Facebook and the influentials

Posted by on Aug 5, 2009 in Uncategorized | 12 Comments

TimeMostInfluential

Pic c/o Prospect.org

Nice lecture here about an analysis of Facebook phenomena which demonstrates what Duncan, Paul, Alex and I have all been saying for a while:

[Mostly] things don't start with special individuals – it is the rest of us and our willingness to adopt something that we see around us that really matters in the spread of behaviours and ideas through populations.

Counter-intuitive I know (and not something very flattering to us self-styled innovators and influentials) but that's what the data suggests. And whatever you'd like to believe about the world and your role in it, it's always good to look at the data to check these ideas out.

Footnote: nice anecdote last night from Goodby's on the same subject The famous Sasquatch dancing man youtube vid was recently sent by a creative director to a planner as if it were an example of "here's to the crazy ones!" (or some such) – a celebration of brave creative heroics, ie. as if what mattered was the first couple of guys.

As we've repeatedly said, the lecture above confirms and Black Swan underlines, it only seems like they're important after the fact – but few of us work backwards through time (Timelords, excluded). Most of us would not be able to tell that these guys – as opposed to any of the other bolloxed groovers around the festival field – would be the seed of a great congregation….

12 Comments

  1. Matt Sadler
    August 12, 2009

    Hey Mark, nice post. I’m sure you’re aware of it already, but this research echoes similar findings by Duncan Watts’ (network theory scientist at Columbia University) whose work with mathematical modelling also provides evidence against the Influentials theory.
    If you haven’t seen it, have a read of my post here:
    http://infomagination.typepad.com/blog/2008/12/how-to-make-sense-of-a-complex-world-and-why-not-to-trust-malcolm-gladwell.html
    Keep up the good work,
    Matt.

  2. Mark Earls
    August 13, 2009

    Thanks, Matt. Appreciate that.
    As it happens, I know DW’s work really well – e.g. I posted about it back here http://herd.typepad.com/herd_the_hidden_truth_abo/2008/01/influencers-the.html when the row blew up over the download experiment.
    You should also check out the other guys linked to above (Paul Ormerod and Alex Bentley have both got really important stuff to add to the debate). And read the award-winning piece Alex & I wrote for Admap summarising how things spread.

