by James Bunnelle (Acquisitions & Collection Development Librarian, Watzek Library, Lewis & Clark College) <[email protected]>
Editor’s Note: This is a follow-up to McElroy & Hinken’s “Pioneering Partnerships: Building a Demand-Driven Consortium eBook Collection,” published in the June 2011 issue of ATG. Readers are advised to consult that piece for information pertaining to the formative stages of the pilot. — JM
In July of 2011, the Orbis-Cascade Alliance (henceforth the Alliance) launched its pilot project for demand-driven acquisition of e-books at the consortium level, the culmination of nearly two years of planning. The Alliance is comprised of 37 member institutions; 36 in Washington/Oregon, with the University of Idaho joining post-launch. At the end of 2009, the Alliance’s Council of deans and directors created an e-book team and charged that body with the following:
• Leverage the existing relationship with YBP to create an entirely new e-book consortial purchasing model that allows consortium-wide access to titles purchased by individual member libraries.
• Focus on developing and implementing the new model and on addressing access, collection development, financial, and technical issues outlined in the first e-book team’s report… Work with the Collaborative Technical Services Team charged with developing technical services operations that support collaborative cataloging/processing for e-book collections.
• Develop a funding model to support the program in an equitable manner.
• Develop a model that prioritizes selection in a way that benefits the most members possible.
• Evaluate the project to determine ongoing viability
• It is broadly understood that Alliance-wide access to e-books purchased through this program will require full participation, including financial support, by all Alliance libraries. We expect that the membership’s shared commitment to collaborative strengthening of the Alliance collection will enable the Team to craft a program all members can support.
As the last point states, it was decided from the outset that if the program was to be successful, it would not be an opt-out model and would require mandatory contributions from all (then) 36 Alliance libraries. This mirrors past and ongoing efforts of the Alliance’s Collection Development and Management Committee (CDMC), the pilot’s umbrella organization, which has focused on cooperative collection building, particularly maximizing existing resources and avoiding unnecessary duplication. Indeed, data collected for several recent CDMC initiatives informed our early decisions; first and foremost, it helped us establish the multiplier, to be discussed shortly.
The funding model for the pilot was done on a tiered FTE scale not unlike that used to calculate our consortial electronic resources. Rather than being a sustainable model for the long-term, it was a comfortable system with which all in the Alliance had some familiarity, and the new team assembled to oversee the pilot, the Demand-Driven Acquisitions Pilot Implementation Team (DDAPIT), felt it would allow us to move forward without getting bogged down in debates on alternative formulas. In the end, all 36 institutions pooled a total of $231,000 in what was slotted to be a six-month pilot. Libraries submitted their payments into a centralized Alliance fund, with all short-term loans and multiplied purchases generated by demand-driven usage charged against this account. This allowed for easy centralized tracking of data by the DDAPIT and alleviated the need for localized bookkeeping practices within the various acquisitions units.
Building the Profile
For the initial retrospective record load of 1,700 titles, and for the ongoing updates of new releases, the team constructed a profile whose broad subject content reflected the diversity of the consortium members. In the end, very few LC ranges were excluded, with content ranging from Basic through Professional, and encompassing 2011 imprints. Caps were put on cost, but the team decided not to dedupe for any e-books purchased by individual member libraries, under the reasoning that they could not be shared and therefore undermined cooperative collection development. EBL did rough calculations on how much our pool of funds would last, which is where we arrived at the 1,700 number for the back load. Admittedly, these were educated data-driven guesses stemming from situations quite different from our own, since this had never been attempted before. The team developed several contingency plans, should things move too quickly.
