Over the past six PR Friendly Indexes: mostly declining
Now I’ve published a number of editions of the PR Friendly Index it strikes me that I have enough data to look at changes over this period.
So, I thought it would be interesting to take a look at the metric that seems most accurately to determine position in the index – Technorati Authority – and see how the blogs in the index have fared recently.
I’ve looked at the blogs which have been in the index since its inception (and since I got the spreadsheets properly sorted), and which have remained at the same URL throughout, so that a fair picture can be drawn – quite literally.
The charts use the funky Google Chart API which I’ve used before but might use for future PR Friendly Indexes. They are independently scaled – so a blog with maximum authority of 100 can be drawn in the same space as one with 1000 – and show the absolute change. Blogs are ranked according to percentage change from the first score in November 2007 to the latest in August 2008, and this is what the figures in brackets show.
As with the PR Friendly Index, take a look at the data below first, then I’ll add more description below that (please note that sometimes the Google API flakes out and the charts don’t appear – I don’t think Google are a particularly happy set of bunnies right now):
So, the most improved blog is, I’m happy to say, Barbara Rozgonyi who has continued to climb throughout the entire period. Her latest authority score is 83 and, given that she started at 11, this means she’s approaching a 7x increase in authority over this period.
I know, I know, small changes to small figures give big percentages – just look at the annual report for any start-up to see such claims. In fact I’m bracing myself for claims and counter-claims as to the validity of this approach.
But there are some heavyweight blogs that have experienced substantial percentage decline. Again, there could be many reasons for this, not least quite simply the challenge of finding time to maintain a blog, perform professionally and have some sort of social life.
I also thought it might be interesting to look at the total authorities across all core blogs over this time:
| Nov 07 | Dec 07 | Jan 08 | Feb 08 | Jun 08 | Aug 08 | |
| total | 16007 | 15521 | 16103 | 14485 | 13401 | 11658 |
| change | -486 | +582 | -1618 | -1084 | -1743 | |
| percent | -3 | +4 | -10 | -7 | -13 |
Not looking too great really is it? Comparing August this year to November of last year, there has been a 27% decrease in Technorati Authority. The average change month-on-month has been a decline of 870 Authority points, which equates to an average month-on-month decline of 6 percentage points.
What does this tell us? On the face of it, that the PR blogosphere is shrinking. Or maybe our core bloggers are just finding it difficult to maintain attention when there are new blogs around (note how only about three-quarters of the index is included here because other bloggers have stopped, started or changed in this time, such as my own blog for example).
Or maybe the ‘blog equity’ is also experiencing something of a credit crunch as other services such as microblogging and aggregators such as Friendfeed come to the fore. Perhaps we’re all just producing more quality rather than quantity, in which case perhaps this is a good thing.
Oh, and since you ask, my chart looks like this – bearing in mind I ‘came out’ as Brendan Cooper instead of being Friendly Ghost and made my primary URL brendancooper.com around December last year:
What I find weird here is how I have many more subscribers now, but a declining authority. I even get fewer hits generally now than before. But that’s for another day…
Brendan once again you continue to astound me with your penchant for researching PR blog performance. Good to know you and thanks a bunch for all you do for all of us. You rock!!!
@wiredprworks on twitter.com
I’m actually increasingly confused by Technorati Authority – I don’t think it works as a metric and if I was using it as a marketing metric, I’d be a bit miffed.
Generally, Getting Ink has had an authority score of around 80. Over the last couple of months it sort of went up and up, eventually reaching 111 about two weeks ago. That makes sense, my readership was increasing, too.
About three weeks ago, the authority score dropped overnight from 111 to 105, then 98, then to 96 and today it’s at 92. Despite there being 5 more links to the site showing up as responses than there were a week ago.
In fact, some days simply clicking refresh can alter an authority score and ranking by 30 or more points. It’s weird, but since I don’t really understand how technorati works in the first place, I’m not sure why it happens – but I wouldn’t rely particularly on it as a metric of anything.
Ooh, having said that I’m obviously still preening that my site looks dead impressive according to your graphs….
…while my site just looks dead.
@Barbara – thanks! In truth I’ve managed to get the workflow to a state where I just need to pull data in from worksheets and build html queries from them so it doesn’t take a huge amount of work. Interestingly enough a lot of the techniques I’ve developed of my own, I find extremely useful in my work nowadays.
@Sally – you’ve hit on the real problem with any metric – none of them are totally reliable. I’d say the ‘weird’ performance can occasionally be due to spam blogs linking up with other blogs, and occasionally Technorati quite simply goes a bit oopsy-la. Even Google have had difficulties recently, so much so in fact that I had to completely abandon Feedburner for some important work I was doing and am not going to return to it until I’m confident it’s passing the right information.
@ike – as I said, I’ve only included blogs that have been constant through this period – ie haven’t come in, dropped out or changed – simply so I’ve got comparable data, especially for the totals. It’s not personal – even I’m not in the ‘core blogger’ set.
Oh no, my graph is doing the dip
You are so right Brendan, no metric is totally reliable and I think that Technorati is increasingly off. It is funny that when I look at your graph for my blog it does coincide with a ramp up on my end in RSS subscribers and (more importantly IMHO) of comments left by folks. But it also runs with a dip in actual blogging for myself, which is interesting looking at all the other data.
No matter what I love reading this stuff and the conversations it sparks!
/kff
Great effort Brendan; did this in any way consider the ‘softer’ measurement metrics we discussed a few months back? Gerry
Lordy, no idea. I can only remember back as far as last week nowadays. Remind me again, what were these?
[...] Last Saturday, I discovered Wired PR Works blog led a group of 100 PR blogs –by over 650% – in the most improved Technorati ranking category. [...]
You said: “…as far as I’m aware you need pretty sophisticated algorithms married to computing grunt and supplemented with substantial manual work to do this properly. That’s why entire companies are devoted to it (Onalytica, Cymfony etc). I’ve looked into different ways to ‘tweak’ the PR Friendly Index. I’m wondering whether I should move away from web stats and more towards social media stats. For example, if I could grab del.icio.us stats then at least that would go some way towards establishing what people think about a blog, rather than what computers do.
http://prdisasters.com/blog-leagues-a-magic-bullet-for-online-popularity/
Remember?
Ah yes!
In which case, the charts above don’t include any of these ‘softer’ metrics – they’re just Technorati Authority – but the August PR Friendly Index looks at more wide-ranging measures. I compiled figures from HowSociable.com which generates a ‘score’ from across 22 social media sites.
I looked into grabbing data from del.icio.us but I couldn’t get consistent results. Since it’s been redesigned I might look into this again, but I think the problem I found with HowSociable – that is, how to derive a unique search term that is mostly fair across the board – will be even more intractable with del.icio.us.
For example, I get wildly differing results if I search for brendancooper, brendan cooper, brendancooper.com and http://brendancooper.com. Which is correct? How do I do this with other sites that have generic names? And if I should do all four searches, for all blogs, then that suddenly becomes four hundred calls. I just don’t have time/willpower/brainpower.
It’s still something I’d like to include though. Maybe one day I’ll wake up and think of something clever. Meantime, we’ve got HowSociable which kind of helps.
[...] Over the past six PR Friendly Indexes: mostly declining [...]
[...] get the blog back-on-track and no longer take the mantle of the blog with the fastest reducing authority. Oh and as Steve’s post (recruitment consultants dream list) shows I’m also [...]