Most of my "job" involves communicating science, all the way from lecturing undergraduates to what most people consider wasting time on Facebook. In the course of a typical week, tens of thousands of people read what I write. But increasingly, what I write seems less important.
Occasionally, I write about my own primary research, but the vast majority of the time I am performing a curation role, shaping people's attention to what I, as a seasoned practitioner (never "expert", please) consider to be important, significant, and indeed, entertaining. Over the past few weeks I have been conducting an experiment based on what I have observed. This has involved leading not with text, but with a single image. The text accompanying the image is minimal, accessible, and contains a link to the source. The text also needs to be above the image (rather than as comments), so that when the image is reshared (see below), the text stays with it. On MicrobiologyBytes, this is normally a scientific paper, for AoB Blog it may be a paper in the journal or it could be purpose-written original blog content. Life is easier when sourcing from open access journals, which I favour on MicrobiologyBytes whenever possible, and of course, in AoB PLANTS we have a pipeline of high quality images published under a Creative Commons licence.
I'm not claiming this "discovery" is in any way original, examples of people who do it well include ASM on Facebook and many astronomers who have used the deluge of beautiful celestial images to their benefit. Some professional media outfits such as the Financial Times do a good job, others - not so much, e.g. Reuters. The theory here is that social networks such as Facebook are not primary publications, only pointers to content elsewhere. With Twitter, it's easy - 140 characters doesn't leave you enough space to do more than point. Facebook and Google+ are more tempting, and pointing is not the only way to do this, it's possible to make them your primary publication platform rather than an owned site such as a blog, although most people would consider this to be a dangerous long term strategy.
I have not formally measured outcomes as a good analysis of the drivers would be so multifactorial (content, colour, time of day, day of week, platform, etc, etc) as to be virtually uninterpretable, but I'm happy with the results I'm getting. Image-led pointers not only get more views, they generate more in platform engagement such as comments and shares. I'm personally surprised that image should generate more engagement than text but that is the way the world seems to work these days.
So why am I writing this post, giving away all my "secrets"? Because I'm shocked that more science communicators have not figured this out for themselves, or observed it by watching others as I have done. I still see so many blog posts and press releases without images. I pass them by without stopping - why did people waste time writing them? Sure, finding a suitable image may take some work and imagination. Writing a press release on the world's most boring metagenomics paper? Unless it fails to name a single species (in which case, why publish it?), there's your picture lead. It just takes a little effort and imagination.