Securing and implementing a transformative agreement: Perspectives from library leaders

Preparing the groundwork — getting ready for discussions

Identifying priorities

In 2019, Carnegie Mellon University (CMU) opens in new tab/window in the US decided to pursue a strategy of securing open access agreements where they provided additional value to the university. That same year, it signed its first with a publisher: Elsevier. Keith Webster opens in new tab/window, Dean of University Libraries at CMU, was closely involved in developing that deal. And prior to discussions opening, one of his first steps was to map out CMU’s options and goals.

According to Keith, this process proved invaluable in helping him determine how to balance and prioritize budgets for reading and publishing, and to maximize readership access to content. “While in an ideal world, you would get everything you want across all three, in reality that’s generally not the case,” he explains. “For example, if you have a limited budget, then you are probably not going to have readership or open access publishing rights to everything that you want. That’s why it’s important to be clear on what you want to achieve.”

Keith believes his preparation played a key role in helping CMU emerge from discussions feeling that they had secured value with an agreement they were happy to move forward with. 

Bob Gerrity opens in new tab/window, former Program Director of the Procurement Program for the Council of Australian University Librarians (CAUL) opens in new tab/window, recalls undertaking a similar analysis exercise during early discussions with Elsevier in 2022. Bob and his colleagues were developing an opt-in open access agreement for CAUL’s 47 member institutions in Australia and New Zealand. Although they were initially aiming for an agreement with an uncapped number of article publishing charges (APCs), after weighing the three factors, CAUL settled on a progressive open access transition throughout the agreement, where the percentage and number of articles published open access rises each year of the agreement.

Analyzing the data

While Keith hesitates to describe it as a “pain point,” he does remember some difficulties around sourcing the data required for his preparation. “I wanted to understand how Elsevier would try to value the agreement financially, and part of my modeling was to look at how many articles we were likely to publish under a multi-year deal. The problem was, I honestly had no clue,” he admits. While the data sources at his disposal told him how many articles CMU-affiliated authors had published with Elsevier in the past, they could not identify how many had a CMU-affiliated researcher as the corresponding author. This was important because the agreement covers the open access publishing costs for the latter. Keith says, “It’s an area where we would have benefited from more insight” as we continue to refine and improve the processes for agreements.

Data also proved a challenging topic for CAUL, says Bob. The three-year deal he was discussing with Elsevier included an annual increase in the number of articles covered. To determine those yearly caps, projections were made based on historical publishing rates. But Bob reveals: “Once the agreement was in place, we saw lower publication levels than were predicted, including a lower percentage of eligible authors choosing the open access option than expected.” It turned out the levels were overestimated due to higher publishing activity caused by COVID, uncertainty about publishing trends as the pandemic continued, and the open access coverage of the agreement needed to be communicated more clearly to eligible authors.

Elsevier has been exploring the best way to deliver accurate and reliable data to aid open access agreements — particularly in terms of identifying corresponding authors affiliated with a university and predicting future output.

In the case of Tulane University opens in new tab/window in the US, the historical publishing data supplied by Elsevier during 2022 discussions proved very similar to its own, says Dr Andy Corrigan opens in new tab/window, Associate Dean of Libraries. “In fact, I think it worked better with Elsevier than it did with some of our other agreements,” he reveals. According to Andy, Elsevier also supplied detailed data on historical corresponding author numbers. “We were really surprised to learn just how many Tulane authors were publishing in Elsevier titles — the number was pretty large.”

Like CAUL, the terms of the agreement that Andy and his colleagues were discussing with Elsevier included a yearly increase in the number of articles that can be published under the agreement. Accurate data played a vital role in setting the article amount. “There’s an initial period in these kinds of deals where it’s a little risky — you want open access publishing to catch on with your authors but not to the point where you exceed your cap,” Andy says. “Turning people away is the last thing you want to do. If we’d got the numbers wrong, we would probably have had to subsidize those APCs.”

While access to data proved relatively simple for Andy, achieving alignment on the terms of the agreement took more time: “The deal with Elsevier was one of our first, so it was never going to be an easy thing. I think what got us over the line was perseverance on both ends. We didn't really give up, which would have been easy a couple of times. We stuck with it, and so did our counterparts.”

With preparation done and discussions underway, our interviewees moved on to resolving the final details of the agreements. In the case of Carnegie Mellon University, Keith recalls two in particular:

• Author rights: A potential challenge arose in the form of CMU’s federally funded research and development center: the Software Engineering Institute opens in new tab/window. Keith explains: “There were questions from their staff about their eligibility under the agreement. Now, at a simplistic level it was clear cut — they are Carnegie Mellon employees, and they were, in Elsevier’s eyes, covered. The sticking point was around their authors’ ability to transfer rights, as most of the institute’s work remains the intellectual property of CMU with the SEI’s government customer granted certain rights pursuant to its federal contract. We made it work, but it took discussions between the university, the institute, the Department of Defense and Elsevier to find a solution.”

