Activity Is Not Influence

from the tweet-this! dept

I recently wrote about my concerns with the rise of services that try to measure "social media influence" by giving it a number. I had a few concerns about this, including the fact that trying to put a number of something that is not quantifiable inevitably leads to problems, but also in that this would lead people to change how they use certain tools. I don't use Twitter for the sake of "influence," but as a communications vehicle. Yet, that harms my "scores" on these services and gives me incentive to do things that I'm not interested in doing.

It appears I'm not the only one concerned about this. Jeff Nolan points us to a writeup by Mack Collier, in which he express similar concerns, specifically noting that these services measure activity, which is different than influence -- and, in fact, can be antithetical to influence. He notes this because one of these services, Klout, told him his "score" was dropping, and the way to increase it was to tweet more things:
Essentially, Klout and Empire Avenue are measuring your level of social media activity, not your level of online influence.

Simply sharing more content and engaging with my network isnít going to make me more influential over them. In fact if itís not the type of content and engagement that they are looking for, my influence over them will fall, not rise as I become more active.
And yet, because these sites and their made up numbers declare that they're measuring "influence" lots of people just believe them. It's really unfortunate, and it's going to lead to people changing their behavior in ways that don't increase influence at all, but decrease it.


Reader Comments (rss)

(Flattened / Threaded)

  •  
    icon
    Marcus Carab (profile), Jun 3rd, 2011 @ 5:26pm

    I think you could measure influence scores in a more useful way by tracking the number of retweets and replies, weighted according to the influence scores of the people doing the re*ing. You could also slowly subtract influence points for tweets that garner no response, thus encouraging quality over quantity.

    It would still be an incomplete picture (a tweet can start lots of conversation that doesn't go through the built in reply/retweet channels) but it would be better than basing it on volume.

     

    reply to this | link to this | view in chronology ]

    •  
      identicon
      Anonymous Coward, Jun 3rd, 2011 @ 11:12pm

      Re:

      It's kinda like with peer review journals, the number of citations is an important quality evaluation ingredient. But there are all sorts of problems with it too. It tends to lead to a domino like effect where people tend to cite those that cite them and so people cite each other for the sake of being cited and patting each others backs. These sorts of citations are weakly related to the quality of the published work.

       

      reply to this | link to this | view in chronology ]

      •  
        icon
        Marcus Carab (profile), Jun 4th, 2011 @ 10:29am

        Re: Re:

        I think when you were working with tweets, which have a much higher volume, it would make the back-patting more difficult. I guess there would be a danger of natural "clumping", where groups of influential people with similar interests/fields would always retweet/discuss each others' tweets, so the algorithm would have to be designed to compensate.

        There's probably a name for the kind of scale I'm envisioning but I'm not a real mathematician so I don't know it. What I'm thinking is a scale where retweets by people a the same influence level are worth 1, then for people below your influence level the value trends towards 0, and above you it trends towards infinity. Thus an influential person has to get a huge volume of retweets from followers below them for it to be worth much, whereas an uninfluential person can get a huge boost if Oprah or someone retweets them.

         

        reply to this | link to this | view in chronology ]

  •  
    identicon
    Anonymous Coward, Jun 3rd, 2011 @ 5:45pm

    Somebody sounds peeved that they don't have a whole hell of lot of followers on twitter...

     

    reply to this | link to this | view in chronology ]

  •  
    icon
    E. Zachary Knight (profile), Jun 3rd, 2011 @ 5:50pm

    I always liked the score of follows/following. A person who is following 4000 with only 100 followers is less influential than someone with the reverse.

    Of course that is still pretty meaningless. But oh well.

    I do think it is far better to have a quality social media presence, which you are advocating, over a quantity presence, which these people are advocating.

     

    reply to this | link to this | view in chronology ]

    •  
      identicon
      taoareyou, Jun 4th, 2011 @ 11:55pm

      Re:

      There are plenty of sites that let you sign up and get tens of thousands of followers with no effort. The number of followers you have is an absolutely useless number for determining anything about the tweeter.

       

      reply to this | link to this | view in chronology ]

  •  
    icon
    PW (profile), Jun 3rd, 2011 @ 6:05pm

    Closing the loop

    One of the BIG probs that Klout and others face with advertisers and marketers that consider these influence scores is that they're missing something important. Influence only matters to marketers to the extent that it moves some needle, whether that be sales of their products or brand awareness. Unfortunately, there's no easy way to measure if these influence scores correlate in any way to either of those metrics.