  3. Jan Van den Bergh
    August 13, 2009

    Mr Sun “proves” with his study on “page fanning” how wrong it is to believe that you only need influentials – the selected illuminati – to spread your ideas to everyone else for free.
    Luckily his study and his restricted influential-concept are far away from the proven successful usage of recommenders, recommendations and reviews in marketing. But since his memes might spread like wildfire within the marketing & advertising world let me add a few words.
    Page fanning is very easily done. In a blink. Alice fans a page and Bob and Charlie see that and they also fan that page. And then their friends do the same and so on and so on. That’s it. Indeed why the hell would you need influentials to help you to take a decision whether or not you should fan the same page too. It probably takes less than a second to decide and to click. Or not to click. You don’t run any risk. It’s free. If afterwards you think twice and you feel you were wrong you leave the fan page. No harm done. Maybe nobody even noticed you were a one-night-stand-fan.
    Buying something with real money on the other hand can be very risky. If you make the bad choice you will not only feel bad and angry (angry with yourself, the brand and the sales person) but you also lost a bit or even a lot of your savings. You also lost the time you invested to select first and later to make up your mind. You maybe even lost face within your peer group. Buying is risky. The higher the price and the more visible the product the riskier it is. Buying the wrong toilet paper is less risky than buying the wrong (up to you to fill in the blank). And that’s where influentials, recommenders, reviews, brand scores play a crucial role.
    In marketing we know for 100% sure and since many years that people actively search for external and trusted sources to evaluate their initial shortlist. We know it, but not all marketers take it into account when they write their plan. We know that reviews by trusted people (those who have experience as a user or knowledge because they are a specialist) are of utmost importance. These people are the trusted sources during that evaluation stage. These recommenders don’t necessarily actively and permanently push their opinion (they can write it down on a review site e.g.) but their opinion is pulled at the right moment by those who need the additional information.
    In fanning none of it is needed. You can even become a fan of an endless amount of pages. In theory millions. You don’t have to make a choice. In this process, indeed Mr. Sun, the influencers, recommenders and other evangelist play no role at all. They’re like a matchbox when you already have a spreading wildfire. Useless. That’s why the fanning test cannot prove the non-existence of influentials in marketing.
    PS. If you haven’t read anything on this, the best you can do is start with Reichheld and the Net Promoter Score team who have been evangelizing like hell about this phenomenon. Additional recent research can be found in:
    1. “Do friends influence purchases in a social network” (Harvard Business School – Working paper 09-123 by Raghuram Iyengar, Sangman Han and Sunil Gupta is a good start to make you a true believer of the existence of influencers and even more the real influence of these influencers on the rest.
    2. Another one which goes into the same direction is “Opinion leadership and social contagion in new product diffusion” by Raghuram Iyengar, Christophe Van den Bulte and Thomas Valente (Wharton – October 2008). They also confirm that influentials exist and that targeting them in a campaign is useful since they influence more peers. Even more: they are often early adopters and heavy users i.e. that their long lifetime value is higher than of any other consumer.
    3. And the last – less “scientific” hence more readable- is the McKinsey article of june 2009 “The consumer decision journey” by David Court, Dave Elzinga, Susan Mulder, and Ole Jørgen Vetvik.
    Jan Van den Bergh
    Holaba.com.cn
    Chairman

  4. Mark Earls
    August 13, 2009

    Thanks, Jan for your excellent response.
    A couple of comments.
    First, the point I’m making here (and in the Herd book and e.g. this article http://herd.typepad.com/herd_the_hidden_truth_abo/2008/11/free-gift-influence-and-how-things-really-spread.html) is not that the idea of social influence – that each of us take the guidance of others’ behaviour in shaping our behaviour and opinions – is wrong or unimportant. Quite the opposite: I’ve written repeatedly about how social influence is the key factor shaping adoption of behaviours and that it is (as you describe) a pull rather than a push-force.
    Nor am I denying the importance of recommendation and endorsement e.g. Amazon or Cialdini’s Bathroom Towel experiment (though I would suggest that what people say is generally much less important than what they are dong). Or the usability of Reichheld’s work (whatever his detractors say).
    What I’m suggesting is that Sun’s is another piece of counter evidence to the idea that “special individuals” – the Influentials – have a disproportionate influence on the rest of us. One way of thinking about this is Ormerod’s work that demonstrates that underlying structure of social networks is most commonly NOT hub-and-spoke but rather effectively more random or clustered. Another is the cascade work of those like Mike O’Brien (archeology), Alex Bentley (anthropology & marketing), Matt Salganik and (yes) Duncan Watts (sociology & marketing) who look to describe the spread of behaviours across populations without assuming the network structure a priori.
    The important thing here is that in this broader-based literature, it is not that the hubs never appear, it’s just much rarer than we marketing folk would like to believe.
    Of course, life would be much easier for us if the Influentials hypothesis does exist but it doesn’t seem to be evidenced that often in the data. Equally it doesn’t make sense of the reality of our lives: most humans today live in a rich social soup in which “doing what the Romans do” is a more appropriate strategy than “doing what the Roman Senators/centurions do”.
    Whatever you think of Sun’s work (and I think you’re being a little harsh), I hope this articulates a broader context for it’s central conclusion.

  5. Matt Sadler
    August 13, 2009

    Thanks Mark, I thought you’d probably be well ahead of me! Got your article, looking forward to reading…
    Jan, good points about different processes for different decisions – reminded me of a paper called “Decision Watch” by Blades & Phillips that won an MRS Award in 2005. It proposed that purchase decisions are a game of snakes and ladders affected by different kinds of influencer (which they segmented into category experts, owners, advocates, role models, researchers, reassurers and professionals). I remember you writing a few pages on it in Herd, Mark. Is it still something you believe in?