• Start date: Although it may seem a small detail, according to Keith, it’s “been a topic of discussion in all our agreements.” He explains: “If an agreement goes live on 1st January, the question is whether articles that are accepted after that date are eligible for inclusion, or only articles submitted after that date. Elsevier’s position was accepted, which is the neater one.” The reason for this is that the time between acceptance and publication is short, so using acceptance date eligibility aligns better with annual published output.

Preparing to roll out the agreement

The driving principles

For all three of the library leaders we spoke with, securing open access agreements for their institutions not only required new ways of thinking but new workflows and processes for faculty and staff.

In the case of CAUL, the deal with Elsevier is one of more than 20 the organization has negotiated with publishers and vendors on behalf of its members. And for Bob, one of the most important principles when it comes to rolling them out is to “keep it simple”:

We are a small organization, so we want as little administrative overhead as possible.

Simplicity is a sentiment echoed by Andy at Tulane University, whose institutional deal with Elsevier came into effect on January 1, 2023. But in Tulane’s case, it’s all about reducing the burden on researchers: “We want to make it as seamless as possible for the author.”

Administering author approvals

The author approval process is typically one of the first new workflows institutions introduce after a deal has been signed. While publishers enable authors to indicate their eligibility for an open access agreement during the manuscript submission process, if the article is accepted, the author’s institution is asked to confirm that eligibility.

CAUL manages the process centrally on behalf of its member institutions. “Then, if there is any question over eligibility, we send it on to the author’s institution to confirm that person is affiliated, but it’s pretty lightweight in terms of time commitment,” Bob reveals.

This solution is reflective of CAUL’s cooperative approach to open access agreements. For example, there is no allocation of articles to individual institutions: All members share the allowance and institutions can monitor progress on CAUL’s Read and Publish Agreement LibGuide.

At CMU, with transformative agreements now in place with six major publishers, “approvals have become a noticeable call on staff member time,” Keith admits. “That’s something we hadn’t fully appreciated, although it’s an administrative exercise we can currently absorb.”

For Andy and his colleagues, the “back-end part” has proved easy to implement, and he’s satisfied they’ve achieved a fairly seamless process for Tulane’s faculty. But he notes: “I’m sure there are some who don’t realize that the library has been involved in working this out for them. I’m OK with that, but I would maybe like a bit more visibility for the library in this. I guess as long as the people who fund the deals know, that’s what’s important.”

Educating and engaging authors

While publishers like Elsevier typically provide agreement pages and journal finders for researchers to understand agreement parameters and eligibility criteria, additional LibGuides form the backbone of all three organizations’ communications to researchers on their open access deals.

At CAUL, each time the organization is close to concluding a new transformative agreement, CAUL staff develop a briefing document for members. This includes adding details of the agreement to CAUL’s dedicated LibGuide. Bob says: “We post basic materials supplied by the publishers: usually information about the deal and what the author will see. Our member institution libraries then typically reformat and rebrand that information for their own LibGuides.” He adds: “We also ask publishers to run a session on the author interface for our content coordinators (CAUL’s liaisons at its member institutions).”

At CMU, in the case of the agreement with Elsevier, the LibGuide was compiled using screenshots of the publisher’s submission system. “They were helpful to show authors, ‘You are entering this workflow, here is what it will look like,’” says Keith. In addition, CMU’s Library provided its staff with a slide deck explaining the terms of the deal, along with a series of talking points to share with faculty.

But with the number of transformational agreements between CMU and publishers increasing, Keith has bumped into an issue with variations in publication workflows. “I want the information we provide to researchers to be as publisher agnostic as possible,” he says. “However, we’ve found that every publisher does things differently. I spoke about this at Charleston Conference opens in new tab/window in 2022: What can we do to standardize the author experience so researchers don’t have to remember different steps and license options? It’s something I’d like to see change.”

It’s a view that resonates with Andy at Tulane: “Elsevier has its own systems and other publishers do too, so that’s been a little tricky. It’s become part of the territory, but in an ideal world, our folks wouldn’t have to follow different processes to manage this.”

The interviewees

Andy Corrigan

Associate Dean of Libraries at Tulane University, USA 

Bob Gerrity 

University Librarian at Monash University, Australia, and former Program Director of the Procurement Program for CAUL (Council of Australian University Librarians)

Keith Webster 

Helen and Henry Posner, Jr. Dean of the University Libraries at Carnegie Mellon University, USA