Home > idea generation, inspiration, viral, viral campaign > Virals – No one knows anything

Virals – No one knows anything

All virals are Black Swan. Or, why the use of past references and case studies might make you less likely to come up with the next big viral phenomenon.

I’m talking here, of course, about the ubber-virals. The ones that have been seen by millions of people, that everyone talked about, that have changed pop culture and have redefined the way we do advertising.

I’m not restricting my thoughts to online virals as now, any piece of communication can become viral – the internet is just the facilitator.

I’m also not just talking about ‘advertising’ virals, but phenomenons that have ended up being virals (Chocolate Rain, Mentos and Diet Coke, Blairwitch…).

There are only very few real virals every year. The subservient chicken, Kylie’s Agent Provocateur video, John West Salmon, the Mentos and Diet Coke experiment are a few examples of that.

What do they all have in common? They were totally unexpected, and their success is near impossible to intentionally reproduce. We’ve all analysed their success and come up with pretty solid arguments as to why they were so successful. Yet no one seems to have found the right formula to reproduce this kind of phenomenon.

But let’s go back to the Black Swan theory from Taleb’s new book:

What we call here a Black Swan (and capitalize it) is an event with the following three attributes.

First, it is an outlier, as it lies outside the realm of regular expectations, because nothing in the past can convincingly point to its possibility. Second, it carries an extreme impact. Third, in spite of its outlier status, human nature makes us concoct explanations for its occurrence after the fact, making it explainable and predictable.

And here’s his argument summarised (whole article on Moneyweek here):

Taleb argues that humans are ‘hard wired’ to see the world through the lens of the ‘Platonic fallacy.’

We look for structure where there is none and comprehension where none is possible.

Such people are condemed to live in the realm of ‘Mediocristan’. There, their understanding will be conditioned by “Platonified economists with their phoney bell-curve-based equations”.

More, they will constantly fool themselves with the “narrative fallacy” – the human drive to impose a post hoc explanation on even the most shocking events.

Let’s come back to the last point. The “narrative fallacy”. Which is, how humans beings like to find patterns where none exist, trying to make sense of the unexpectable.

If you were to ask 10 advertising gurus why the subservient chicken was so successful, you’d probably get 10 different points of view. The surprise, the innovative use of the technology, the awkwardness, the comic factor… All making for a pretty sound explanation of why this one piece of work, out of hundreds of thousands, had achieved such a grand scale success.

Yet, for all these explanations, no one seems to be able to use this knowledge to recreate similar effects.

Being able to understand and analyse great virals is in no way a factor in predicting whether a piece of communication will go viral, let alone trying to manufacture another one.

I will go further than that. I actually think it’s counter-intuitive. As soon as you have some big principles as to why communication have turned out to be great virals, you are even less likely to come up with the next one.

As I said, all the virals listed above had one point in common. Their success was totally unexpected. You could not predict that, overnight, kids around the world would start popping Mentos into Diet Coke bottles following an online video. It’s easy to post-rationalise why they did. But it in no way guarantees that you will be able to come back with the next one (especially if you are one of the two brands involved).

Some agencies or really talented people have a better feel for what will is likely to become viral. But they are just doing better work than the rest. No one has found a formula to consistently turn communication into virals. Crispin Porter and Bogusky have never been able to reproduce another viral of the scale of the Subservient Chicken. The Viral Factory has not done something as successful as the Trojan Olympics. (And I’m happy to be proved wrong), Fallon will find it hard to come up with something as hugely popular as the bouncing balls or the drumming gorilla.

There is a big difference between great work, getting numbers and people talking, and highly successful virals. Yes, having some principles help us get to better work, but not to amazing work. It is only 0.01% of communication that has a major impact on our industry and on our audience’s lives.

So when trying to turn a communication into the next big viral, don’t just look at what has been successful in the past: it will give no indicators of what will be in the future and will make you less likely to achieve it. You need to be brave and look into other areas, new directions, outside of advertising. Don’t try to predict success by comparing it with the 10 most successful virals, it just won’t work.