  6. Mark Earls
    August 13, 2009

    Thanks, Matt
    In putting together the latest edition of HERD, I spoke at length to Stephen P and we agreed that the data now seems to suggest a very different picture. Even less clear and importantly, much more mutual but still at heart nothing like strong and fixed version of the “influentials hypothesis”
    While there are occasions when it does seem to be a good description of how influence works and how things spread, this is far from being the norm and really shouldn’t be our default setting!
    Let me know what you think!

  7. John Dodds
    August 13, 2009

    The point is that if we cannot be sure where the hubs will appear, we cannot and should not focus our efforts on locating them.
    Rather we should ensure that our actions will reach and be speadable by whomever turns out to be an influencer this time round.

  8. James Cherkoff
    August 13, 2009

    I think it’s more interesting to talk about passion than influence. Networked media allows people who share passions to find one another. The Facebook guy is describing patterns of passion emerging when when he talks about ‘page fanning’ – (albeit in the dry language of a network engineer). Understanding how people organise themselves around their passions in the modern era is a more interesting starting point than who is influential and who is not.

  9. John Dodds
    August 13, 2009

    Well said James.

  10. Daniel Goodall
    August 14, 2009

    @James
    I totally agree with you point about passions: genuine advocacy is a function of an individual’s emotional connection.
    Understanding how people organize themselves around passions is an exciting extension to the Social Objects concept.

  11. Samina
    September 1, 2009

    Thanks Jan for the useful articles. After reading a few articles on influence I feel that it’s a mix between using those who are considered more influential (because as humans we for some reason feel that their opinions are important and we have a desire to mirror them) and peer-to-peer (the mass). I’m yet to read the psychological theory on my assumptions on human behaviour- article suggestions are very much welcome!
    The sociologist in me can easily see the power of herds and tribes but what I find difficult to ignore is celebrity trendsetting. Do celebrities or influentials just add fuel to the fire? Does the trend exist before but they bring it to the forefront?
    I agree with @james that passion plays a big part and I wonder if passion itself becomes contagious. If you tell me with genuine enthusiasm that something is fantastic am I more likely to take it on?
    In regards to the Facebook research (by Stanford Uni) even if the spread of fan pagging may seem irrelevant compared to actual purchase decisions it still shows how an idea might spread and gain exposure on a real social network. If we’re to look at how easily people would click buy, it’d be interesting to see how Facebook gifts spread over a particular event (e.g. Valentines).

  12. Samina
    September 1, 2009

    Thanks Jan for the useful articles. After reading a few articles on influence I feel that it’s a mix between using those who are considered more influential (because as humans we for some reason feel that their opinions are important and we have a desire to mirror them) and peer-to-peer (the mass). I’m yet to read the psychological theory on my assumptions on human behaviour- article suggestions are very much welcome!
    The sociologist in me can easily see the power of herds and tribes but what I find difficult to ignore is celebrity trendsetting. Do celebrities or influentials just add fuel to the fire? Does the trend exist before but they bring it to the forefront?
    I agree with @james that passion plays a big part and I wonder if passion itself becomes contagious. If you tell me with genuine enthusiasm that something is fantastic am I more likely to take it on?
    In regards to the Facebook research (by Stanford Uni) even if the spread of fan pagging may seem irrelevant compared to actual purchase decisions it still shows how an idea might spread and gain exposure on a real social network. If we’re to look at how easily people would click buy, it’d be interesting to see how Facebook gifts spread over a particular event (e.g. Valentines).
    I end up looking a bit cross-eyed when reading into this because in my opinion each social network is unique. The activity and strength of ties will vary across networks and so I question whether we can apply a standard model to it. We are indeed complex beings!