear. This situation was underscored in the shared online catalog Prospector, where users affiliated with one institution could see, but not access, ebooks purchased by another institution. While the DDA model ensured that books purchased would have at least some use, that use was limited to users affiliated with a single institution.
A demand-driven pilot project at the consortial level provided an opportunity to gain experience with shared access to ebooks in a consideration pool, shared triggers for purchases, and shared ownership of the titles purchased. It was also an opportunity to learn about cross-institutional demand. Were there some subject areas of interest across institutions or did subject interest tend to be associated with specific institutions? Cross-institutional demand also raised the broader question of whether DDA makes sense in a consortial environment. In the local environment most titles are used once or twice. Does it make sense to aggregate low usage across multiple institutions and then pay for ownership? In addition, most titles in a local DDA program appear to be used by one person for one class. Does it make sense to share ownership of titles used at one institution? Finally, for some Colorado Alliance libraries the pilot was an opportunity for some libraries to learn about DDA, and for some that had not purchased many ebooks, it was an opportunity to learn about ebooks.
The possibility of a Colorado Alliance demand-driven pilot project was explored in 2008–2009 by some members of the Shared Collection Development Committee, including Michael Levine-Clark from DU and Patricia Smith from CSU, and representatives from EBL and Blackwell Book Services, but the sale of Blackwell to YBP put those discussions on hiatus.
By spring of 2011, the Colorado Alliance Shared Collection Development Committee (SCDC) again began discussing a consortial DDA program. In the summer of 2011 the committee met with representatives from YBP to learn how YBP could support such a program. By the close of the meeting the committee identified several project goals:
• Create a simple system to test demand-driven acquisition of ebooks on a consortial level.
• Make a large number of academic titles available.
• Create minimal barriers to accessing ebooks for users from the participating libraries.
• Lower the unit cost and aggregate cost of monograph acquisitions per participating library.
• Collect data from the pilot to support future planning.
• Looking to the future, develop a model to expand access to Prospector libraries that are not part of the Colorado Alliance project.
At the meeting the committee also developed preliminary parameters for the project. The project would be publisher-based and begin with 2012 imprints. It would be managed by YBP, which would control duplication with approval plans at the participating institutions and serve as a single source for record loads and invoicing. The publishers would be divided evenly between two aggregators, ebrary and EBL, to test patron and library staff satisfaction with the different platforms and services. Pricing for purchases would be calculated using a multiplier of the list price as was done in the Orbis Cascade project. The books purchased would be owned by all of the participating libraries. Funding for purchases and STLs would come from a central pool managed by the Colorado Alliance, with contributions from the participants. Purchases and loans would be tracked for future planning, but costs for the pilot beyond initial contributions would not be charged back to the institutions. The institutions that agreed to participate were Auraria Library (which serves the University of Colorado Denver, Metropolitan State University, and the Community College of Denver), Colorado College, Colorado Mesa University, Colorado State University, Regis University, University of Colorado Colorado Springs, University of Denver, University of Northern Colorado, and University of Wyoming. The University of Colorado Boulder, an Ingram Coutts/MyiLibrary customer, opted not to participate.
Before the project could begin, a multiplier needed to be determined and publishers needed to be recruited for the project. YBP produced lists of publishers that participated in DDA programs through EBL and/or ebrary, and the SCDC used these lists to identify possible publishers for the project. The committee assigned each targeted publisher to one of the aggregators based roughly on overall availability of ebooks by that publisher on the particular platform, and tried to have a fairly even spread of publishers, titles, and subject types through each aggregator.
YBP gathered data about purchasing patterns across the Colorado Alliance to help determine a multiplier. Although participating Colorado Alliance libraries typically bought fewer than two copies per title for this mix of publishers, the SCDC decided to negotiate a multiplier of 2.5 rather than 2 as a negotiating tactic to give publishers something up front. EBL and ebrary then negotiated with the possible publishers. Some publishers declined to participate, and publishers that required a multiplier above 2.5 were excluded.
The SCDC was unable to achieve an equal division of publishers between EBL and ebrary based on publication output eligible for DDA. Of the publishers that agreed to participate, a larger number were on the EBL list, so the subsequent pilot was heavily weighted toward EBL. The result was an imperfect mix of publishers between the two aggregators. The EBL publishers were: Continuum, DeGruyter, Edinburgh University Press, Facts on File/Infobase, Oxford University Press, Princeton University Press, Rodopi, SAGE/CQPress, University of California Press, and Wiley (multiple imprints); for ebrary: ABC-CLIO, Ashgate and Gower, Harvard University Press, Jessica Kingsley, John Benjamins, McFarland, and Stanford University Press.
The remaining decisions involved the specific DDA parameters and the budget. The free discovery would follow the practice already established by each aggregator, five minutes for EBL and ten minutes for ebrary. The committee agreed that there would be six STLs for each aggregator, with a purchase occurring after the sixth. To begin the project, the SCDC agreed that each participating library would contribute $12,500, for a total of $112,500, an amount that was expected to be enough for at least one year. The platform fees for the aggregators were waived.
