Social media's limited impact on promoting scientific articles: the case of 'Circulation'
Why social media isn't an effective way to promote scientific articles?
Social media is a powerful way to promote mass-market content and increase exposure. Facebook, in particular, has a content-friendly algorithm that rewards users for writing articles, with a preference for long-form ones.
Given these factors, it's easy to assume that Facebook would be an effective platform to drive views to your scientific articles. Increased views can lead to higher citation rates, which is desirable for academic content. However, the reality is different.
The experiment: how 'Circulation' tested social media's effectiveness
Facebook encourages users to remain on its platform rather than directing them to external sites. This means it's actively discouraging its 1.86 billion active monthly users from leaving the platform to access your content.
Additionally, a two-month study conducted last year revealed that Twitter, another major social network, is among the least effective platforms for engaging users with scientific content.
The most compelling evidence comes from a randomized, controlled study conducted by Caroline S. Fox and her colleagues on articles published in Circulation in 2016. Their research found that social media had no effect on the pageviews of a medical journal. Even with the use of paid ads, no significant benefit was observed from using either Facebook or Twitter.
This post breaks down the experiment conducted by Fox et al. (Journal of the American Heart Association) and its implications for content creators and consumers.
Key findings: why social media didn't boost page views
The experiment
In their study, 152 original articles published in Circulation, a leading weekly journal on cardiovascular medicine, were used to test the effectiveness of social media promotion. These articles, published between January 13 and September 22, 2015, were randomly assigned to either a control group (n=78) or a social media group (n=74). The social media group's articles were promoted via Facebook and Twitter, while the control group received no additional promotion.
Design schematic from Fox et al. 2016
The social media group promotion strategies included:
- Posts with images and statistics linking readers to newly published articles on both Facebook and Twitter
- Three posts per article—once at publication and two more at 11 a.m. and 3 p.m. the following day
- Facebook posts were promoted using $10 worth of paid ads targeted to their audience
- Outside the experiment, Circulation's Twitter and Facebook accounts maintained engagement by posting and reposting content (2-7 posts per day). Both organic and paid traffic were used, and their marketing efforts grew followers but didn't raise article page views.
Implications for publishers: exploring alternative strategies
The results
As a whole, the combined efforts—paid ads, social media activity, and more followers—didn't move the needle for Circulation's page views. The mean 30-day views per article were nearly identical: 616 for the social media arm and 597.2 for the control arm. Median views were 499.5 vs. 450.5 respectively, but neither difference was statistically significant.
Section of Figure 2 from Fox et al. 2016 depicting the cumulative percent of 30-day page views by treatment arm.
Potential implications and questions
The study revealed no significant difference in 30-day page views, despite best-practice promotion across Twitter and Facebook. While social media marketing drove a 50% increase in followers, those new audiences didn't translate into more article views.
While this isn't ideal for those banking on social networks to drum up readers for academic content, there are still effective ways to market scientific articles—and one involves social media in a different way.
Here are the key takeaways for publishers and readers:
Takeaway 1 – Social media doesn't drive organic traffic
Facebook continues to shrink organic reach, and Twitter's algorithm is on the same path, with Ogilvy and Mather forecasting that organic reach will approach zero. In the Circulation study, going from 46,000 to 87,000 Facebook likes and from 6,700 to 10,000 Twitter followers still didn't increase article views. Social media simply didn't deliver organic readership for scholarly content.
Takeaway 2 – Social platforms want you to stay put
Twitter bought Periscope for $120 million to keep live video in its ecosystem. Facebook countered with Live streaming and Notes for long-form posts. Both platforms want you to consume more content through them, not on your site, making it even harder to drive journal traffic.
Takeaway 3 – It's time for other solutions
Publishers, researchers, doctors, and pharmaceutical companies need approaches beyond traditional social feeds. Owning a contact list lets you reach readers directly. Specialist services like TrendMD and PubMed—platforms that surface relevant content to interested readers—drove more engagement than Twitter, Google, or Google Scholar in our own two-month study.
Key takeaway:
- Optimize Facebook and Twitter for brand awareness, but consider other options if readership is the goal.
- This post summarizes the research conducted by Caroline S. Fox and her colleagues in the Journal of the American Heart Association.
- Fox, C.S., Gurary, E.B., Ryan, J., Bonaca, M., Barry, K., Loscalzo, J., Massaro, J. (2016) Randomized Controlled Trial of Social Media: Effect of Increased Intensity of the Intervention. J Am Heart Assoc 5:e003088 doi: 10.1161/JAHA.115.003088
