Sunday, March 03, 2013


Spikes doesn't have a Mission Statement (or I'd have to resign as Internet Consulting editor ;-) but if it did, it would be something like To bring plant science to a wider audience. For that reason, we care about the traffic the site receives - not for egotistic reasons, but because otherwise, well, there's not much point.

Last year the site experienced a big reddit traffic spike (much discussed at solo12), and we've just had a Tumblr spike. These periodic events stand out above the steady traffic the site receives from a range of referrers, including email subscriptions. But the magnitude of these events is such that they dominate the overall year-on-year stats. Consequently, it's difficult to avoid our actions being biased by these relatively rare events, and I admit, we have been sufficiently conscious of them to direct some of our effort to courting traffic from referrers by ensuring our content is represented there. But the reality is that spike traffic is chaotic, and therefore unpredictable. If I knew how to dial-up a traffic spike once a week or once a month, I'd be rich. Well, actually, I do know, but I'm not interested in writing the sort of low rent celebrity-driven content that requires (Celeb O'tweek splits from long term boyfriend: "My love of Arabidopsis is greater than my love for Justin"). Consequently our content seeding strategy delivers ... patchy results.

In contrast, over at MicrobiologyBytes, some of the best performing content is non-newsworthy items written years ago, which consistently gathers enough views to stay at the top in the face of occasional spikes. I've written before about how, when I started MicrobiologyBytes in 2005, I thought I was writing a weekly magazine - tomorrow's chip wrappers. It didn't take me more than a year to realize that I was actually producing a string of reusable learning objects as blog posts. A good example of this are the MicrobiologyBytes podcasts, which I abandoned three and a half years ago, but which still receive over 5,000 downloads a month without any further attention from me in the meantime.

So our strategy at needs to change course slightly. We need to stop chasing spikes by targeting individual platforms and instead we need to produce spike-friendly content. To reach many audiences, this needs to be diverse, and could well be visual (e.g. graphics, video, animations) as well as text-based information. For my contribution, I will be concentrating on writing explanatory appraisals of papers to make them accessible to a wider audience, but I encourage my co-contributors to continue producing a diverse range of items, including the steady flow of content that is the life blood of AoBBlog. Our target is in the total value of the content of the items we produce rather than in achieving those oh-so-tempting content spikes. If we build it, the spikes will come. 

No comments:

Post a Comment