The loading of provisional MARC records from the ebook aggregators into a library catalog or discovery interface is a key component to the success of the program. The concepts are simple but the execution has required continual attention by the technical services staff at each participating library. For EBL, YBP provides a provisional set of MARC records on a weekly basis, which are deposited on an FTP server in directories for each site. Libraries may download these records, and they are most frequently edited at a site level in MarcEdit before loading into an integrated library system and/or discovery interface. In most cases, sites are adding proxy server wrappers to the URLs. Once records are loaded into a local catalog, they are automatically transferred to the Prospector union catalog (http://prospectorhome.coalliance.org) where holdings are merged into a single record with URLs for each participating library. Users may access these ebooks from many different avenues, including their own local catalogs or discovery interfaces or the Prospector union catalog.
After purchase of a title, YBP sends a new MARC record for the “owned” title, which can be loaded into the local catalog as well as the union catalog to replace the provisional record. However, many sites have preferred to formally catalog the purchased titles in OCLC. In this case, one site has agreed to be the lead cataloging site and shares these records with the other libraries for local loading. Posting ownership holdings back to OCLC is determined at the site level.
Since the beginning of the DDA pilot, ebrary has had a separate workflow, which has been more problematic. Ebrary has done much of the site-level modifications for provisional MARC records. However, due to a variety of issues, ebrary has not always gotten the customized MARC records out in a timely fashion, making their current additions to the program less timely. Plans are in the works to change the workflow to match that for EBL.
Table 8.1 | Cost data by institution
Institution |
Contribution, $ (2 years) |
Contribution, % |
Auto purchase EBL, $ |
STL EBL, $ |
Total Cost EBL, $ |
Total Cost ebrary, $ |
Total Cost Both Programs |
|---|---|---|---|---|---|---|---|
|
Auraria Library |
16,500 |
10.8 |
6,608.51 |
10,248.47 |
16,856.98 |
3,179.21 |
20,036.19 |
|
Colorado College |
12,500 |
8.2 |
1,015.55 |
3,036.31 |
4,051.86 |
513.62 |
4,565.48 |
|
Colorado Mesa University |
12,500 |
8.2 |
1,249.62 |
2,381.31 |
3,630.93 |
318.05 |
3,948.98 |
|
Colorado State University |
32,500 |
21.3 |
23,633.24 |
35,383.58 |
59,016.82 |
2,368.31 |
61,385.13 |
|
Regis University |
16,500 |
10.8 |
3,158.96 |
5,462.66 |
8,621.62 |
948.55 |
9,570.17 |
|
University of Colorado Colorado Springs |
12,500 |
8.2 |
844.76 |
2,009.70 |
2,854.46 |
636.95 |
3,491.41 |
|
University of Denver |
16,500 |
10.8 |
5,784.90 |
8,370.90 |
14,155.80 |
1,711.51 |
15,867.31 |
|
University of Northern Colorado |
16,500 |
10.8 |
2,442.25 |
4,250.14 |
6,692.39 |
1,496.28 |
8,188.67 |
|
University of Wyoming |
16,500 |
10.8 |
5,593.79 |
7,123.06 |
12,716.85 |
1,465.06 |
14,181.91 |
|
TOTALS |
152,500 |
50,332 |
78,266 |
128,597.71 |
12,637.54 |
141,235.25 |
Note: 4 titles, totaling $478, were firm ordered from EBL, with that amount not assigned to any specific library.
During the first year of the pilot in 2012, there were nine participating libraries. Each library contributed $12,500 to a central fund held at the Colorado Alliance, which was then used to pay the corporate partners on a weekly billing statement issued by YBP. With no historic use data, an even split of the anticipated costs was intended as a way to get the program underway with the idea that some type of pro-rated cost split based on use would be instituted later.
At the close of the first year of the pilot the entire expenditure for EBL and ebrary STLs and auto purchases, along with a small amount for YBP cataloging and administrative fees, was almost $53,000, well below the amount collected (see table 8.1). This was due in part to a slow start-up by some libraries in getting the MARC discovery records into their local catalogs, and in part to a delayed launch by ebrary, caused by the need to implement some technology to support the necessary DDA functionality. Ultimately, ebrary launched almost six months after EBL. At the end of the first year of the pilot, of the $53,000 spent, only $3,152 was spent on ebrary titles. Most spending in the first year was on STLs; only $8,875 was expended on ebook purchases (16.75 percent of the total). Over the two years of the pilot, a larger portion of spending (about 38 percent) covered purchases. The vast majority of spending (91 percent) was with EBL. See table 8.1 for a breakdown of allocations and spending by library over the course of the pilot.
As a result of the slow start-up and very uneven use by the participating libraries, the group decided to change the funding contribution structure for the second year, asking five of the nine libraries to contribute additional funds to allow the pilot to continue through the second year. Based on its high usage in year one, CSU, the one ARL library in the program, would contribute an extra $20,000 (its use was 40 percent of the total; see table 8.5 later in this chapter). The four libraries with the next highest amount of usage would each contribute $4,000 more, and the four libraries with the lowest amount of usage would contribute no more during the pilot. By the end of the second year of the pilot in June 2014, the Colorado Alliance had expended $141,235 with 37.4 percent of the total expended on ebook purchases and 62.6 percent spent on leased access, or STLs.
After the two years of the pilot, the Colorado Alliance decided to continue the project into a third year with the original group of nine libraries. Since the deposit account had been largely depleted and with rising STL rates from some publishers, the group estimated that approximately $144,000 should be collected to keep the program going. After looking at the detailed use data, the Colorado Alliance developed a new cost-sharing formula in which each library would contribute a base fee of $3,000 plus additional costs based on a percentage of historic use during the first two years of the pilot. This allowed small users to stay in the program for reasonable fees, and larger users to pay their fair shares for what was being used.