    The way this is handled w/TV today, is also a bit of a black art, where every few month surveys are run to poll users on brand awareness shortly after a campaign has been run. This isn't a great way to measure but advertisers are accustomed to these surveying methods. This is why online advertising has seemed so attractive to advertisers. With online advertising, it's easier to measure whether people clicked on an ad and took other actions.

    Until companies like Klout, Peerindex and Empire Avenue can substantiate the correlations between their scores and useful metrics for the advertisers, I don't see the level of activity they're seeing as anything more that marketers putting their experimental budgets to work. These don't last for ever, and if they don't find a way to measure results soon, they won't last much longer ;)

     

    reply to this | link to this | view in chronology ]

  •  
    identicon
    abc gum, Jun 3rd, 2011 @ 6:18pm

    Some people will believe anything

     

    reply to this | link to this | view in chronology ]

  •  
    identicon
    Travicane, Jun 3rd, 2011 @ 7:25pm

    Tweet-a-Fart

    -- the new way for the flatulent to increase their influence, and attract an avid twitter following due to their daily increasing influence rating!

    /sarc

     

    reply to this | link to this | view in chronology ]

  •  
    identicon
    GabbyQ, Jun 3rd, 2011 @ 8:39pm

    sorry 4 them!

    i know of a "couple" of users with a "high" klout score and their tweets are like... "F*(&" or.."i'm drunk" you know.... a lot of stupid things that seriously do not "influence" anybody... still.. they are happy to be "Influencers" for Klout... and they bluff about it!

     

    reply to this | link to this | view in chronology ]

  •  
    icon
    CheMonro (profile), Jun 3rd, 2011 @ 10:17pm

    Activity is not equal to influence... TRUE.

    But, can you have influence with no activity? No. And if you look at the commenters who regularly post on this forum, the ones who comment on story after story after story... Won't they tend to build up influence?

    So activity is not infulence, but... activity is certainly related to influence, and... oh... a person's maximum potential influence increases with activity.

     

    reply to this | link to this | view in chronology ]

    •  
      icon
      Jeni (profile), Jun 4th, 2011 @ 9:10am

      Re:

      "But, can you have influence with no activity? No."

      True. Absolutely. But if the biggest SE in the world doesn't think you are "active" enough, which supposedly means "influential", and drops your rankings so low no one can find your "stuff" any more, you can't have any influence, either, no matter how helpful or "good" your content is.

      It's a 2 way street. Activity can take many other forms than a social link, too. Email, for one. Letters, for another. Newsletters, phone calls...

       

      reply to this | link to this | view in chronology ]

  •  
    identicon
    Anonymous Coward, Jun 3rd, 2011 @ 11:06pm

    "Essentially, Klout and Empire Avenue are measuring your level of social media activity, not your level of online influence.

    Simply sharing more content and engaging with my network isnít going to make me more influential over them. In fact if itís not the type of content and engagement that they are looking for, my influence over them will fall, not rise as I become more active. "

    Maybe a slightly better measure of influence is views per new thread, and the length of each new thread could also be factored in (ie: maybe views per 1000 words or 10,000 letters posted per author. I'm just making up these numbers of course). Or maybe a combination of these things.

    Posting more things can naturally lead to more traffic, but that's not necessarily a measure of how influential someone is.

    Then again, posting nothing tends to get people to neglect the blog, and consider it abandoned, and so people will stop regularly visiting it and over time they will ignore it when it does post. So more posts could indeed lead to a larger audience and hence more 'influence'.

     

    reply to this | link to this | view in chronology ]

  •  
    identicon
    TooLazyToSign, Jun 4th, 2011 @ 1:55am

    AC Grumble

    Ok, I admit to posting AC because I'm lazy, I don't really reply much and I'm too lazy to get another secure, 9-14 digit nonalphanumeric password to remember.

    Still, lately I've started at least using the same handle because of one particular AC:

    "Somebody sounds peeved that they don't have a whole hell of lot of followers on twitter..."

    This voice is familiar and seems to take the position that 1. Mike suxor 2. Anything mentioned on Techdirt suxor because Mike suxor.

    I mean, is this a former jilted lover of Mike's? An abandoned child who is now acting out the Oedipus Complex here? A Contrarian-bot? Reminds me of a two-year-old that suddenly discovered the word "No".


    So, open comment to Mr AC Grumble:

    1. I skip AC posts now that smell like you. I don't read you. You're wasting your breath.

    2. If you hate Mike so much or TD so much, why don't you just... go read something you like? I mean really! It's so 4th grade.

