Strategic Combination

Strategic Combination

1

Solving for X

2

As we bear witness to the decline and fall of Twitter (we at The Brief are still struggling to call it “X”) in real time, questions were raised this month about how useful it really ever was, at least in terms of driving traffic. NPR (National Public Radio), with some 8.7 million followers, left the platform and six months later reports the impacts as negligible. Closer to home, Angela Cochran of the American Society of Clinical Oncology wrote about her disappointment in the returns from ASCO’s social media marketing efforts.
 
But perhaps the picture here isn’t quite so simple.
 
First, given the latest developments, determining whether your community still primarily resides on “X” and, if not, identifying its current whereabouts, poses a challenge. Recent surveys suggest that Mastodon, LinkedIn, and Instagram/Threads are attracting some of the previously dispersed science traffic, although it’s still early in the post-Twitter era (we’ve been enjoying Bluesky ourselves). We don’t have a crystal ball, but it appears that the landscape may become increasingly fragmented (at least for the next 12 months). Whether the scientific and scholarly community will coalesce around a primary platform remains unknowable.
 
From our vantage point, average social referrals for academic and professional journals are in the 1%–6% range (meaning that 1%–6% of referral traffic to a given journal typically derives from social media platforms in the aggregate). These percentages vary based on community dynamics and the effectiveness of the social strategy. Interestingly, even with the notable dispersion of the science community from Twitter, these referral rates have not undergone significant changes.
 
It’s also crucial not to overlook the realm of “dark social,” which includes direct messages, message groups, emails, Slack or Teams channels, or comments within communities. Like dark matter, dark social is difficult to measure but still significant. Indeed, the vast majority of the conversation triggered by social posts – typically exceeding 80% – occurs off the public faces of social platforms. Social media may influence direct traffic (or what is currently categorized as “direct”) much more than previously realized. This complexity poses challenges in tracking social media through traditional methods.
 
The best campaigns are multichannel, and social is an important channel in the campaign mix. In the realm of consumer media, for instance, monitoring social referrals and social campaign engagement signals a higher likelihood of subscriptions. Research by Piano showed that “the more channels a visitor is referred from, the higher the conversion rate. In fact, a user who comes from both social and search is likely to have a conversion rate 10X higher than one who comes through just one channel. The more channels they use, the more loyal they become.” Social media analytics, and audience insights, have become much more sophisticated and go beyond just engagement and measuring source-by-source referral traffic.
 
A proficient social media strategy must therefore be highly nuanced, involving various strategies and metrics tailored to specific objectives. We have seen the rise of OA and the desire to attract authors changing the way many organizations think about their social media strategy. When the goal is readership, social will always be a small percentage of traffic referrals due to the strong discoverability channels available to researchers. However, when the goal is building community and brand affinity with current and prospective authors, social is a much more important lever, as it is a great way for journal brands and their editors to engage with their audience.
 
The art of social media is knowing what you are trying to accomplish with it. Once you know this, you can determine what you might automate and when – and where – to invest in more time-consuming community engagement. Investing in the right technology is critical to reduce effort (and cost) and improve effectiveness and will become more so as the landscape fragments. Emerging AI tools are poised to make navigating (and automating) aspects of social strategy even more efficient, although an authentic, human voice will remain essential for community engagement.

People

3

Copyright Clearance Center (CCC) has named Duncan Campbell as the new Executive Director of its Client Engagement and Business Development team.
 
Tracie Hall has resigned from her role as the Executive Director of the American Library Association. The Association will name an interim director as it prepares to search for her successor.
 
The American Physical Society announced that Dylan Moulton has joined the society as Global Sales Director of Publications. (C&E was thrilled to have supported APS in the recruitment process.)
 
Perhaps a casualty of the ongoing Hindawi fallout, or maybe an indication of a shift in overall strategy, Wiley has parted ways with CEO and President Brian Napack. Former Group Executive and Board Chair Matthew Kissner has been named as Interim CEO. 
 
Wellcome has appointed John-Arne Røttingen as its new CEO.
 
Steve Sherry is now the Acting Director of the National Library of Medicine.
 
Rafael Sidi has been appointed by Wolters Kluwer as Senior Vice President & General Manager of the Health Research segment.

Briefly Noted

4

Just as The Brief was going to press, cOAlition S released a Halloween treat – a proposal titled “Towards Responsible Publishing” that outlines a vision for a “community-based scholarly communication system.” We will put to the side all of the community-based scholarly journals from independent societies driven to seek deals with commercial publisher over the last 5 years due to Plan S, and read this carefully before commenting further in the next issue of The Brief.
 
One doesn’t often see a press release announcing the intention to sign an agreement (rather than an actual signed agreement), but Wiley has declared that they intend to do so with Projekt DEAL. No details are offered in the press release about the terms of the agreement (so we can’t draw any conclusions about how those terms will compare to Elsevier’s new arrangement with Projekt DEAL).
 
While we at The Brief remain proponents of building sustainable-in-the-long-term business models around essential community infrastructure, we were still pleased to see arXiv receive some US$10 million in grant funding for much-needed technology upgrades.
 
EBSCO’s annual Serials Price Projection Report reflects the growing decline of (and increasing costs associated with) print journals, and suggests an overall price increase for 2024 to be in the range of 3%–4% (subject to impacts from currency exchange rates).
 
eLife, no stranger to controversy, was back again in the news for firing Editor-in-Chief Mike Eisen. The move was precipitated by Eisen’s reposting of a satirical article in the Onion addressing current events in Gaza. The eLife Board has issued a statement on their reasoning. Lost in the chaos of the Eisen kerfuffle, eLife celebrated its 10th anniversary this month, and issued its annual report and financial statements.
 