At the start of the project in May 2012, there were 419 titles available from EBL. The initial load of records for ebrary, in September of that year, had just nine titles. By October 2014, there were 8,572 titles available from EBL, with an average of 59 titles loaded in the 146 weekly loads. Ebrary loads were less regular, with thirty-four of them occurring by May 2014, at which point there were 2,833 titles available. On average, these were larger loads, with eighty-three titles per load.
Within the Colorado Alliance project, EBL’s spread of titles available by publisher is uneven, with 65.69 percent coming from John Wiley and Sons. Together Wiley, De Gruyter, and Oxford University Press represent 83 percent of EBL titles (see table 8.2).
For ebrary, Ashgate and McFarland account for 55 percent of the titles available. Table 8.2 shows the breakdown of availability by publisher.
Table 8.2 | Percentage of titles available by publisher
EBL publishers |
% of paid usage |
% of coverage |
|---|---|---|
|
Continuum International Publishing |
1.97 |
1.90 |
|
De Gruyter |
0.89 |
8.38 |
|
Edinburgh University Press |
0.28 |
1.20 |
|
Editions Rodopi |
0.40 |
2.67 |
|
Infobase Publishing |
0.39 |
0.20 |
|
OUP |
9.26 |
8.88 |
|
Princeton University Press |
3.39 |
4.44 |
|
SAGE |
0.92 |
2.26 |
|
University of California Press |
4.12 |
4.38 |
|
Wiley |
78.39 |
65.69 |
ebrary Publishers |
||
|
ABC-CLIO |
6.30 |
2.58 |
|
Ashgate Publishing Ltd |
13.18 |
29.72 |
|
Harvard University Press |
17.50 |
9.40 |
|
Jessica Kingsley Publishers |
17.08 |
11.46 |
|
John Benjamins Publishing Company |
4.23 |
10.67 |
|
McFarland |
30.93 |
24.94 |
|
Stanford University Press |
10.78 |
11.24 |
At the beginning of the pilot, usage was relatively low, with only one auto purchase in the first four months. By the end of the pilot on June 30, 2014, 3,984 EBL titles had been used at least once, including free browsing. For the 2,745 EBL titles with some form of paid use, there were 260 (3.85 percent) purchases and 6,489 (96.15 percent) STLs. By contrast ebrary had 20 (1.66 percent) purchases and 1,186 (98.34 percent) STLs across 671 titles. Tables 8.3 and 8.4 show patterns of usage and corresponding spending over the course of the pilot.
Table 8.3 | EBL month-by-month activity
Month |
Auto Purchase |
STL |
Total |
Auto Purchase |
STL |
Total |
|---|---|---|---|---|---|---|
|
2012-05 |
10 (100.00%) |
10 |
$105.52 (100.00%) |
$105.52 |
||
|
2012-06 |
66 (100.00%) |
66 |
$549.98 (100.00%) |
$549.98 |
||
|
2012-07 |
1 (1.23%) |
80 (98.77%) |
81 |
$74.98 (8.68%) |
$788.84 (91.32%) |
$863.82 |
|
2012-08 |
111 (100.00%) |
111 |
$1,172.48 (100.00%) |
$1,172.48 |
||
|
2012-09 |
4 (2.21%) |
177 (97.79%) |
181 |
$577.36 (21.84%) |
$2,065.91 (78.16%) |
$2,643.27 |
|
2012-10 |
6 (2.42%) |
242 (97.58%) |
248 |
$2,011.90 (38.96%) |
$3,152.39 (61.04%) |
$5,164.29 |
|
2012-11 |
4 (1.83%) |
215 (98.17%) |
219 |
$599.52 (18.85%) |
$2,580.36 (81.15%) |
$3,179.88 |
|
2012-12 |
6 (3.31%) |
175 (96.69%) |
181 |
$879.82 (32.34%) |
$1,840.46 (67.66%) |
$2,720.28 |
|
2013-01 |
6 (3.14%) |
185 (96.86%) |
191 |
$884.62 (28.97%) |
$2,169.42 (71.03%) |
$3,054.04 |
|
2013-02 |
4 (1.36%) |
291 (98.64%) |
295 |
$912.34 (21.20%) |
$3,390.91 (78.80%) |
$4,303.25 |
|
2013-03 |
10 (3.79%) |
254 (96.21%) |
