What is the usage profile like for a typical blog post on this blog? I suspect the statistics for the post on “Are You Able?” is fairly typical (although, as I confessed recently I did pimp up this post on Twitter.
What we can see is the post was viewed by most users on the day it was published, with a steady drop after that, although there was a slight increase in the numbers of viewers on the Monday after the weekend. It was also pleasing to note that most uses have been via the syndicated RSS feed. This is good news and provides evidence that much greater use is being made of RSS readers, by readers of this blog at least.
But let’s look at another blog post to see a very different usage profile. As can be seen we again saw a peak of about 200 views (again mostly views of the syndicated feed) on the day the post was published. But since the post was published there has been a long tail of daily views of the post and with a current total of 1,946 views on 23rd February 2009, most of the views of this post have taken place in the weeks and months after it was published.
What’s the reason for the difference? Although it may be felt that there aren’t significant differences as, over time, the Are You Able post might have a long tail of views. I’ll address that point by concluding this post with the usage statistics for a post published around the same time as the one mentioned above. And the title of that post “Pinky and Perky and Swedish Topless Model Caught in Use as Learning Objects” might give a indication for its popularity.
Yes, you’re right. This post is so popular because of the numbers of people searching for “topless model” and “Swedish topless model“. And I’ve been caught out for the unethical approach of using an inappropriate title with the sole intention of boosting the blog’s usage figures. Well, not (quite) true in my opinion. I did have a legitimate buy antibiotics in mexico interest in how use of such phrases could effect the amount of traffic. But I also have a need to think of new titles for blog posts (I’ve published over 500 posts, and I can’t call them all stuff I think is interesting about Twitter, Facebook, …). And I’ll continue to think of puns and word plays for the titles of the blog posts – and I know I’m not alone in this. I will, though, try to ensure that the titles are relevant to the post (there was a photograph of a topless Swedish model included in the post) – but a post called “Britney Spears nude” purely to pull in the traffic would be inappropriate (as well as showing that I’m not up-to-date with the latest pop babe).
But what about concerns that although it may help to motivate me as the author to think up interesting titles, this can skew the usage figures which may be requested by funding agencies? My response is that there are many ways to enhance usage statistics – as I illustrated in a post on Lies, Dammed Lies, Blog Statistics and Unexpected Spikes. So for me, if funding bodies wish to request inappropriate metrics, then that is their prerogative. But at least I’ve been open about my awareness that the usage statistics are flawed. And hopefully going public about the dangers of over-simplistic metrics will discourage the civil servants and bean-counters from mandating their use.
As I mentioned above in order to provide a meaningful comparison a graph of the usage statistics for a post on Butler Group Report on “Enterprise Web 2, published on 11th December 2008, a week after the Swedish model post, is shown. As can be seen, after the first week the number of views dropped off sharply, confirming, I believe, the reasons for the popularity of the Swedish model post.
Now isn’t it strange that the Swedish topless model has the long tail and not Pinky and Perky? I guess she must be a mermaid.