The Library Research team recently delivered some training on Increasing Citations. As part of that session I spoke about using SEO for research publications. The below is a survey of that content. If anyone is interested in us running that session for a particular department/school, please get in touch – Chris
In a world where researchers increasingly use the web to search for literature, and where that web is increasingly crowded, researchers need to do whatever they can to get their research discovered. SEO techniques can help to get your research outputs discovered.
A research output may be full text indexed by a search engine. However, different sections of it (e.g. title, abstract and keywords) may be weighted differently by a search engine. Judicious wording and selection of key phrases and concepts in these sections may render your article higher up the returned results in a web or database search.
Titles should describe what the research is about – they should give the reader a clear idea as to what the paper is about. That might sound obvious, but here are some (real) article titles:
- “A message from Titanic”
- “From lemonade stands to 2065”
- “Hot potato endgame”
Any ideas what they might be about?… no me neither. If you do stumble across a creative title for a research paper don’t dismiss it, maybe you can use it in social media, but it’s probably best not to use it as the formal title.
The title is likely to be the first thing a potential reader will see about your article so it’s crucial to let them know what it’s about. Patrick Dunleavy offers some great advice in his blog post Why do academics choose useless titles for articles and chapters? :
- The title should be relevant
- The title should be consistent with named concepts in the abstract and sub-headings
- Consider using a full narrative title e.g. ‘New Public Management is Dead — Long Live Digital Era Governance’. This has 2 specific topics, memorable non-academic language and lends itself to citation e.g. “Some commentators think Public Management is dead (Dunleavy et al, 2006)”
- If you cannot do 3 above, at least provide some narrative clues
Keywords are terms used to describe the key concepts articulated in the research output. They can appear in the title, abstract or body of the work – but do look for terms that may not appear in full text. These terms may be synonyms or acronyms or possibly larger conceptual terms that are not specifically named in the body of the text. (There is a story of an article that appeared in the Washington Post about Arnold Palmer scoring successive holes in ones – unfortunately the article omitted the term “golf” – so a search on golf didn’t return this article!)
- Think of keywords as potential search terms
- Use keywords that are common terminology in your research field
- Include relevant synonyms as keywords
- Include keywords in your abstract and body text
But, be aware of cramming keywords or repeating them too much – search may exclude items they consider to be keyword stuffed and, ultimately, the reader of your paper is a human not a search engine. Don’t repeat keywords to the point they detract from the flow of the text.
For example, useful keywords for this blogpost might include SEO; Search Engine Optimization; journal articles; research outputs.
An abstract is a piece of text that should convince a potential reader to read the whole article – its function is to aid selection – and always provide one when you can.
However, it is also likely to be more heavily weighted than the body of the article, so in an online world of full text indexing it is also choice text that is indexed by search engines – its function is also to aid return in search.
A 2015 study by Weinberger, Evans and Allesina in PLOS Computational Biology looked at how longer abstracts containing terms that used superlatives and signaled novelty or importance were more likely to be cited than shorter simpler abstracts, they conclude: “Longer, more detailed, prolix prose is simply more available for search.”
So think about how the abstract is not just a representation of the paper but also text that will be indexed by search engines.
Don’t be afraid to cite your previous work – but don’t overdo it and make sure all works cited are relevant. According to this Wiley guide, search engines factor citations into how they rank your work.
Graphics used in a paper should be vector (e.g. *.svg, *.ai, *.eps, *.ps) rather than raster (e.g. *.bmp, *.png, *.tif and *.jpg) as text in raster images cannot be read by search engines and are therefore not used to aid the ranking of the work in search results.
Be consistent with your name across publications and with names of authors you cite in any particular paper. Advice from the publisher Sage is to use your full name including middle names – try to make sure it is distinguishable from other names of researchers. Even better get an ORCiD, associate all your research with that persistent identifier and use the ORCiD when submitting work to publishers.
The work you do on SEO is contingent on the platform your research output sits on, some sites are better indexed by Google than others. In Google search ORO items consistently come up alongside or above the same output on the publishers website. ORO is a great discoverability platform and so are other repositories and academic social networking sites like ResearchGate. So make sure the article is on a platform that is well indexed by search engines, and that platform might not necessarily be the publisher’s site.
Finally, timely reminders from Witold Kieńć on Open Science… [SEO] will only work if the publication itself is good and interesting enough. Academic SEO does not substitute but supports the quality of content.
and from seanrnicholson…
Some Further Reading:
The Wiley guide: Search Engine Optimization: For Authors
The Elsevier guide: Get Found – optimize your research articles for search engines
Excellent commentary on the publisher SEO guides from Wouter Gerritsma: Academic search engine optimization: for publishers
A good recent overview of SEO and research from a workshop at British Ecological Annual Meeting in 2015: Maximising the Exposure of Your Research: Search Engine Optimisation and why it matters
Witold Kieńć on Open Science, Why and how should you optimize academic articles for search engines.
This is a really interesting read. As someone that does SEO full time I had never even considered how powerful it could be for research publications. I just wanted to back up some of the information provided here – top notch.
I truly want to echo the previous comment because I am SEO Services Provider myself. My focus thus far has been on commercial enterprises, and I could potentially have lost out on much more opportunities by not considering the use of SEO for research publications. It is worth contemplating the comment that “academic SEO does not substitute but supports the quality of content”. After all SEO does have to do with relevancy and a great article in the “wrong” publication, can do more harm than good.
Google is much more likely to spot the keyword-heavy articles which are being referred to in the cartoon. A quality article on a subject is much more likely to rank well (and to be shared naturally – which is another indicator to Google of a good piece of work). I’d have to disagree that a great article in the wrong publication can do more harm than good – I can’t see any way this could cause a negative impact but I’m happy to be proved wrong.