264 |
$1,971.07 (37.65%) |
$3,264.40 (62.35%) |
$5,235.47 |
|
2013-04 |
9 (2.90%) |
297 (95.81%) |
310 |
$1,509.48 (25.20%) |
$4,002.37 (66.82%) |
$5,989.85 |
|
2013-05 |
8 (2.72%) |
286 (97.28%) |
294 |
$1,146.96 (25.93%) |
$3,276.20 (74.07%) |
$4,423.16 |
|
2013-06 |
7 (3.06%) |
222 (96.94%) |
229 |
$1,704.70 (38.04%) |
$2,776.87 (61.96%) |
$4,481.57 |
|
2013-07 |
14 (5.81%) |
227 (94.19%) |
241 |
$2,309.72 (46.26%) |
$2,683.37 (53.74%) |
$4,993.09 |
|
2013-08 |
11 (4.55%) |
231 (95.45%) |
242 |
$1,856.92 (39.08%) |
$2,895.26 (60.92%) |
$4,752.18 |
|
2013-09 |
24 (6.72%) |
333 (93.28%) |
357 |
$5,098.22 (54.62%) |
$4,235.37 (45.38%) |
$9,333.59 |
|
2013-10 |
17 (4.59%) |
353 (95.41%) |
370 |
$3,616.58 (47.18%) |
$4,049.11 (52.82%) |
$7,665.69 |
|
2013-11 |
22 (5.06%) |
413 (94.94%) |
435 |
$3,782.66 (45.03%) |
$4,618.55 (54.97%) |
$8,401.21 |
|
2013-12 |
13 (3.80%) |
329 (96.20%) |
342 |
$2,127.00 (34.86%) |
$3,973.69 (65.14%) |
$6,100.69 |
|
2014-01 |
19 (5.03%) |
359 (94.97%) |
378 |
$3,741.84 (44.89%) |
$4,593.35 (55.11%) |
$8,335.19 |
|
2014-02 |
23 (6.65%) |
323 (93.35%) |
346 |
$4,716.48 (53.53%) |
$4,094.76 (46.47%) |
$8,811.24 |
|
2014-03 |
16 (4.30%) |
356 (95.70%) |
372 |
$3,758.99 (45.06%) |
$4,582.45 (54.94%) |
$8,341.44 |
|
2014-04 |
19 (4.49%) |
404 (95.51%) |
423 |
$3,834.84 (44.94%) |
$4,699.24 (55.06%) |
$8,534.08 |
|
2014-05 |
14 (4.88%) |
273 (95.12%) |
287 |
$2,474.55 (43.65%) |
$3,194.15 (56.35%) |
$5,668.70 |
|
2014-06 |
3 (1.07%) |
277 (98.93%) |
280 |
$741.03 (17.43%) |
$3,510.72 (82.57%) |
$4,251.75 |
|
TOTAL |
260 (3.85%) |
6,489 (96.09%) |
6,753 |
$50,331.58 (38.99%) |
$78,266.13 (60.64%) |
$129,075.71 |
*Overall totals include four titles firm ordered in June 2013 at a total cost of $478.
Table 8.4 | eBrary month-by-month activity
Month |
Spent, $ |
Purchase |
STL |
Total |
|---|---|---|---|---|
|
2012-10 |
139.26 |
0 |
6 |
6 |
|
2012-11 |
69.09 |
0 |
20 |
20 |
|
2012-12 |
393.50 |
0 |
15 |
15 |
|
2013-01 |
353.54 |
2 |
13 |
15 |
|
2013-02 |
270.19 |
2 |
14 |
16 |
|
2013-03 |
247.22 |
1 |
26 |
27 |
|
2013-04 |
722.60 |
3 |
42 |
45 |
|
2013-05 |
372.73 |
0 |
36 |
36 |
|
2013-06 |
283.53 |
0 |
35 |
35 |
|
2013-07 |
230.03 |
0 |
36 |
36 |
|
2013-08 |
280.57 |
1 |
30 |
31 |
|
2013-09 |
654.76 |
0 |
75 |
75 |
|
2013-10 |
977.51 |
1 |
104 |
105 |
|
2013-11 |
1,243.20 |
1 |
124 |
125 |
|
2013-12 |
589.96 |
0 |
57 |
57 |
|
2014-01 |
457.25 |
2 |
41 |
43 |
|
2014-02 |
852.23 |
0 |
105 |
105 |
|
2014-03 |
1,370.61 |
1 |
127 |
128 |
|
2014-04 |
1,342.95 |
4 |
115 |
119 |
|
2014-05 |
988.43 |
2 |
108 |
110 |
|
2014-06 |
498.38 |
0 |
57 |
57 |
|
TOTALS |
12,637.54 |
20 |
1,186 |
1,206 |
Usage patterns varied by institution, with CSU accounting for a disproportionately high percentage of usage for EBL titles, with 42.7 percent of the paid usage and 40 percent of all consortial usage, including free browsing. CSU accounts for 19.7 percent of the consortia’s total full-time equivalents (FTEs) while Auraria Library accounts for 30.6 percent of FTEs, but 13 percent of the consortia’s EBL usage (see table 8.5). Other institutions tend to be more closely aligned to their FTEs. For ebrary, usage more closely matched FTEs, with Auraria still lower than would be expected, but above CSU. Tables 8.5 and 8.6 show usage by institution.
