Tipping-point skeptic says that super-Influencers are overrated

Clive Thompson has a long feature in Fast Company about Duncan Watts, a researcher who disputes Malcolm Gladwell's vaunted "Tipping Point" model of how social ideas spread in society. His experiments and computer models suggest that the spread of ideas is a lot less linear than good ideas in the hands of influential people:
Watts wanted to find out whether the success of a hot trend was reproducible. For example, we know that Madonna became a breakout star in 1983. But if you rewound the world back to 1982, would Madonna break out again? To find out, Watts built a world populated with real live music fans picking real music, then hit rewind, over and over again. Working with two colleagues, Watts designed an online music-downloading service. They filled it with 48 songs by new, unknown, and unsigned bands. Then they recruited roughly 14,000 people to log in. Some were asked to rank the songs based on their own personal preference, without regard to what other people thought. They were picking songs purely on each song's merit. But the other participants were put into eight groups that had "social influence": Each could see how other members of the group were ranking the songs.

Watts predicted that word of mouth would take over. And sure enough, that's what happened. In the merit group, the songs were ranked mostly equitably, with a small handful of songs drifting slightly lower or higher in popularity. But in the social worlds, as participants reacted to one another's opinions, huge waves took shape. A small, elite bunch of songs became enormously popular, rising above the pack, while another cluster fell into relative obscurity.

But here's the thing: In each of the eight social worlds, the top songs--and the bottom ones--were completely different. For example, the song "Lockdown," by 52metro, was the No. 1 song in one world, yet finished 40 out of 48 in another. Nor did there seem to be any compelling correlation between merit and success. In fact, Watts explains, only about half of a song's success seemed to be due to merit. "In general, the 'best' songs never do very badly, and the 'worst' songs never do extremely well, but almost any other result is possible," he says. Why? Because the first band to snag a few thumbs-ups in the social world tended overwhelmingly to get many more. Yet who received those crucial first votes seemed to be mostly a matter of luck.

Reading through the piece, it seems to me that Watts is primarily concerned with those ideas that don't "break out" and swamp the mainstream -- if you're going to have a modestly successful idea, how can you increase that modest success two- or three-fold? It may be that the combination of a hugely influential person; a simple, easy-to-communicate idea and a receptive market can go viral and be on everyone's lips in a few days. But what if you've got a hard-to-communicate, subtle idea and you want to maximize its spread? Link


  1. Sure. I can name about thirty artists who broke out into the mainstream solely on the basis of a hit single, while their album quality remained steady (or even fell).

    Even groups with no concrete “hit single” (think Arcade Fire) explode from a couple of glowing reviews.

    I won’t lie – critical acclaim often leads me to investigate a band far more patiently than I normally would. It’s difficult to try something new when your effort hasn’t already been validated by another.

  2. “But what if you’ve got a hard-to-communicate, subtle idea and you want to maximize its spread?”

    Seems obvious – you do some or all of the following:

    1. Make your idea easier to communicate
    2. Make your idea less subtle
    3. Hope luck is on your side

  3. I’ve always been intrigued by the rise of euro-synthpop in the US in the early 80s. MTV didn’t have enough domestic videos keep in rotation, so they had to play stuff like Der Kommissar. I doubt that people would have gravitated music like that normally, but the forced exposure and the perceived coolness of MTV made it succesful.

  4. But what if you’ve got a hard-to-communicate, subtle idea and you want to maximize its spread?”

    use bribery

  5. You can see this trend in other social websites like Digg or even the threads in Slashdot.

    On Slashdot the first posts get seen the most and if they are reasonably good they get modded up. The key to a 5 rating in Slashdot isn’t being the best but being reasonably good and being first.

  6. @ #4

    Bribery eh? Even if it’s a simple idea, well explained, even bribery doesn’t work – my favourite example is stamps.com – $20 in free postage and $$$$ later they died.

    After a quick spell in the special Rehab place Companies who’ve snorted huge amounts of Venture Capital go; they returned looking much better and even appeared on here: http://www.boingboing.net/2005/04/27/second-try-for-perso.html

  7. Not sure why anyone is surprised at this. Gladwell’s book is not a science text, unlike Blink it’s not a summation of pre-existing research, and Gladwell is a journalist (with a background in psychology) not a psychologist. The tipping point is a fun little meme, a way of construing certain kinds of social networks – I don’t think even Gladwell would suggest it has empirical validity.

  8. IMHO, memes spread in a fractal mode, but can, at times spread in a linear fashion. Depends on the ideas and how they’re being pushed. Global warming is an example of how ideas spread like a fractal, growing, branching, changing a bit while keeping their core geometry.

  9. This model sort of mirrors the voting system at http://www.electricsheep.org Animated fractal processes, “sheep”, get downloaded into the computer client’s screen saver program after they become rendered by the masses and assembled on the host server. Their life and breeding potential with other sheep are then determined by a voting process rating system where as the sheep is displayed you can press a thumbs up or thumbs down. You can also rate sheep via the website. Higher rated sheep stay in the gene pool longer. However, not all sheep rendered have a chance of downloading to everyone’s computer; mainly the higher rated sheep are downloaded (unless they actively seek them out via the website). So this leaves out the potential higher rating of the few that didn’t get a few thumbs up in the beginning. It’s an interesting dilemma, as many gems will go unnoticed to the major populace, however the high rated ones are generally always very good.

