Site moved to www.rohitbhargava.com/2006/03/finding_a_new_f.html, redirecting in 1 second...

« Are Domain Names the Next Tagging Phenomenon? | Main | 5 Ways Marketing Can Save the US Airline Industry »

Thursday, March 23, 2006

Finding a New Formula of Blog Influence

In a world of personal media where the technical constraints of publishing content are becoming nearly nonexistent - some have recently discussed the rising problem of information overload.  Seth Godin suggests that we may all be adding to the clutter, and though self-filtering, or relying on content creators to tag their content is ideal - it also depends on an objectivity that most people don't have.  To get around the human factor, Technorati is among the companies trying to lead the way by launching tools such "filter by authority" feature - which uses the number of people that link to a blog as a barometer for influence.  Using inbound links as the measure is problematic firstly because there is no prioritization given to different types of inbound links, and more importantly, because it does not account for bloggers who are posting influential discussion oriented content that results in large numbers of comments, but relatively few inbound links or trackbacks.  Guy Kawasaki's posts get lots of links back to them.  Joel Achenbach's posts get lots of comments.  Which is more influential? 

To answer this question of influence, I believe bloggers as a whole need to start a discussion about the metrics (beyond inbound links).  Ultimately, the real formula will depend on putting all of these things together. To get started, here is my attempt to rate the importance of an inbound link based on the type of link and who it comes from.  I would love to see an automated tool that could quantify inbound links based on which type of link they are.

Inbound links (in order of importance):

  1. From blogs in the Top 1000 as listed by sources like Technorati
  2. In blogroll of Top 1000 as listed by sources like Technorati
  3. From blogs or sites that also reprint some body text from a blog post (full text reprints, however, do not count as they are likely to be spam blogs)
  4. From tagging sites like del.icio.us or Digg.com to blog homepage
  5. From tagging sites like del.icio.us or Digg.com to blog homepage to individual blog posts)
  6. From blogs or sites referring to blog homepage
  7. From blogs or sites referring to a blog post

To complete the picture, below is my own list of other criteria that I also look at beyond a blog's links in order to determine influence (in no particular order):

  • Age of blog (how long has it been live?)
  • Affliation/employer of blogger
  • Number of blog posts on the blog
  • Average number of comments
  • Distribution of syndication of site (through RSS or email subscription)

Do you have other metrics for influence that you look for which I missed?  Have you seen a better measurement for influence out there?  Let me know.

TrackBack

TrackBack URL for this entry:
https://www.typepad.com/services/trackback/6a00d8341c4f1253ef00d8341c70c153ef

Listed below are links to weblogs that reference Finding a New Formula of Blog Influence:

Comments

I have to say I also like to use sites like Egosurf to basically do the calculations for me. Also, the conversations tool on Blogpulse is another favorite. Beyond just how many people are blogging about it, I like to see how far reaching the conversation is and how long it goes on.

I also use the Alexa rankings as an indication of reach -- if I'm dealing with a particular website.

Measuring influence has been done in the academic world using citation analysis for over 30 years. Citation analysis measures measures both direct and indirect influence.

Leontief won the 1973 Nobel Price in Economics for the input/output method that measures how connected entities such as sectors of the economy influence each other.

I'm amazed how many people simply "invent" metrics based on their gut feeling or conventional wisedom. Measuring influence has been done in the academic world using citation analysis for over 30 years. Citation analysis measures measures both direct and indirect influence.

Leontief won the 1973 Nobel Price in Economics for the input/output method that measures how connected entities such as sectors of the economy influence each other.

I'm amazed how many people simply "invent" metrics based on their gut feeling or conventional wisdom. What happened to good quality research?


Hi Flemming,

Thanks for commenting. I followed the link to your blog and read your post about the most influential authorities on Business Blogging with interest. Your efforts to create the Issue Influence Index through your company speaks to the need that I wrote about in this post. A need to move beyond equating popularity with influence (as you also discuss) and having a more scientific approach to determining influence. ultimately, the problem that led me to creating my own list of how I have heuristically determined a blog's influence is because I have not yet found a relatively simple tool for doing this.

As you suggested, true measure of influence has to come from good and thorough research. Unfortunately, marketing teams do not always have the time, funding or (sadly) the inclination to perform this research. People seek quicker, simpler and higher level tools to perform some amount of this analysis without the necessity of exhaustive research. I'd be interested to hear your thoughts on whether there can ever be a "one-click" method of determining influence, or whether it will always require the academic discipline inherent in a solution such as your Index.

One way I look for signs of a healthy blog is looking for "Blog Resonation"

I've listed out several items below:

http://jeremiahthewebprophet.blogspot.com/2005/11/signs-of-healthy-blog-resonation.html

This transcends beyond just being 'influential' as all blogs are influential in a specific market. I see your focus on 'top bloggers' but remember to 'think small' there are A-listers in every industry.

Thanks Jeremiah - it's a great concept to view a blog in relation to its resonance with readers. You are right that the layers of influence can be very thinly sliced to far more niches than just the "A-list." My view is that an interesting extension of this will be when more celebrities and "offline A-listers" start to blog. I wonder how this will shift our current discussions of influence? Perhaps a new metric will emerge where influence can also be based on who you are rather than the quality (or resonance) of the content you produce. If that does come to pass, it will make the process of measuring influence that much more difficult.

Great article. Heres a thought about the Technorati "filter by authority" feature. It may appear a 'fair' ranker of blogs, but the higher ranking blogs get even more traffic as a result of their high ranking, which ends up being a circle of the top blogs getting further and further away from lower onces because more people visit them, which results in more backlinks, which results in even more visitors etc.

But thats the internet, its all about braking into a higher level.

Very interesting posting and it seems you've gone further into this than I recently managed (see http://thefriendlyghost.wordpress.com/2007/05/25/how-influential-is-your-blog/). I'd be really interested to know if you've since come across a 'one-click' solution?

I'm not sure there is a one-click way to measure influence, but I'm working with two other partners to create a visual map of influence based on incoming links.

Sites like Technorati measure total value, but the type of link, frequency, and breadth of connection of top sites to other sites would register influence.

The feeders for the instapundit site - bloggers who have Glenn Reynold's confidence, if not his numbers are good examples of influencers that would not be found in a traditional search.

Our goal is to analyze the relationships between blogs, and the sad fact is that much of the information going into a large engine is going to be garbage. Filtering sites with human editors, than running them through an analysis program will hopefully give us some better insight into what "influential" truly means.

Rohit - you'll be one of the first to know if we're successful. Jeremiah too.

Blogs are good for every one where we get lots of information for any topics nice job keep it up !!!

The comments to this entry are closed.