Table 8.5 | EBL usage by institution
Institution |
% of total FTE |
Auto purchase |
STL |
Total purchase use (auto + STL) |
% of total paid usage |
Owned loans |
Unowned browse |
Owned browse |
Total usage |
% of total usage |
|---|---|---|---|---|---|---|---|---|---|---|
|
Auraria Library |
30.6 |
42 |
851 |
893 |
13.2 |
761 |
1,792 |
689 |
4,135 |
13.1 |
|
Colorado College |
1.7 |
7 |
223 |
230 |
3.4 |
74 |
441 |
121 |
866 |
2.7 |
|
Colorado Mesa University |
6.6 |
7 |
229 |
236 |
3.5 |
116 |
848 |
311 |
1,511 |
4.8 |
|
Colorado State University |
19.7 |
113 |
2,766 |
2,879 |
42.7 |
1,945 |
5,884 |
1,887 |
12,595 |
40.0 |
|
Regis University |
7.7 |
19 |
543 |
562 |
8.3 |
299 |
1,385 |
340 |
2,586 |
8.2 |
|
University of Colorado Colorado Springs |
7.0 |
8 |
191 |
199 |
2.9 |
175 |
410 |
150 |
934 |
3.0 |
|
University of Denver |
8.2 |
26 |
690 |
716 |
10.6 |
909 |
1,580 |
892 |
4,097 |
13.0 |
|
University of Northern Colorado |
9.5 |
14 |
397 |
411 |
6.1 |
192 |
811 |
234 |
1,648 |
5.2 |
|
University of Wyoming |
9.3 |
24 |
599 |
623 |
9.2 |
568 |
1,451 |
505 |
3,147 |
10.0 |
|
TOTALS |
260 |
6,489 |
6,749 |
5,039 |
14,602 |
5,129 |
31,519 |
Table 8.6 | eBrary usage by institution
Institution |
% of Total FTE |
Auto purchase |
STL |
Total purchase use (auto + stl) |
% of total usage |
|---|---|---|---|---|---|
|
Auraria Library |
30.6 |
5 |
278 |
283 |
23.5 |
|
Colorado College |
1.7 |
0 |
43 |
43 |
3.6 |
|
Colorado Mesa University |
6.6 |
2 |
18 |
20 |
1.7 |
|
Colorado State University |
19.7 |
3 |
250 |
253 |
21.0 |
|
Regis University |
7.7 |
1 |
110 |
111 |
9.2 |
|
University of Colorado Colorado Springs |
7.0 |
0 |
84 |
84 |
7.0 |
|
University of Denver |
8.2 |
4 |
142 |
146 |
12.1 |
|
University of Northern Colorado |
9.5 |
3 |
142 |
145 |
12.0 |
|
University of Wyoming |
9.3 |
2 |
119 |
121 |
10.0 |
|
TOTALS |
20 |
1,186 |
1,206 |
With a trigger of six STLs before a purchase, the Colorado Alliance DDA program is clearly tilted toward access over ownership. Year one of the pilot saw 16.75 percent of the expenditures made on ebook purchases, and year two saw purchases rise to 35.5 percent of the total expenditures for the year. This difference is easily understood in that STLs accumulate on titles that eventually trigger a purchase. As time goes on, a greater percentage of the expenditure should be for purchases due to accumulating STLs. This may be mitigated by other factors such as the addition of front list titles, whether the consortium decides to remove older titles from the consideration pool, or because older ebook titles may be found in packages such as the ebrary Academic Collection that some libraries license.
One goal for this pilot was to determine whether there was a wide enough spread of demand across institutions to make a consortial DDA program viable. One way to measure this is simply to look at the level of paid usage, STLs, and auto purchases across libraries. Clearly, a title used by only one library but then purchased at a multiplier of 2.5 would not make sense for a consortial purchase, while a title used by all nine libraries would be a great deal at a multiplier of 2.5. Since the usage of ebrary titles was relatively low, this portion of the analysis looks just at EBL titles.
Across all 260 of the EBL titles that had an auto purchase, for 28.85 percent (seventy-five titles), all paid leased STLs and owned auto purchases were at a single institution. An additional 33.85 percent (eighty-eight titles) of all leased and owned purchases were at just two institutions. Consequently 62.69 percent of the titles had paid usage by fewer institutions than the multiplier. On average, across these 260 titles, paid usage occurred at 2.23 institutions, which is below the multiplier of 2.5. See table 8.7 for a breakdown of paid usage by institution for titles with an auto purchase.
Table 8.7 | Paid usage of EBL titles with an auto purchase
Titles with an auto purchase—EBL |
||
|---|---|---|
|
All STLS/auto purchase by one library |
75 |
28.85 % |
|
All STLS/auto purchase by two libraries |
88 |
33.85 % |
|
All STLS/auto purchase by three libraries |
69 |
26.54 % |
|
All STLS/auto purchase by four libraries |
19 |
7.31 % |
|
All STLS/auto purchase by five libraries |
8 |
3.08 % |
|
All STLS/auto purchase by six libraries |
1 |
0.38 % |
|
TOTAL |
260 |
|
When factoring in an additional measure of usage, the post-purchase “owned loan,” i.e., any use of more than ten minutes after the auto purchase has occurred, the spread of usage improves among institutions. By this measure, only 17.69 percent (forty-six titles) of the auto-purchased titles were used by a single institution, 23.85 percent (sixty-two titles) by two institutions, and 58.46 percent used by three to seven institutions. On average, post auto-purchased use titles were used at 2.94 institutions, above the multiplier of 2.5. See table 8.8 for a breakdown of paid usage plus owned loans by institution for titles with an auto purchase.