Partnerships and the Multiplier
With the funding and profile finalized, several challenges confronted us immediately. Chief among these was engaging in ongoing conversations with publishers and requesting their participation. Our close working relationship with EBL and YBP was vital to success in this area, and both worked very hard to build a pool of publishers for the pilot that could meet the diverse and demanding needs of the Alliance membership, which runs the gamut from community colleges to ARLs. That being said, it proved challenging; after all, part of the impetus of the pilot was a general dissatisfaction with the high-priced “big deal” e-book packages being offered by some of the very publishers with which we were initiating discussions. Although many publishers were participating in DDA acquisitions at the local level, the consortial model was an entirely different (and untested) affair. Furthermore, the high multiplier being suggested by some publishers showed that they held inflated expectations of the consortium’s holdings for publications on their title lists. This was clear from analyzing the data gathered for our 2011 Threshold Pilot, which aimed to minimize consortial duplication by setting a maximum number of print copies for titles obtained through YBP. Data culled and analyzed for this project, with the help of YBP, showed that for 2009 there were, on average, no more than four print copies of any given title purchased within the consortium (note that this did not include purchases from other vendors, since the Alliance had already moved to YBP as our primary book vendor in 2008.) The team used this as a basis for its multiplier, increasing this slightly to five, in part to compensate for the above caveat. Thus, having recent and reliable data on duplication conveyed the reality of the Alliance’s collecting patterns and helped immeasurably when negotiating what we considered to be a fair and equitable multiplier. More broadly speaking, it took publishers’ trust in our collective abilities, and a certain degree of faith, that this new model would operate and succeed at the consortium level. In the end, the DDAPIT was excited about the pool of participants moving forward to the launch; these included: ABC-CLIO, Ashgate, BRILL, Cambridge UP, Earthscan, Hodder Education, John Wiley & Sons, Oxford UP, Pharmaceutical Press, Sage, Taylor & Francis, and The Policy Press. The Alliance is indebted to our vendor partners on the DDAPIT team who worked on these and other efforts, including Robin Champieux, Sadie Williams, and Alison Bobal from EBL, and Joan Thompson, Barbara Kawecki, and John Elliott from YBP.
Short-Term Loan Threshold
Another difficult and imprecise task was deciding upon the short-term loan threshold. For those unfamiliar with this process, when a user exceeds a five-minute browsing period while accessing an EBL e-book online, they are asked if they would like to initiate a loan (also occurs for copying, printing, or downloading any portion). At that point, a short-term loan (STL) is generated and charged to your account, which is a small percentage of the book’s retail price. Once “X” number of STLs is reached, the title is purchased for permanent access. The DDAPIT already knew from the Council’s initial charge that ownership was important, and prior e-book teams had operated under that assumption during the vendor review process. But there was a second important factor to balance this against, namely, the rate of spending and the assurance that funds would last long enough for a) all contributing members to get some amount of usage, and b) for the team to gather data spanning both peak and slow periods of the school year; this had been the rationale for running the pilot from July to December. Ultimately, the team decided to go with an STL threshold of ten, with the understanding that adjustments might be necessary once concrete patterns of patron usage emerge.
Discovery and Loading
As team chair Emily McElroy and member Susan Hinken mentioned in their June 2011 piece, the DDAPIT and portions of the Collaborative Technical Services Team (CTST) worked closely during the past year. As launch approached, discovery options and record-loading replaced profile building and publisher involvement as the central concern of both teams. The DDAPIT quickly realized that we had sorely underestimated the time it would take to sort through the myriad of technical issues that would arise, and the CTST’s expertise in this area was instrumental in moving forward. Alliance members had three ways to achieve discovery, each based on existing practices: WorldCat Local, typically in conjunction with OCLC’s knowledgebase; MARC records for Innovative’s Millennium; or using Summit, our consortial catalog via the Navigator platform. In consultation with all areas of the library, particularly reference and public services folks, each institution selected one and reported this back to the DDAPIT, which then determined the type of record-loading workflow needed. These workflows and their implications for user discovery were covered extensively in the training materials by CTST members.