Happy birthday wishes also go out to The Lancet, which, for a 200-year-old publication, seems rather spry and lively! The journal offered a celebratory issue looking back on its history and forward to where it intends to evolve.
 
Richard Sever offers a tour de force overview of the history of science publishing, just what it is that journals do, and the costs that are involved in his PLOS Biology article “Biomedical publishing: Past historic, present continuous, future conditional.” Along with this essential knowledge comes Sever’s vision of a web-based future of science publishing, which begins with preprints (it is perhaps not a coincidence that Sever is a cofounder of both bioRxiv and medRxiv). The atomization of dissemination and the ongoing evaluation of the literature has many appealing merits, though we wonder if it would do away with the efficiencies of the current system. Would submitting research to a wide variety of badging/evaluation systems, for example, require more time and effort than sending it to one journal? Ultimately, we suspect that organizations would bring all of those disparate services under one roof and create a “one-stop-shop,” thus largely reinventing the journal. Sever’s article calls to mind a quip attributed to Jim Barksdale (the former CEO of Netscape) that “there are only two ways to make money in business: One is to bundle; the other is unbundle.” (Barksdale clearly would have done well as a scientific and scholarly publisher). Our only substantive quibble with this paper is that we could not find a preprint version on bio/medRxiv!
 
Another important acquisition announced this month was the purchase of the Charleston Hub, Against the Grain, and the Charleston (Library) Conference by Annual Reviews. While this was not surprising given the previous acquisition of purchase of The Charleston Advisor by Annual Reviews, it marks the end of an era (and beginning of a new one) for what has become among the most important gatherings of librarians and publishers under one roof. 
 
The big question around the University of California (UC) library’s “multi-payer model” for OA deals is whether campus authors will agree to be one of the “multi-payers” and put some of their research funding in with the library’s contribution to cover article processing charge (APC) costs. If the results of UC’s agreement with IEEE (Institute of Electrical and Electronics Engineers) are any indication, then the prospects look poor. UC is ending their OA deal with IEEE and reverting to a “read-only” agreement, as only 20% of UC authors publishing in IEEE journals chose to publish their papers OA, and fewer contributed grant funds toward the APC than was expected. Meanwhile, the Association of Swedish Higher Education Institutions has declared that Sweden will no longer be willing to sign transformative agreements for hybrid journals and will only enter into agreements for fully OA journals. While “increasing costs” are given as the primary rationale for this decision, no information is offered as to how fully OA journals will reduce those costs.
 
The more things change, the more they stay the same. Back in 2011, Phil Davis’s landmark randomized controlled trial looking at the effects of making papers OA concluded that OA greatly increased readership of papers, but had no impact on citations, generally because the majority of readers at research institutions who were likely to do the next experiments and cite those papers already had access to them. But OA was declared to have worked well because it extended access beyond research institutions to the general public. Cut to 2023 where a group of researchers looked at what happened to US Department of Energy National Laboratory research after the 2013 Holdren Memo went into effect, requiring public access to funded papers within 12 months of their publication. Unsurprisingly, the freely available papers were not cited any more than subscription articles, but they do turn up significantly more (42%!) in patents, particularly patents from smaller companies. So, once again, the message is that OA really works, but less so for academia than for commercial development of research. Economic development is a key reason why governments fund research, so this offers clear evidence of OA ROI (return on investment).
 
A new preprint from Hanson et al. offers some tremendously useful data on the current state of scholarly publishing. Expect to see Figure 2, detailing “the rise of the special issue model of publishing” featured prominently at every meeting you attend over the next few years. One quibble – in their analysis, it is difficult to discern what constitutes “growth” versus what is actually consolidation (which continues to occur at a rapid pace). The “growth” (publishing more papers) shown for many publishers in the study is not due to the launch of new publications or growth in the size of existing publications, but rather the moving around of publications between publishers and from independently published to partner-published journals. This is mitigated somewhat by the authors’ work measuring articles per journal, but remains a confounding factor. The elephant in the room of the article though, is that while much of what the data shows describes the impact of activities from MDPI, Frontiers, and Hindawi, the authors attribute the results to the entire publishing industry, rather than calling out the clear driving forces behind the problems they note.
 
In a recent issue of Nature, researcher Dritjon Gruda suggests that publishers start allowing authors to submit the same manuscript to multiple journals at the same time. The reasoning here seems flawed: the problem Gruda hopes to solve – how to get research results out faster – can largely be solved by posting one’s work as a preprint, rather than asking an already strained peer review system to massively increase its workload. The industry seems to have responded in the negative to this request, if only coincidentally, through the release of a new duplicate submissions detector tool from the STM Integrity Hub, primarily aimed at detecting the activity of paper mills.
 
This month’s AI food for thought includes a fascinating article in The New Yorker that looks at what happens to AI after it has ingested all the material out there and largely replaces human-generated writing. Where will it then find data for further training? One suggestion is “synthetic data,” which the AI systems will generate themselves, but many are skeptical that this is a real thing, given that, “no matter how smart you are, you can’t learn new facts about the world by reviewing what you already know.” On a more positive note, Business Insider looks at the fears about AI-driven job loss through similar concerns around the launch of Excel and ATMs. The short answer is that while both made some jobs go away, in the end each created more jobs than were lost.
 
Here’s one we’ve never seen before – a preprint exposes the latest method for manipulating journal article citationsthrough adding a long list of fake citations to a paper’s metadata at the point where Crossref mints its DOI. While we at The Brief do not endorse this unethical behavior, we do at least have to give some respect to the hustle and innovation shown here.

***
I have certainly not always responded in an ideal manner. —Michael Eisen