Table 8.8 | Paid usage and owned loans of EBL titles with an auto purchase
Titles with an auto purchase—EBL |
||
|---|---|---|
|
ALL STLS/AUTO PURCHASE/OWNED LOANS BY ONE LIBRARY |
46 |
17.69 % |
|
ALL STLS/AUTO PURCHASE/OWNED LOANS BY TWO LIBRARIES |
62 |
23.85 % |
|
ALL STLS/AUTO PURCHASE/OWNED LOANS BY THREE LIBRARIES |
73 |
28.08 % |
|
ALL STLS/AUTO PURCHASE/OWNED LOANS BY FOUR LIBRARIES |
41 |
15.77 % |
|
ALL STLS/AUTO PURCHASE/OWNED LOANS BY FIVE LIBRARIES |
20 |
7.69 % |
|
ALL STLS/AUTO PURCHASE/OWNED LOANS BY SIX LIBRARIES |
14 |
5.38 % |
|
ALL STLS/AUTO PURCHASE/OWNED LOANS BY SEVEN LIBRARIES |
4 |
1.54 % |
|
TOTAL |
260 |
|
Looking more broadly at all titles with a transaction, there were 2,745 titles with at least one STL. Of these, 1,271 (46.30 percent) had a single STL, and so could only have been used at one institution, yet 1,775 (64.66 percent) of all titles with any STL had all paid usage plus owned loans at one institution. And 88.08 percent of these titles were used at one or two institutions. See table 8.9 for a breakdown of paid usage plus owned loans by institution for all titles with paid usage.
Table 8.9 | Paid usage and owned loans of EBL titles with at least one STL
All Titles with an STL—EBL |
||
|---|---|---|
|
All STLS/auto purchase/owned loans by one library |
1,775* |
64.66 % |
|
All STLS/auto purchase/owned loans by two libraries |
643 |
23.42 % |
|
All STLS/auto purchase/owned loans by three libraries |
224 |
8.16 % |
|
All STLS/auto purchase/owned loans by four libraries |
63 |
2.30 % |
|
All STLS/auto purchase/owned loans by five libraries |
22 |
0.80 % |
|
All STLS/auto purchase/owned loans by six libraries |
14 |
0.51 % |
|
All STLS/auto purchase/owned loans by seven libraries |
4 |
0.15 % |
|
TOTAL |
2,745 |
|
*1,271 of the titles use by one library only had a single STL.
Clearly, most titles do not get used across a wide range of libraries, but the spread may be great enough for consortial DDA to make sense.
Another way of thinking about the value of a consortial DDA program is to measure what usage and costs would look like if each library had put in place the exact same DDA program, but with no multiplier, at the local level. In this calculation, all STLs, owned browses, and auto purchases triggered by a particular institution are added together to determine that library’s activity as if it acted alone. The auto purchase is counted as an STL, and only counts as a purchase if there is a combination of seven STLs, owned browses, and auto purchases. Table 8.10 shows this calculation for each institution, and indicates that, with $111,388.22 in potential spending on EBL titles for libraries going it alone versus $128,597.71 in actual spending, the Colorado Alliance would have saved money by not undertaking the two-year DDA pilot.
While it could be argued that most individual libraries would have set up their programs with fewer than six STLs, this analysis confirms Doug Way and Julie Garrison’s findings that DDA programs become cheaper with more STLs prior to purchase.31
While this analysis of usage at the title level seems to indicate that shared purchasing of ebooks in a consortial DDA program is more expensive in the short term for the participating libraries, it fails to take into account the potential for usage in the third year and beyond, when titles already purchased incur no additional costs. It seems likely that there will be a leveling off point in the next few years, at which time the cost of usage by the group matches or even improves on the aggregate cost for the individual libraries. In the meantime, the costs are close enough, and the resource sharing benefits are clear enough that the Alliance opted to continue the pilot into the third year.
Table 8.10 | EBL purchasing if each institution had acted alone
If acting alone (STL loan transactions + owned title purchases) |
||||||
|---|---|---|---|---|---|---|
|
STL |
Auto purchase |
Titles with a trigger |
STL cost, $ |
Auto purchase cost, $ |
Total cost, $ |
|
|
Auraria Library |
— |
34 |
598 |
12,600.83 |
2,546.26 |
15,147.09 |
|
Colorado College |
256 |
4 |
192 |
3,283.21 |
188.89 |
3,472.10 |
|
Colorado Mesa University |
293 |
6 |
172 |
3,220.64 |
400.89 |
3,621.53 |
|
Colorado State University |
3,097 |
118 |
1,517 |
38,866.15 |
10,157.50 |
49,023.65 |
|
Regis University |
651 |
19 |
404 |
6,842.96 |
1,088.51 |
7,931.47 |
|
University of Colorado Colorado Springs |
242 |
7 |
153 |
2,539.62 |
279.94 |
2,819.56 |
|
University of Denver |
839 |
28 |
465 |
9,935.45 |
2,427.01 |
12,362.46 |
|
University of Northern Colorado |
506 |
11 |
281 |
5,465.98 |
785.02 |
6,251.00 |
|
University of Wyoming |
756 |
26 |
425 |
8,948.45 |
1,810.91 |
10,759.36 |
|
TOTALS |
111,388.22 |
|||||
As table 8.2 shows, there was uneven usage by publisher. On one extreme, Wiley accounted for 78.39 percent of the paid usage in the EBL program, while representing a smaller but still large 65.69 percent of the titles available. On the other end of the scale for EBL was De Gruyter, accounting for just 0.89 percent of the use but 8.38 percent of the coverage. As noted above, CSU accounted for a disproportionately high percentage of the overall use for EBL, but not for ebrary. Wiley alone among these publishers has a strong STEM focus, matching the strengths of CSU. A next stage in assessment of the program is to look at subject coverage of the titles used and correlate that to institution.