The Training, Evaluation, and Exit Strategy Teams
Three subteams were developed to handle areas identified as essential to the pilot, and all proved crucial at various points throughout the year. The Training Team would oversee the coordination of local and regional onsite training sessions, in addition to webinars and distributable materials. The Evaluation Team would identify measures of assessment and organize the collected data in such a way as to be easily consumable by the rest of the consortium. Finally, the Exit Strategy Team discussed possibilities of slowing spending if necessary, including the temporary suppression of some content, and how to execute this in a non-disruptive and centralized manner. For Training, two main sessions were held, one in Portland, the second in Seattle, and an online webinar tried to accommodate those who could not attend. YouTube desktop recordings were made for various sections, such as record loading and using EBL’s LibCentral to examine statistics. All were posted on a DDA dedicated Alliance website, along with an extensive FAQ. The Evaluation Team worked out calculations for all institutions ROI and offered them additional techniques for self-assessment if they chose to do so. Ultimately, the Exit Strategy Team, which was intended more as an emergency brake, became a more complex mechanism for tweaking STLs and purchases at key points of the fiscal year.
Changes Midstream: Lowering of STL Threshold and Extension
As Fall hit, it became clear to the team that the STL threshold of ten was too conservative for the amount of records included. A mass of STLs was moving forward, but most of these would not trigger by the pilot’s December deadline, and we would therefore fail to meet the Council’s charge of acquiring, not just accessing, content. After debate, it was decided to incorporate e-book titles from 2009 and 2010 that also fit our existing profile. We would also lower the STL threshold to five, or having 5x copies purchased as the “sixth STL.” Perhaps more importantly, in its November report to Council the team recommended that the pilot be extended to July 1, 2012, with the stipulation that additional funds may be required to close the gap between fiscal years. We emphasized that these funds would not exceed the amounts of the original contributions at the pilot’s launch. As a byproduct of the extension, we would also have the opportunity to analyze usage patterns for an entire calendar year. Council agreed on both recommendations, and funds were generated to match the initial pool.
From the outset, it was important for the team to communicate usage updates to all, so title-level reports were created by EBL using LibCentral and distributed by the team to Alliance member contacts. For the final report to Council in spring, the Evaluation Team completed its usage analysis thus far and wrote the following: “Across the Alliance, a diverse clientele is discovering and using a broad range of immediately accessible scholarly content. Every participating library has seen a positive return on investment, in many cases quite substantial. Through sharing of expertise, Alliance libraries representing the spectrum of experience with e-books have successfully collaborated to overcome considerable technical challenges.” Particularly high users were the community colleges, which the team tentatively interpreted as the unleashing of years of pent-up demand after continued and sustained cuts to materials budgets. The distribution otherwise showed predictable peaks at certain times of the year, with a slow start in the summer and spikes at the end of semesters, and to a lesser extent quarters.
New Funding Model for 2012-2013
With several months of data to analyze, it was clear that a new funding model would be needed as we moved forward with our recommendation to Council. While sufficient for the pilot, all agreed that member contributions needed to be increased considerably if we intended to expand, particularly since the limited publisher pool had been a common complaint throughout the year. The DDAPIT discussed the advantages and disadvantages of various formulas. Any model slanted towards usage would hit community colleges disproportionately, while materials budget and FTE models placed financial strains on the two ARLs, UW and UO. Abandoning a usage component completely in the spirit of shared collection building, the team agreed to recommend in its report to Council a budget distributed as follows: 30% Equal Split, 35% FTE, 35% Materials Budget. This was accompanied with two funding-level recommendations: one which would maintain the status quo ($515,150), and a second which would allow for expansion ($975,150). We submitted our final report to Council for review in advance of its March meeting.
Pilot to Program
The DDAPIT is excited to announce that Council has unanimously agreed to move forward with the DDA E-book Program for FY2013, at a funding level of $750,000 to be distributed according to the team’s recommended 30-35-35 formula. More importantly, it has reaffirmed its strong commitment to a shared e-book collection for the Orbis-Cascade Alliance and declared its intent to increase funding for FY2014 to $1,000,000. A reconstituted e-book team has been created to administer the program moving forward, as there are still many technical challenges that need to be addressed in the coming year.