Posts Tagged ‘alertthingy’

Down with Twitter Authority: Up with Influence

Jan 18, 2009 Digital Marketing & Metrics, Microblogging, Social Media

Twitter “Authority”

numberoneA lot has been written recently on so-called “Twitter Authority”.  Apparently a few Social Media “A-Listers” thought it would be great to be able to rank the people on Twitter by authority. Why? Because, more companies are searching the web, including  Twitter – to find brand and product mentions. This is done to monitor brand reputation and to facilitate customer service.  Even a few years ago, when people talked about our brands, it was behind our backs and we only found out about it if they contacted us through 1-800 numbers, letters, or e-mails. Today, those dialogues are online and can be accessible through a variety of search processes. It is not an easy task by any means and certainly not a 9-5 job! There are innumerable ways to search the web for your brand name, for example I have used Tweetscan, Twitter Search for Twitter (more on using Corporate Tweeting), Technorati and Google Alerts for Blogs and News, and BackType for Blog Comments. But, there is also Social Mention, AlertThingy and some great paid Services such as home-grown Radian6.  With so many ways to search, and even more possible “mentions” – it makes sense, to find ways to narrow the search, evaluate the comments that we should tackle, and find a way to rate and analyze the messages coming through.  At first glance it may seem logical to assume that number of friends, followers, or potential eyeballs would be the best way to do this. This is how we have measured traditional media for years. With publicity we say an article about our product had X number of impressions, or an ad on TV or in print had X number of potential eyeballs. In these traditional forms of media there wasn’t much more we could do to measure one against another or to give meaning to our marketing efforts to the big wigs, and the sales force.

But the web is different. We can do so much more, then measuring Twitter authority by how many followers you might have. And, is authority truly what we want to measure? Virginia Miracle makes a great point on her blog Marketing Environmentalism when she says it is “less about authority and more about influence.”

Authority vs Influence

talking1

Check out the definitions:

authority: an expert whose views are taken as definitive; or a person accepted as a source of reliable information …

influence: The power to affect, control or manipulate something or someone; the ability to change the development of fluctuating things such as conduct …

We might say that Martha Stewart Living Magazine is an authority in ribbon and paper crafts, however does it influence its readers to start paper crafting? Hard to say without doing some market research on the subject! Online however, including Twitter we can get an idea of a person’s influence, but unless we can nail down the parameters of what having authority means – and it doesn’t mean how many followers you have (Mack Collier has some great points on why) – then we won’t actually have a clue who has it. In a vlog post by Jackie Huba and Ben McConnell they question- if someone has written a book do they have authority – and the decision – yes.  Therefore, being an author can make you an authority (on that subject) – but what other factors make you an “authority”? Who decides?

So perhaps we forget this concept of “authority”. Go beyond numbers of followers – this eyeball/impression mentality is best left to television and print. Examine who has the influence to effect change, make someone else take action.

TwitterRank

Huba and McConnell suggest that perhaps what we are looking for is a complex algorithm, and that one factor may be the “retweet”. How often are you submitting content worth “retweeting”, how many people are “retweeting” it?  In this case it is more about the quality of what you are sending into cyber-space then the quantity of your followers.  It also has more to say about the quality of your followers than the quantity. An active base of similarly interested followers is more likely to “re-tweet” your tweets, than followers who have simply followed for the numbers, who are robot tweeters (here’s a blog post, here’s a sales promotion….), or who set up an account and have since moved on. Once “retweeted” your tweet can be seen by innumerable others outside of your followers, so having 50 or 5000 followers will not matter.

As (authority – he has written a few books so clearly he is that!) Guy Kawasaki said in a recent interview “I’m sure people will figure out a way to game retweeting too but until they do, there is no better measure of the quality of a person’s tweets than retweeting.”