Another item for further study is the degree to which broader ebook availability at an institution drives usage. DU and CSU both have strong local ebook programs, and, as tables 8.5 and 8.6 show, these institutions used ebooks in the consortial DDA program at a higher rate than their FTEs would have predicted.
The Colorado Alliance DDA program does not exist in a vacuum, and the Colorado Alliance continues to explore other ebook options. Publisher packages remain appealing because they often have less restrictive rights attached. While publishers generally cannot do DDA in the same way that ebook aggregators such as EBL or ebrary can, they do offer evidence-based selection (EBS) models. Using an EBS model, the Colorado Alliance would deposit a set amount of money at the start of a program and then choose titles based on usage at the end of the program. While the Colorado Alliance has not yet begun such a program, negotiations are ongoing with several publishers.
Subscriptions to aggregated ebooks are also appealing to many libraries, with several Colorado Alliance members having subscriptions to the ebrary and EBSCO packages. As an extension of the DDA program with ebrary, the Colorado Alliance has discussed adding a consortial subscription to the ebrary Academic Complete package. This has the potential to complement an active consortial DDA program, with older unused titles moving to the subscription package.
DDA itself is evolving. As the second year of the pilot drew toward a close, many publishers announced that they were going to increase fees for STLs.32 Publishers have discovered that the decrease in sales under DDA is not offset by an increase in STL revenue, and their overall income on any given title has decreased to the point that they are concerned about their ability to continue publishing some books. As with libraries, DDA has been an experiment for publishers, and the experiment has shown them that their initial level for STLs was set too low. While this is not easy for libraries to deal with, the reality is that libraries need to adjust DDA programs, whether local or consortial, in some way. The group is watching this closely.
Many libraries have implemented DDA programs in the last five years or so, including several in the Colorado Alliance of Research Libraries. As libraries moved to this new acquisition model for ebooks, the question inevitably arose of what the consortial role in DDA should be. The Colorado Alliance, like most library consortia, has a strong history of resource sharing in terms of lending print books and other physical items among libraries, as well as in collaborative licensing of electronic resources to decrease costs for the group. Ebooks and DDA programs at local institutions challenge these core consortial values. Libraries generally cannot loan ebooks to another library, and this is especially true if they are titles made available but not yet purchased through a local DDA program. DDA, by its very nature of purchasing content only at the point of need, is at odds with group licensing, which generally is based on bulk purchase of titles. In this environment, the Colorado Alliance decided to undertake a consortial DDA pilot beginning in May 2012.
The Colorado Alliance DDA program has been a success by some measures. The group has purchased 280 titles from ebrary and EBL that are now perpetually available to the nine participating members. Unlike some early DDA programs, including the 1999–2005 Colorado Alliance netLibrary program, the current Colorado Alliance DDA program has stayed well under budget, and analysis of usage data has allowed a reasonable reallocation of funds. Usage overall is quite strong, with over 3,400 titles with at least one paid use, and many of which are used often. By some measures, the spread of usage across institutions is wide enough that there is clear benefit to doing this program at a consortial level.
By some other measures, the success of the program is still being evaluated. Perhaps most importantly, some Colorado Alliance members did not participate, meaning that only a subset of the consortium has access to the books in the program. Also many non-Colorado Alliance members that use the Prospector union catalog still are unable to use ebooks that they discover through this shared resource. Finally, by some measures, usage of most titles is too localized to warrant a consortial DDA program; in the first two years, it would have been cheaper for each library to go it alone.
While the overall success of the program is still debatable, the Colorado Alliance decided to continue it into a third year. The strengths of the program seem to outweigh the weaknesses, and the potential benefits to the community, and even to other consortia, are great enough to warrant continuation. The third year of the program saw the introduction of several new publishers on the ebrary platform, will see the impact of STL rate increases, and will see the merger of ebrary and EBL content onto the single ProQuest Ebook Central platform now that both programs are managed by ProQuest.
Notes
1. Kizer Walker, “Patron-Driven Acquisition in U.S. Academic Research Libraries: at the Tipping Point in 2011?,” Bibliothek Forschung und Praxis 36, no. 1 (February 2012): 125–129.
2. William H. Walters, “E-book in Academic Libraries: Challenges for Acquisition and Collection Management,” Portal: Libraries and the Academy 13, no. 2 (April 2013): 187–211.
3. Judith Nixon, Robert S. Freeman, and Suzanne M. Ward, “Patron-Driven Acquisition: An Introduction and Literature Review,” Collection Management, 35, no. 3-4 (2010): 119–124.
4. Candice Dahl, “Primed for Patron-Driven Acquisition: A Look at the Big Picture,” Journal Of Electronic Resources Librarianship 24, no. 2 (April 2012): 119–126.
5. NISO DDA Working Group, “Demand-Driven Acquisition of Monographs: A Recommended Practice of the National Information Standards Organization.” Approved June 24, 2014. (Baltimore: National Information Standards Organization, 2014), www.niso.org/apps/group_public/download.php/13373/rp-20-2014_DDA.pdf.
6. Karen S. Fischer, Michael Wright, Kathleen Clatanoof, Hope Barton, and Edward Shreeves, “Give ’Em What They Want: A One-Year Study of Unmediated Patron-Driven Acquisition of e-books,” College and Research Libraries 73, no. 5 (September 2012): 469–492.
