Skip Navigation

People also read…

Related Content Linking and Oxford Journals

By Richard O'Beirne
Electronic Publishing Manager

In recent years, OUP has been working with our publishing partners to adapt to a changing online world. With deep linking to content now the norm, the article is becoming the primary point of entry to the journal. In 2011, 8% of our website's page views were tables of contents and 6% home pages, compared to 83% for article content. This has influenced the design of journal websites so that today 'every page is a home page.'

Another strategic question we are addressing is sometimes referred to as the attention economy, a phrase (coined in 1971, well before the invention of the World Wide Web) that describes the situation where:

...a wealth of information creates a poverty of attention and a need to allocate that attention efficiently among the overabundance of information sources that might consume it. source

In other words, attention is a scarce commodity and once we get a reader's attention, we need to maximize the time they spend on our website by helping them to find content of interest. If we can do this successfully, it will be mutually beneficial: your journals' readers find useful and interesting content quickly, and the journal's website benefits from increased usage – vital in order to ensure institutional subscription renewals in today's competitive market.

Related content linking example

Early Trials and Current Beta

In order to meet this need, in early 2011 we trialled three possible solutions using different technologies. Following the trial, we are now running a live beta test of a usage-based recommendation system. This technology anonymously analyses user behaviour throughout all of OUP's online journals, taking into account time spent on any particular item, or whether the user scrolled down the page. The software uses various algorithms to identify, for any given article, other articles which may be of interest. The results can include articles from across the range of OUP journal titles, as well as the journal the user is currently reading.

Our approach is similar to that used on websites such as Amazon.com, The New York Times, and the Harvard Business Review. We are interested to see how well it performs on the OUP website. One important caveat to bear in mind is that because reader behaviour can be unpredictable, occasionally results will reflect that. In theory, as the database of behaviour grows in size the results will become more focused, and because the software observes user behaviour in real time, 'hot' articles are quickly brought to readers attention. We are currently assessing user feedback to the recommendations service. Based on 80 responses to an online poll, 78% found the recommendations relevant.

Try it out

The recommender service is currently live on a number of sites including International Immunology, the British Journal of Anaesthesia, Plant and Cell Physiology, and the Journal of Biochemistry. Further Society-owned journals will be invited to opt in going forward and in 2012 we will review and report on the conclusions of this extended trial.