There are a growing number of tools to follow “retweets”

Beyond the retweet I think there are a few other factors that indicate someone’s influence. Certainly the numbers game should count to a certain extent, but they would count even more if we could look at it based on their rank as far as influence. We know that Google’s PageRank is a complicated (unknown) algorithm that is based on over 200+ factors (or so the experts say), perhaps something like this could be created around Twitter users. It could include factors such as the retweet, but also include the @ replies to and from the twitter user (interaction with others has to count for something), number of updates, number and quality of followers, and number we follow (can’t just talk to ourselves!) and if possible click-through on links shared (this one is likely technically impossible since we all use different URL shortening services). What other factors might indicate influence?

Is it possible? I don’t have the technical background to know. But, if someone can create “Most Retweeted Lists” and create aps that follow  Twitter conversations by programs like TweeTree, perhaps it is feasible. It is likely easier to find these influencers online than it is finding them offline which of course marketers have tried often to do in order to facilitate WOM.

Would this change how we follow? Would it change how we interact? In a good way? Maybe it wouldn’t be friendly Twitter anymore?

Update:

Thanks to Amber and Matthew for reminding me about another key issue and that is if the “influencer” is directly related to the business.  If a company is monitoring brand mentions for reputation analysis a person who may be seen as an “expert” (or in this case we could also say “authority”) then they will hold more weight than someone who has little connection to the category or is not part of the target demographic.  I use “expert” and “authority”, but in this case an expert is not one in a traditional sense of a talking head on the news.

An example perhaps? I will use what I know – small appliances. Lets say that a Silicon Valley tech geek tells their followers that he just loves his “X Brand Blender because it makes the smoothest drinks in the morning getting him off to a great start in the morning. ” This particular tech geek has an enormous and engaged following. But, is the tech geek an influencer in the realm of blenders? Will he influence his tech geek followers to run out and buy X Brand Blender next time? What if on the other hand  cooking blogger, also with an large and involved following tweets that she used “X Brand Blender to make Butternut Squash soup and it was so lovely and smooth, yummy!” Will she influence her followers to buy X Brand Blender next time?

When I did some brand monitoring online I would pay far more attention to the cooking blogger than the tech geek in this situation. That is not to say that the popular tech geek wouldn’t matter, but I see these mentions as potential eyeballs, possible points for brand awareness. They aren’t authorities on the subject, but neither are the celebrities we use to sell make-up, cereal, vacuum cleaners, and other consumer products. The tech geek will likely get his followers to test the latest web tool, but unless they are in the search for a new blender the mention will not mean much (sort of like TV and Print ads!) The cooking blogger on the other hand is an “authority” on cooking to her followers – what she says holds a lot of weight. Even if the followers aren’t looking for a blender at the moment, they may recall this statement down the road when they are.

So while we can perhaps come up with a ranking on Twitter, but how do we differentiate between the different influencer circles? Unfortunately, Twitter doesn’t have a whole lot of information on us in fields we fill out (compared to Facebook!). We tell Twitter our location (and here we could say anything from country to town), and a brief bio which could say almost anything or nothing. However, Twitter is information rich when it comes to our Tweets. If we could harness the information in the tweets we might be able to determine interests and determine “authority” to the influencers. Twitscoop gives us clouds of popular words, what if something could pull these for individual Tweeters? Sort of a Keyword ranking for Twitter users? The cooking blogger might Tweet more often about cooking, recipes and such while the tech geek might tweet about web tools, social networking and Apple.  A company could then choose a plethora of keywords that they want to locate the influencers for within Twitter. As a small appliance company I might choose Tweeters that talk frequently about “recipes, cooking, mixing, cups, food prep…”

Is this possible? No idea. I also want to mention, I don’t like the idea of ranking for regular use (for example I only follow people who rank number “1″) but for business purposes this could be helpful IF done correctly.

Remember to connect on Google Connect (right hand bar!) And, of course follow me on Twitter!

Popularity: 34% [?]

Rate this:
3.2

Tags: , , , , , , , , , , , , , , , , , , , ,