Morgan McLintic discusses various metrics for assessing blog influence – and gains plenty of useful comments on the issue. At its heart is the question ‘does size matter?’. We’ve always known in public relations that talking to the right people was more important than talking to the most people.
Or, to put it another way, how do you judge what’s a happy family? By the number of children, aunts and uncles? Or by some other measure of coherence and shared values/jokes/stories?
Corporate blogs will be asked to define their influence as a means of measuring ROI. Luckily, this doesn’t apply to personal blogs.We’re answerable to no one but ourselves.
PR Studies 
Thanks Richard – I think assessing the impact or ROI of your own blog (personal or corporate) is easier than assessing the influence of one run by a third party – say an industry commentator. You have the traffic stats, you have the conversations and connections, you have the RSS subscription numbers.
But mapping out the influencers in a particular sector or about a particular topic is still at an early stage. The tools we have all lack in one area or another. It’ll be interesting to see how they evolve over the next months and years.
This is very interesting and was discussed at length at the blogging conference last week. Obviously businesses want assurances about ROI, but it cannot clearly be assessed other than by checking out the number of hits on their website after the blog has been launched and any extra business that might result.
P.S. I shall post this on my blog today too and credit you.
Richard, nice thoughts. I agree with Morgan’s take on blog influence as well. I had the opportunity to talk with blog author Ted Demopoulos today and realized in that conversation that our ability to perceive the relative influence of a blog really is limited by the versatility of the monitoring tools we have at our disposal.
And, for those of us without deep pockets, our disposal isn’t very big, so we have to head for the more lightweight, free solutions like Sitemeter or StatCounter.