Can we make valid and useful comparisons of blog traffic based on
subscriptions statistics reported by rss readers and feedburner?
Bloglines, just one of many popular RSS readers, reports that there are 182 of it’s users subscribed to the feed from cybersoc.com. Feedburner, on the other hand, shows 89 people subscribed via bloglines – 11% of the total of 821 subscribers.
That’s a discrepancy of just under 100 subscriptions, 19 of which can be accounted for by the fact that two bloglines users are subscribed to the podcast version and 17 are subscribed to the old, typepad feed.
In early November, Adrian Monck compiled a list of the top 10 most subscribed to journo blogs in the UK. Cybersoc appears at number four on that list with 73 google reader subscribers. Feedburner, as you can see in the screenshot above, reports 252 subscribers using google – a discrepancy of around 180.
I understand why people want to compare the traffic of different blogs but I’m not convinced that the figures for RSS subscriptions, which appear to me to be unreliable, is the way to go about doing it.
Another difficulty, however, is that normal web stats only show part of the picture. See the page view and unique users stats, compiled by statcounter, for cybersoc.com over the past year. Month on month traffic has held fairly steady except for a big spike in April following my posts about the Virginia Tech shootings.
Yet although the stats remain fairly flat, rss subscriptions to cybersoc, as reported by feedburner, continue to rise steadily. This makes sense – as more people adopt RSS into their toolkit, page views and other measures of people reading a website will remain flat or even decline.
Technorati is another candidate for the measurement and comparison of blogs but, as any blogger knows, it is far from perfect and misses some inbound links. Technorati rank is also skewed towards the benefit of older, more heavily trafficked blogs than the many excellent newcomers. Maybe I’m just saying that because, having hit a high of around 9500, my rank has been in free fall and is now around the 27,000 mark. What does seem more meaninful than technorati rank is the number of inbound links but, again, this skews the results in favour of older blogs which may or may not be still generating lots of inbound links.
Lists, of course, are one way to generate lots of those inbound links. But lists of blogs, compared based on spurious rss subscription numbers, web stats, or technorati rank aren’t, in my opinion, very likely to tell us much other than the inadequacy of current tools for doing the job.
Here is another example from Paul Bradshaw in his words: Inspired by Martin Belam’s extensive charts of popular RSS feeds, and Adrian Monck’s list of popular UK journalism bloggers, I’ve grabbed the baton and produced a chart of the top ten American journo-bloggers, based on combined subscriptions via Google Reader and Bloglines…link
As for Technorati, i think that they have changed the way they count links now. There is some form of reset every month or so.. I`m not sure but the reason for the big decrease may be that.
Odd discrepancy given Google own Feedburner! I agree it’s another rough and ready yardstick and I have to say I knocked my list up more as an exercise in curiosity – the post that no-one picked up on (probably because they weren’t listed!) was what Google reader revealed about Times columnists compared to bloggers, where the problems of RSS subscription come home to roost.