    3. Sounds like somebody needs their nappytime! *mock stern hands on hips look*

    OK, sorry TDers for venting in forum. I'm just bored with this chimp.

     

    reply to this | link to this | view in chronology ]

  •  
    identicon
    Bryan O'Doyle, Jun 4th, 2011 @ 6:51am

    Makin' De Link

    So why is it a bad thing if vanity leads to a fall?

     

    reply to this | link to this | view in chronology ]

  •  
    identicon
    staplegun, Jun 4th, 2011 @ 7:31am

    I agree using activity to indicate influence is less than perfect. But Google's page rank does exactly the same thing - uses page link activity to indicate page quality. It's pretty hard to measure what goes on in people's minds (influence/page quality) unless you poll them. Activity is a rule of thumb to get a ballpark figure.

    Page rank is less than perfect, but a lot of people seem to be pretty happy to use it and get reasonable results. I expect Klout, et al. will achieve similar results, though I think they could do with some further refining of their algorithm to increase its plausability.

     

    reply to this | link to this | view in chronology ]

  •  
    icon
    Jeni (profile), Jun 4th, 2011 @ 8:47am

    Echo...

    In my head - Mike, this is why I enjoy reading your writings SO much - not once have you failed to echo a thought or sentiment of my own that I can only sit here and think (and too often get upset about with no outlet).

    Anyway, this has been an issue that has driven me batty for years. Now it's costing me web site rankings for what I do online since Google's Panda update. I am NOT into the "social marketing" scene - now I'm paying for that. Literally and dearly.

    Thankfully I have some resources that will hopefully allow me to get by but suffice it to say this has greatly affected my ability to give and to donate and to SHARE as I've always so loved to do and that, in turn, is so frustrating. For example, I do a local Dial-A-Prayer site and love to donate the hosting/domain; I had to ask for the $ this year. I hated that. I REALLY hated that.

    You see, I would far more like to work than sit around "socializing". If I sit here and "socialize" all the day long, I can't work - am a one-woman operation here and can only stretch myself so thin. I would far rather work on quality content and good web design skills that need constant updating as anyone who's done that work knows but now I find myself babysitting "twitter" and considering other avenues should I ever have the time. But it goes against everything in me and I can't afford - especially now - to outsource this.

    And really sorry for rambling but please just a sec or two more - I was surfing a site called freelancer.com to see what the deal is and what do I see? People PAYING others to write content. People paying others to tweet, to post FB profiles and get "fans" - NONE OF IT IS EVEN FOR REAL; IT IS ALL FOR "SCORES" OR "RANKING" SO HOW DO WE EVEN KNOW WHO IS FOR REAL MUCH LESS HOW "GOOD" THEY ARE!?

    Sorry. Didn't mean to shout. Deep breathes...

    /offsoapbox

     

    reply to this | link to this | view in chronology ]

    •  
      identicon
      Anonymous Coward, Jun 4th, 2011 @ 11:17am

      Re: Echo...

      You see, I would far more like to work than sit around "socializing". If I sit here and "socialize" all the day long, I can't work - am a one-woman operation here and can only stretch myself so thin.

      Ah, but you see, "work" isn't the American way anymore. It's all about finding ways to get money from someone else's pocket to yours without earning it. Collectively, as a country, the U.S. is doing this by running up the national debt with plans to just pass it on to to the children of the future. (A huge Ponzi scheme). Individually, people are trying to do it by gambling (Wall Street, etc.), using legal processes (IP laws, lawsuits, financial instruments, government protection rackets, etc.), and socializing, among others.

      People paying others to tweet, to post FB profiles and get "fans" - NONE OF IT IS EVEN FOR REAL; IT IS ALL FOR "SCORES" OR "RANKING" SO HOW DO WE EVEN KNOW WHO IS FOR REAL MUCH LESS HOW "GOOD" THEY ARE!?

      Yep, it's just a big game, producing little of real value. But, hey, that's the American way now.

       

      reply to this | link to this | view in chronology ]

      •  
        icon
        Jeni (profile), Jun 4th, 2011 @ 11:31am

        Re: Re: Echo...

        You know AC, you're right. In fact, while mopping my floor just now I was thinking how hey, the big daddy gov't goons are giving away grants and bailing out big corps etc., PLUS they think the rich should pay more more more more MORE ...

        Well, the music and movie industry seem to fall under "Rich" to this little poverty peon so hey, why isn't this whole mess reversed? Why shouldn't the gov't encourage RIAA and MPAA and all their cohorts to be more giving, to pay more, to SHARE (as Odrama said to Joe the Plumber)? Makes sense to me.