  10. @Gareth – True, but the article takes some heavy hitters in marketing head on. I work in that field, myself, and I find Watts’ ideas as expressed here very compelling. In fact, they echo what I’ve been thinking about marketing for some time. My POV has always been that if you have a product or something you’re trying to sell, an idea, a service, whatever, you should get it into the hands of as many people as possible and let it speak for itself. You never know who the real “influencer” is going to be, so the more the merrier, which is I think Watts’ critical point.

  11. What do you mean, “Malcolm Gladwell’s vaunted ‘Tipping Point’ model”? He was just expounding upon a thirty-year-old model created by Thomas Schelling (using a term coined, according to Wikipedia, by Morton Grodzins). Gladwell is a generally good writer, but let’s give credit where it is due– a popularizer of an idea is not the idea’s creator.


  12. This is fascinating, but the example of Madonna in 1983 doesn’t fit.

    I remember suddenly seeing on every episode of a music video show in 1983 “Borderline” repeatedly. Madonna had been seen as a potential star, so she was promoted to the saturation point.

    If this experiment were to repeat, I’s suggest another group with one outside person telling all, here’s the hit song, you have to listen to this before you hear the others. In an hour, you will listen to it again…

  13. @Backload

    I think there is a very, very good reason tha Stamps.com failed. The people most likely to find out about stamps.com had stopped using stamps! I do almost everything via the internet and have for over 10 years. My bills, my correspondence, everythigng gets where it needs to go without the postal system. I haven’t purchased a physical stamp in…let’s see…4 years. I used one stamp, and gave the other 9 to my cube neighbor.

  14. Since it’s semi-on topic I will recommend reading Connie Wilson’s charming novel “Bellweather” which is an amusing take on researching trying to understand why things become popular and how fads start.

  15. Just to be precise, the author is Connie Willis (not Wilson), and the novel is Bellwether.

    And I definitely second the recommendation for the book.

  16. It is an interesting analysis based on social simulations, but I think the premises may be flawed. Influentials are not merely socially promiscuous people, they are explicitly being watched for trends etc., so their larger coterie is also more likely to be infected by them.

  17. I have always suspected that culture is only driven by the object–the work of art, the song, the “good idea”–to a limited extent, and that what really matters the role the object can be made to play in activating and facilitating social interplay between individuals, groups, and–perhaps most of all–the idea of “the public” (“everyone,” “my fellow citizens,” “the sheep,” “the mass,” whatever you like to think of them/it as).

    The particular objects that get chosen to play this role are often chosen for reasons that have nothing whatsoever to do with their merits as ideas or aesthetic objects.

    Someone like Britney Spears, say, may be popular not because folks really like her music so much, but because she helps us play out our hopes and fears about adolescent sexuality.

    Kind of depressing, but the take home message seems to be that a lot of the people who are involved with music or wine or ideas–the people who determine the “success” of these things–don’t really care about them in themselves. They care about other people–What are they like? (Vain? Trivial? Stupid?) And what do they think of me?

  18. The Madonna example isn’t so bad, btw.

    If you work in media, you know that WAY more things are promoted heavily than actually become “hits.”

    How many Americans remember Die Toten Hosen?

    Similarly, Madonna might have turned out to be a well-promoted miss rather than a well-promoted hit.

  19. “But what if you’ve got a hard-to-communicate, subtle idea and you want to maximize its spread?”

    Make it secret. Refuse to divulge it. Charge for the privilege of even hearing about its existence. Once you’ve made it clear that nobody is good enough to know it, leak it. Why do you think that Scientology is growing so fast? It’s not because Tom Cruise makes a fool of himself in public every week. It’s because it’s esoteric ‘wisdom’ that you have to earn, by which I mean pay for, the right to discover.

  20. I never much took to Schelling/Gladwell’s ‘tipping point’ idea. It’s too pat, and smacks of post hoc data rationalization. It’s just as easy to find major memes of all kinds that can’t be ascribed to the tipping point model (as a number of people have demonstrated here).

    Certainly there is some kind of crucial nexus that is often needed for an idea to propagate rapidly but not all major changes happen fast, nor dramatically. Success does not depend on speed.

  21. Ok, being something I’m greatly interested in, I did RTFA. Some thoughts on the subject:

    Influencers did dramatically increase the extent of the spread when compared with the average person. This study doesn’t at all disprove viral marketing– it merely calls into question how and why it works.

    Having actual merit will always be vitally important to spreading a meme.

    Reward word-of-mouth in a creative way if you can reasonably do so. Make it damn easy and enjoyable to spread your meme, and your audience will help you.

    The most powerful influencers are also often the easiest to get to pass around your meme. I’m referring to journalists both amateur and professional, of course. Re: slashdotted.

    If you can get influencers to spread your meme, then you should do so. But it is always better to have a normal person who is a real fan than an influential person who only kind of likes your meme, and it is usually easier to cultivate real fans than to convince people who are invested in ‘being cool’.

    When it comes to judging the power of an influencer, it doesn’t matter how many people a particular person knows or how well regarded that person is. It only matters how many people someone will tell about your meme and how interested they are in it.

    Don’t underestimate the power of irreverence and humor in getting your meme spread.

    There is no substitute for being original when it comes to viral marketing.

  22. This exact little experiment is being played out in the primary elections.
    I would prefer not to have Madonna for president.

  23. @#3 OCCUPANT

    You’re right, about MTV in the early 80’s. If the same Euro videos weren’t in heavy rotation we wouldn’t have even had a “New Wave” era in pop music.

    I think a one hit wonder would have been a better example than Madonna. Yes all of the stars must align for an artist to breakout, including luck, but Madonna essentially brokeout several different times after it appeared her career was over. There is something else in play there.

Comments are closed.