This time, we discussed how we can get the article workflow ready for tomorrow’s authors. Joining us in discussion were Michael Casp from J&J Editorial, Kate Horgan from Aries Systems, and Nicole Brown from Silverchair.
Fresh after attending SSP in Portland, our Head of Publisher Relations, Romy Beard jumped straight into hosting the latest in our popular webinar series. This time, we discussed how we can get the article workflow ready for tomorrow’s authors. Joining us in discussion were Michael Casp from J&J Editorial, Kate Horgan from Aries Systems, and Nicole Brown from Silverchair. The experience and knowledge of the panel really shone through, as they offered reflections and ideas on how to get the article workflow ready for the authors of the future.
At ChronosHub we believe that good user experience will increase the speed of the publishing process. Romy pointed out that there is a very close relationship between author experience and article workflow, so a lot of the discussion overlapped. If there is good author experience, then it tracks that it will speed up the article workflow for both the author and the publisher.
Before we get into the details of the discussion, here’s a quick rundown of the key takeaways from the panel:
Michael got things started with a great visualization of the publishing process. A colorful wheel where everything in the process flows perfectly into each other – or so we thought! As Michael took us through each element of the wheel, he commented that “we run into a lot of bottlenecks”, and this observation provided a great opener to the discussion that was to follow. What exactly are these bottlenecks and what can we do to alleviate them? Is automation the solution?
Let’s start at the very beginning with the submission process. Michael thought that if publishers could find a way to have aligned submissions standards, then it would make a difference to the speed of submissions across the industry. A lot of work has already been done to automate submissions. A key innovation has been ‘format free’. This allows authors to submit an article without spending time formatting it to tricky specifications, which can change from journal to journal. Another is automating article ingest, which means taking the data from the article at the time of submission and matching it to, for example, persistent identifiers such as ORCID, Ringgold and ROR. This significantly reduces mistakes that take time to rectify further down the line.
And what about the peer review process – what types of bottlenecks occur there? Anyone who has ever participated in peer review will be familiar with the fact that there are never enough reviewers and editors are always seeking them out. J&J Editorial have seen the use of algorithms and AI to find new reviewers outside of the editor’s usual network. However, Michael remarked that they have heard mixed reviews about these tools, and this takes us back to the proposition – is automation the solution? Well, yes and no.
In this situation, you need a solution that stops system generated emails being filtered out by spam filters. This is especially important when an opportunity to find a reviewer has been missed because an email went to spam. Romy really appreciated that this point had been raised, because you can build the best workflow in the world, but if things fail due to such a simple thing, then it’s important not to leave out these details. The solution then becomes tracking the engagement points with the author so that any issues like this are thought through.
Our greatest opportunity to improve the author experience is actually taking a look at the point where editors are reaching out to reviewers. We have plenty of tools that can automate and support us up to this point, but we are still reliant on humans interacting with the content and giving their expertise to engage.
In an ideal world, authors would not be saddled with making a lot of administrative decisions about OA payments and could rely on their institutions or funders to handle this for them. This is one of the promises of transformative agreements.
Seamless integrations across different platforms and services could be a solution to this issue. According to the speakers communications integrations are an area of growth, as email is being slowly phased out. To begin with Integrations can be tricky, as a publisher may have to work with more than one partner to achieve a consistent workflow. Romy gave some good advice about this, such as avoiding multiple logins and providing a consistent experience even if a platform is running across different systems.
Furthermore, it’s easy to separate authors, reviewers, and editors into three different users, but they are all researchers. In practice, it’s often one person navigating three different roles and publishers need to keep this in mind when they are reaching out to researchers, regardless of the role they are playing in the process at the time.
Something that the panel agreed on is that platforms need to consider ways to incentivize peer review. We often think of value in relationship to money, but this is not necessarily the case in academia. Prestige and integrity have value – so one question to consider is how do we use this knowledge to incentivize peer review and add value to participating in the process?
After covering peer review, the panel moved on to production. Speed, of course, came up as a hot topic. Michael noted that the production side of things has increased in speed within the last decade, reaping the benefits of automation. However, he specifically cited APC payments, OA, transformative agreements, and copyright as slowing the publishing process down, which makes them areas prime for innovation.
Romy talked us through how ChronosHub measured the length of time it can take for an article to travel through the ChronosHub platform. The quickest time was 2 minutes 20 seconds, with a credit card payment. The average length of time is 40 minutes. ChronosHub has made use of automation and innovative user experience to decrease the amount of time it takes for an article to pass through each of the steps.
In this case good user experience will drive speed, if, for example, it’s easy to see what the license options are. A lot of work can be reduced by automation, so it takes minutes instead of days to complete a task in the article workflow. Simple switches like making license selection available on a platform rather than via email, as we have on ChronosHub, saves time.
Interestingly, Romy expressed that you can’t automate everything! She advised thinking about what would benefit from being automated, tailor the automation to your authors, and if there is a manual element, make sure that it is not too slow. At the end of the day, consider how each element – manual or automated – affects the author. One of Romy’s top tips is to map out the whole process and work out when tasks happen in the flow. This is something ChronosHub did at the start of 2023 with a group of publishers. At each step, ask yourself: does it have to happen this way or is there a more efficient way of doing it?
There are of course, other ways to increase the speed of the article workflow. Production must balance so many things: speed, accuracy, fidelity to the original accepted article. All of these things have to remain constant through different handover points between publishing staff and the researchers – who operate as authors, reviewers and editors.
What is the magic solution that will tie all of this together? Time and again, the importance of using XML came up. XML can be used on cloud proofing platforms where authors and editors can make edits directly to the underlying article XML, which saves time and increases accuracy, in comparison to PDF markup.
If everything is working as it should, XML can enable dreams – when the article is sent to production it can be ingested and published on the journal website with little to no human interaction. Then web pages can be generated automatically, metadata can be sent to indexing databases like PubMed or WebScience, and then the articles can be searchable and read by readers – the end goal. Eventually, these platforms can recommend articles to readers improving reach and discoverability.
Nicole from Silverchair sang the praises of standardization throughout the publishing process, and it doesn’t end with XML formats. Using a standard identifier, such as a ROR ID for an institution, means that it will be easier to deposit the article in a repository later. It also means articles can be indexed and found by those who need to read them.
Standards are the holy grail for automation, but not everyone adheres to them! So, what’s next for the article workflow? Well, agreeing on and working towards standards was mentioned as one thing. But it is no surprise that AI was mentioned. AI could be beneficial from the read and publish perspective, by providing even more detailed reports based on users and their interests, and the industry could really take advantage of the analytics available to improve author experience. Nicole would love to see AI used to enhance discoverability and engagement, so articles find their way to the right readers.
It was an interesting note to end on, where the panel were looking towards a future where the article workflow is much more automated and standardized. There was also thoughtful discussion about how much to automate and how to keep in mind the human element of the article workflow. As Romy said at the beginning of the webinar, we can automate as much as possible, but if the engagement points fail with the researcher, the speed of the publishing process will not improve. The panel definitely showed that there’s good reason to believe in a future where increased automation will be better for the researcher and the article workflow.