New Push Towards Social Advertising May Miss The Fact That Influence Isn't Static

from the check-yer-assumptions dept

As more and more people are realizing that banner ads don't work very well as advertising, there's a rush on to find better alternatives. What's now getting a lot of attention is "influential" ads. News.com has two examples of new companies trying to put this into action. The first is a company called 33Across that tries to use some algorithm to figure out who is most "influential" online and target ads towards them, hoping they'll influence others. The other is a company with the most generic name for this space: SocialMedia, who is hyping up the fact that it's come up with a "FriendRank" to determine how influential someone is, and then try to get those influential people to effectively "endorse" advertisements that their friends see.

These approaches have plenty of problems, but the biggest one is the simple fact that studies are starting to show that the concept of "influentials" is overstated. Sure, people are influenced by others, but it's not because some officially designated "influential" influences them. Influence doesn't work that way. People don't trust people because they're suddenly considered influential. They trust people because they know that individual well and trust them on that particular topic. In other words, Bill could be influential on a certain topic to Jill, but won't be influential to others or on other subjects.

But, these services don't seem to do much to recognize that. Instead, they assume that people actually have some sort of universal "influential" rank. What they'll quickly discover is that this won't be very effective, because people won't be influenced by who these services think are influential. And, if anything, these efforts will decrease influence by inserting additional friction. If I were to see a friend in an ad for a product, before making me think that product is more interesting, it will make me wonder what my friend gets out of it, and whether or not he really believes in the product. I trust recommendations that come up unsolicited -- not those that are built into an ad unit.


Reader Comments (rss)

(Flattened / Threaded)

  1.  
    identicon
    some old gy, Jun 25th, 2008 @ 5:52pm

    It will fail

    This "experiment" will fail, and thats fine. There's no reason why a marketing department can't spend resources on experimenting in new methods. That's just healthy research and development.

    What you described doesn't really appear to be a "push" in that direction, so much as a "meander".

    Also, the underlying assumption, is that it might work better than nothing. The assumption does not appear to be "this is the penultimate gold mine".

     

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

  2.  
    identicon
    Anonymous Coward, Jun 25th, 2008 @ 7:06pm

    Popcorn anyone ?

    "the influentials"

    Sounds like a new Disney movie title

     

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

  3.  
    icon
    Michael Sherrin (profile), Jun 25th, 2008 @ 9:11pm

    Influencers not that influential

    These companies are focusing on what is turning out to be outdated thinking. The concept of an influential few has been challenged by some persuasive studies by Duncan Watts who shows that it's more important to have lots of people talking about your product or idea - everyone is influential over someone and somehow connected to more influential people.

    I take it a step farther recognizing the internet has expanded the definition of influentials, negating the value of a FriendRank - you want as many people talking as possible. With blogs, social networking sites, and user reviews, anybody can be easily connected with lots of people and share information, bookmarks, and products. Someone might not be very social in real life, but the internet helps them be influential.

     

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

  4.  
    identicon
    Dave, Jun 26th, 2008 @ 9:57am

    You're Right

    Mike, I agree with you quite a bit.

    However, FriendRank is not designed to find the select few influential individuals within a certain topic category. It's designed to determine, specifically, who influences you, and you alone. Of course, the process is repeated several hundred million times for each respective individual. But, this is not like PageRank in that there is one top answer for a given question and it magically applies universally to all people.

    In short, we're not out to find the world's influentials. We believe that EVERYONE online influences someone. The question instead is, "Who do they influence?" Those are the people we deliver social banners to, containing a message or information from the influencer (aka, sara, my favorite person at my lunch table). And, like you suggested, the closer the individuals are in real-life, the more trust is between them, and the better the social banners perform.

     

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

  5.  
    identicon
    Matt, Jul 3rd, 2008 @ 6:40pm

    Banner Ads

    I disagree with your first sentence. "As more and more people are realizing that banner ads don't work very well as advertising, there's a rush on to find better alternatives."

    I think they don't work because the goals are wrong. Advertisng, at it's best, gets people to like it (it does other things, but liking an ad generally means liking a brand.)

    Banner ads though, are like an Amway Salesman, always asking for the order. Which doesn't work. But that's how we measure them, and how we might always measure them since the consumer is only a mouse click away.

    Anyway, I think banner ads could work if we changed the metric. Advertising has a bunch of metrics like memorability -- if that was the metric for banner ads, they might 'work'.

     

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

  6.  
    identicon
    miami web design, Aug 6th, 2010 @ 10:47am

    web design

    interesting article definitely will take this into account for our next project.

     

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


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