Based on her presentation at ALPSP’s ‘Beyond the rhetoric: New opportunities in open access’ seminar, Kathryn Spiller of BioScientifica Publishing reports on recent experiences of launching an open-access journal
I took over as head of publishing at BioScientifica Publishing in December 2011 on the back of a growth agenda. The starting point for this growth was fairly small: our journal portfolio in 2011 included five titles, three of our owner, Society for Endocrinology, and two for partner societies.
All our titles operated under similar subscription models. In 2011 they all offered an open-access (OA) option – with article processing charges (APCs) of £2000, and half-price APCs if the corresponding author’s institution subscribes to the title. In addition, all our content is made free after 12 months and only the accepted, unedited version can be deposited in repositories at this point if no OA fee has been paid. Our OA options had low take up; less than 10 per cent of content across our portfolio was made OA.
However, during 2011, the Society for Endocrinology and the European Society for Endocrinology strongly felt that they needed to offer more than ‘hybrid’ OA, especially in view of trends within the industry and from policy makers. BioScientifica brought the societies together on a joint project to identify opportunities for a new journal.
The project taught us much about launching a new OA journal. Firstly, there are many ways that it is like launching any new journal, including subscription titles. There is still a need for comprehensive market research of the target community, identifying a ‘real’ gap in the market and identifying what will be meaningful to the target community. In addition, any new journal launch will involve significant investment and risk, and the journal will start its life with no impact factor or indexing.
However, there are many ways that new OA journals differ from subscription launches too. Perhaps most crucially, researchers now take on a new role in the process. They are no longer just authors and readers but also paying customers. It is very important to assess their willingness and ability to pay and to consider the strengths and weaknesses that the OA model brings.
With all this in mind, we carried out market research, working with an external marketing consultancy, to identify a gap and define the scope of a new journal and assess the value that our brands (the two societies and publisher) might add. We also wanted to pick a name and understand attitudes to key terms, as well as assess the factors that are most important to our community in deciding where to submit.
Other market research objectives included: understanding awareness of and attitudes to OA in our community; understanding the current costs of publishing paid by authors in our community; and understanding the level of APC that would be acceptable to our authors and how it would be funded.
Our survey was delivered to just under 50,000, including members of both societies (circa 4000), of which 904 (1.8 per cent) took part. The majority of these respondents were from Europe and North America. Most were based at a university or hospital, with a good spread of clinicians and scientists working in different specialisms related to endocrinology.
Through this research, we identified a gap in the market for a broad-scope journal linking endocrinology with its intersecting disciplines. We found that the two society brands had positive associations for more than half of respondents, even though the majority weren’t members, but our publisher brand was not well known. In addition, it was interesting to find that the word ‘endocrinology’ had very positive associations for most respondents whereas the word ‘hormone’ had more neutral or negative associations.
The research also revealed author preferences. Key factors influencing respondents’ choice of where to publish were, firstly, the reputation of the journal; followed by the impact factor; then the relevance of the publication to your work; speed of publication; free access to your work; reputation of associated society; reputation of publisher; and funding body.
These findings differ from those from research published by Wiley at around the same time, which found that relevance/scope was the most important with impact factor second and reputation of associated society third. This shows the importance of finding out what matters to your own community.
Some differences were also revealed in experiences of OA. According to our research, 78 per cent had some or good awareness of OA as a new model for publishing research articles (awareness was lowest in Latin America and Asia), 44 per cent had published in an OA journal, 26 per cent of which had published frequently or occasionally. In contrast, the Wiley survey found that a third had published in an OA journal.
The difference between the two pieces of author research reflects how biomedicine is further along on the route to OA than many other disciplines, a situation that was echoed in the attitudes to OA that we found in our survey. Of those with experience of publishing in OA, 47 per cent said that they strongly support the principle but have trouble paying fees and 39 per cent said that they support it and are prepared to pay reasonable fees. PLoS ONE was the most popular OA journal amongst respondents.
Cost of publishing
We also found that 72 per cent had experience of paying page charges. Almost half of these said the average cost of page charges per paper is between $101 and $500. Over 50 per cent of these authors fund charges through grant money and just over 30 per cent fund them personally. The range of journals where people have paid to publish OA is large, as is the price range.
When asked what a reasonable APC would be, 20 per cent said it is not acceptable to charge; 41 per cent stated below $500; and only 11 per cent stated over $1000. There is a discrepancy between what people say they will pay and what they often are already paying; people are paying high page charges even if they think they are unreasonably high.
As a result of our research, Endocrine Connections was launched in June 2012. The name was carefully chosen to represent the different connections, between disciplines, between geographical regions and between basic and clinical research.We adopted an online-only model with the CC-BY licence. The full price for an APC is £700, with the first 200 articles published being half price. I think this is sustainable.
We cascade articles rejected by the societies’ high-impact journals. However, we have not marketed this as a cascade journal and are very careful to only cascade high-quality papers that are out of scope. These submissions accounted for 50 per cent of submissions consistently throughout the year. We are also marketing for direct submissions.
We had a very short timeline of six months between internal sign off of the idea to having content live online. The response to the first announcement resulted in more than 700 signing up for more information and 3490 have signed up for alerts since launch. We received 58 submissions in the first six months and published two issues of seven articles, with five more articles online, in the first six months. We don’t have many usage figures yet as the journal has only been going a few months. For me, the number who’ve signed up to alerts has been the really surprising figure but the full-text downloads have been high too.
Our average receipt-to-accept time is 27 days. One of the key things we’ve done in launching this journal, because we believe speed is so important, is to look at the peer-review process. We don’t have a major revision category on the journal. We thought it was important not to have papers going back for six months for major revisions.
Another change that was required in launching this journal was the introduction of e-commerce for payments. An automated payment system is important because it would otherwise be a big administrative burden. We have had some issues with transfer of papers from other journals into Endocrine Connections , in particular in relation to the automation of the publishing process, so we’re looking at ways in which this can be improved. In addition to giving us our first journal launch since 1988, this launch experience has taught us many lessons. An important one is to not be afraid to take risks and invest in new products; not doing anything is potentially a bigger risk.
However, things have not all gone quite as we planned. In terms of the cascade model, only 22 per cent of those who were offered to be cascaded accepted this option and submitted to the journal; our initial research suggested that would be a lot higher (60 per cent) but the biggest barrier today is that the journal is not yet indexed in PubMed. Once we are indexed in PubMed and ISI that should push up the submissions. We are considering including article-level metrics on our site.
One of the biggest surprises to us was the acceptance rate, which is pretty much the same as our other journals (around 25 per cent). We had budgeted for an acceptance that was much higher (70 per cent), thinking that there is already a quality check from other editors cascading articles. However, those editors are also on the board of the new journal and have been much more stringent than we expected with the papers that they have sent over.
I want to finish on a piece of advice given to me that makes a great deal of sense: don’t name a journal after its business model. We don’t know what’s going to happen in the future and what’s going to change, so flexibility in approach is key .