         

        reply to this | link to this | view in chronology ]

  •  
    icon
    Jeni (profile), Jun 4th, 2011 @ 11:17am

    Exactly, Marcus. The whole thing is just silly IMHO.

    I have a twitter account that I use for personal interests and that I enjoy and could give a hoot about "influence". And I just "follow" people with the same interests because everyone finds different stories or what-have-you, not because of any "influencing" factor.

    It just all seems as silly to me as being "popular" in High School.

     

    reply to this | link to this | view in chronology ]

  •  
    icon
    Jeni (profile), Jun 5th, 2011 @ 6:11am

    Socializing W/O Fraud

    Run a small local forum, keep it totally legit, clean and troll free and see how many real people actually join and participate to give you all this 'influence'.

    Social networking has been taken over by hacks, trolls, fakes, $$$ and in some cases, outright sicko's who can literally harm others - and have.

    Yet, we're expected to be "influential" in the social networking scene.

    Well hell, I might as well go join the Hell's Angels then. At least I'd have "real" people in my "network" who make no bones about who they really are.

     

    reply to this | link to this | view in chronology ]

  •  
    identicon
    Aaron@Biebert, Jun 5th, 2011 @ 2:05pm

    Activity has something to do with Influence

    Thanks for the good discussion starter Mike. I would just like to add my two cents that without activity you have no influence. It's not really a debatable fact.

    I think if someone can post an engaging bit of information once every 20 minutes on Twitter, twice a day on Facebook, and once a day on Linkedin, they will have more influence than someone who posts once a month. I doubt anyone would debate that.

    On the flip side, I think that too much is too much and people will get tuned out, blocked, or unfollowed.

    It's common sense and I'll bet that Klout has that figured out...one of the reasons some folks who tweet too much see their scores go down.

    I would be careful of bashing Klout and other metrics just for the sake of saying something. It sounds like those who bashed FICO scores back in the day, even though they obviously have some relevance. I think Klout has relevance.

    Everyone understands that there are limitations, but I for one appreciate having an easier way to filter people in mass.

     

    reply to this | link to this | view in chronology ]

    •  
      icon
      Jeni (profile), Jun 5th, 2011 @ 3:57pm

      Re: Activity has something to do with Influence

      Saw this on Twitter:


      What Would Have Cost $1000.00 And 30 Weeks Of Work, You Can Now Acquire Within Minutes And Jump Start Your Way To Obscenely-HIGH Profits! Grab The Private Label Rights & Master Resell Rights To 900 High Quality Articles And Multiply Your Income Streams For Free! http://ow.ly/1d47zA


      Example of how people are getting their "klout" and I bet all of that so-called "article content" has been stolen off sites where someone thought working hard was the right way to go, and can't rid herself of that bad habit of a work ethic to TWEET TWEET.

      Grief what IS this world coming to.

       

      reply to this | link to this | view in chronology ]

    •  
      icon
      Mike Masnick (profile), Jun 5th, 2011 @ 10:52pm

      Re: Activity has something to do with Influence

      I think if someone can post an engaging bit of information once every 20 minutes on Twitter, twice a day on Facebook, and once a day on Linkedin, they will have more influence than someone who posts once a month. I doubt anyone would debate that.

      I would. Influence isn't how often you do something. Someone can be influential and only speak up every so often. Warren Buffet is influential, but doesn't speak up that much.

       

      reply to this | link to this | view in chronology ]

  •  
    icon
    Shawn (profile), Jun 6th, 2011 @ 6:30am

    This is a common misconception made. I remember running into this very problem in college only instead of "influence" it was "leadership". I was denied a RA position becuase I didn't talk enough in the group interview process. What I said has great impact but no I did not just talk to hear my own voice.

    Just like here more activity does not equal more influence. It can increase the chance to be influential but also increases your chance to irrelevant. I find that when people mindlessly tweet that it becomes annoying.

     

    reply to this | link to this | view in chronology ]


Add Your Comment

Have a Techdirt Account? Sign in now. Want one? Register here
Get Techdirt’s Daily Email
Save me a cookie
  • Note: A CRLF will be replaced by a break tag (<br>), all other allowable HTML will remain intact
  • Allowed HTML Tags: <b> <i> <a> <em> <br> <strong> <blockquote> <hr> <tt>
Follow Techdirt
A word from our sponsors...
Essential Reading
Techdirt Reading List
Techdirt Insider Chat
A word from our sponsors...
Recent Stories
A word from our sponsors...

Close

Email This