7. William Breitbach and Joy E. Lambert, “Patron-Driven Ebook Acquisition,” Computers in Libraries 31, no. 6 (July 2011): 17–20.
8. Dennis Dillon, “Texas Demand-Driven Acquisitions: Controlling Costs in a Large-Scale PDA Program,” in Patron-Driven Acquisitions: History and Best Practices, ed. David A. Swords (New York: De Gruyter, 2011): 157–167.
9. Michael Levine-Clark, “Building a Demand-Driven Collection: The University of Denver Experience,” in Patron-Driven Acquisitions: History and Best Practices, ed. David A. Swords (New York: De Gruyter, 2011): 55.
10. Carol Joyner Cramer, “All About Demand-Driven Acquisition,” Serials Librarian 65, no. 1 (July 2013): 87–97.
11. Kay Downey, “Technical Services Workflow for Book Jobber-Mediated Demand Driven ebook Acquisitions,” Technical Services Quarterly 31, no. 1 (January 2, 2014): 1–12.
12. Kay Downey, Yin Zhang, Cristobal Urbano, and Tom Klinger, “A Comparative Study of Print Book and DDA Ebook Acquisition and Use,” Technical Services Quarterly 31, no. 2 (April 2014), 146.
13. Tonia Graves, Rob Tench, and Anne Elguindi, “Taking Advantage of Every Opportunity: Blending Local and Consortial DDA ebook Programs,” Against the Grain, 26, no. 2 (April 2014): 62–64.
14. Lynn Wiley and Elizabeth Clarage, “Building on Success: Evolving Local and Consortium Purchase-on-Demand Programs,” Interlending and Document Supply 40, no. 2 (May 2012): 105–110.
15. George Machovec, “Consortial ebook Licensing for Academic Libraries,” Journal of Library Administration 53, no. 5-6 (July 2013): 390–399.
16. Greg Doyle and Cory Tucker, “Patron-Driven Acquisition—Working Collaboratively in a Consortial Environment: An Interview with Greg Doyle,” Collaborative Librarianship 3, no. 4 (October 2011): 212–216.
17. Susan Hinken and Emily J. McElroy, “Chapter 2: Consortial Purchasing of E-books,” Library Technology Reports 47, no. 8 (November 2011): 8–13.
18. Jill Emery and Bonnie Parks, “The Demand-Driven Acquisitions Pilot Project by the Orbis Cascade Alliance: An Interview with Members of the Demand-Driven Acquisitions Implementation Team,” Serials Review 38, no. 2 (June 2012): 132–136.
19. Ibid., 134.
20. Tony Horava, “Today and in Perpetuity: A Canadian Consortial Strategy for Owning and Hosting Ebooks,” Journal of Academic Librarianship 39, no. 5 (September 2013): 423–428.
21. Kate Davis, Lei Jin, Colleen Neely, and Harriet Rykse, “16 Schools, $150,000 and 9 days: Experimenting with the Patron-Driven Acquisition Model in a Consortial Environment,” presentation at the Electronic Resources and Libraries Conference, Austin, Texas, February 28, 2011. www.electroniclibrarian.com/past-conferences/2011-conference.
22. Kate Davis, Lei Jin, Colleen Neely, and Harriet Rykse, “Shared Patron-Driven Acquisition within a Consortium: The OCUL PDA Pilot,” Serials Review 38, no. 3 (September 2012): 183–187.
23. Lorraine Huddy, “Striving for Insights and Contending with Limitations: The Assessment of a Collaborative ebook Project,” Against The Grain 24, no. 4 (September 2012): 36–40.
24. Jodi Shepherd and Marc Langston, “Shared Patron-Driven Acquisition of E-books in the California State University Library Consortium,” Library Collections, Acquisitions, and Technical Services 37, no. 1-2 (March 2013): 41.
25. Allen McKiel and Jim Dooley, “Changing Library Operations—Consortial Demand-Driven ebooks at the University of California,” Against the Grain 26, no. 3 (June 2014): 59–61.
26. Steven Carrico, Trey Shelton, and Roy Ziegler, “The FSU-UF Patron-Driven Acquisitions Plan: A Cutting-Edge Collaboration,” Florida Libraries 56, no. 1 (Spring 2013): 18–23.
27. Jeanne Richardson, “The Arizona Universities Library Consortium Patron-Driven e-Book Model,” Insights: The UKSG Journal 26, no. 4 (March 2013): 66–69.
28. George Machovec, “netLibrary Revisited,” The Charleston Advisor 4, no. 4 (April 2003): 21–26.
29. Kari Paulson, “The Story of Patron-Driven Acquisition,” in Patron-Driven Acquisitions: History and Best Practices, ed. David A. Swords (New York: De Gruyter, 2011): 63–78.
30. Hinken and McElroy, “Chapter 2: Consortial Purchasing of E-books.”
31. Doug Way and Julie Garrison, “Financial Implications of Demand-Driven Acquisitions: A Case Study of the Value of Short-Term Loans,” in Patron-Driven Acquisitions: History and Best Practices, ed. David A. Swords (New York: De Gruyter, 2011): 137–156.
32. Avi Wolfram-Arent, “College Libraries Push Back as Publishers Raise Some E-book Prices,” Chronicle of Higher Education (June 16, 2014), http://0-chronicle.com.bianca.penlib.du.edu/article/College-Libraries-Push-Back-as/147085/.