By looking at past examples, you are going to make contrived work that is never going to be any better than the originals you were looking at. You’ve put yourself into the wrong frame of mind. Therefore it won’t be surprising. Or unexpected. This behaviour could not have led to the drumming gorilla or balls, because they broke every single rule of what people thought a good viral was made of.

Which explains why it’s so easy to post-rationalise why something went viral (by retrospectively applying a pattern to it) yet so completely impossible to recreate one.

  1. Digicynic
    February 24, 2008 at 12:43 pm

    I had posted this in my company’s blog, which you can find here (http://www.knitwareblog.com) and got an interesting response from Toni Smith at the Viral Factory.

    So I’d post it here, as well as my reponse to it:

    Toni Smith The Viral Factory 21 Feb 2008
    Great article and one that I will share with our clients. Your invite…

    ‘The Viral Factory has not done something as successful as the Trojan Olympics. (happy to be proved wrong), ‘

    … has of course led me to place a comment. Firstly I can’t prove you wrong, Trojan had over 62 million views, and only our 2005 for Axe/Lynx called Ravenstoke has come anywhere near with 50 million. But Trojan was launched in 2003 when the internet was a very different place. There was no where near the amount of video content that there is today to contend with. Viral distribution methods have become far easier (remember a time before YouTube and 2MB hotmail inboxes!) and there’s more content than ever to share, so to break through and truly go viral just keeps getting harder.

    Therefore success measurement has changed significantly – We regularly see commercial virals launched, targeting the mainstream, that achieve 1.5 million views and are heralded as a phenomenal success by the agency. Although we wouldn’t agree with this we do regularly advise our clients that a viral is successful at around the 3 million mark as this is the point when a film has gone viral and the views exceed what’s possible with a healthy seeding budget.

    All of this backs up your main point that if you want to achieve viral success don’t look back and try to copy past successes. Our continued success relies entirely on innovative creative which comes as a direct result of our team being part of the audience that we are trying to reach – We make stuff that we would want to be sent. Our latest campaign, How we met, for Samsung (http://www.youtube.com/watch?v=dBvDm_JLEcI) has racked up 10 million views. It’s no Trojan but it’s still a modern day viral success story and that’s because we haven’t seen anything like it before.

    Toni Smith
    The Viral Factory

    Jerome Courtial 21 Feb 2008
    Hi Toni,

    Thanks for your feedback. It was something I had in mind and wanted to write down to see people’s reactions.

    Ah! Ravenstoke. I had forgottten that one. Yes, you have a good point.
    I have always thought of your agency and CP+B as the two exceptions to that rule and agree that you are prolific at creating great work.
    I had seen ‘How we met’ and thought it was a fantastic. I have also seen your other work on Samsung, flick it and the ants video. Great stuff.

    Your point I’d like to come back to is that although there is a lot more stuff competing with our audience’s attention, it has never been easier for them to share and distribute our work. So I think these two things actually cancel each other, giving even more credit to what you achieved in 2003, taken broadband penetration, no Youtube…

    But this is exactly my point. It’s easy to post-rationalise that a piece of comm has achieved the status of uber viral, like Trojan and Ravenstoke did, because of the context at the time they were launched. As in it was easier at the time, people are more difficult to entertain nowadays… I don’t think this is totally true. How could one explain that the ‘Charlie bit my finger’ video got almost 8 million views on Youtube? It’s such a random video!

    I think we (as an industry) are just very slow to evolve to what it takes to really surprise our audiences.

    I totally agree that ‘how we met’ is a modern day viral success but not exactly an uber viral. It is a great piece of entertainment, yet it follows some ‘rules’ that refrained it from going straight to the uber viral category. Home-made video, the drawing frame-by-frame… Hence making it not totally unexepected.

    The rules that we can apply to increase the likeliness of a communication going (naturally) viral, like making it extreme, funny… are also the ones that stop us from uncluttering our minds and coming up with the next ones.

    The only rule is that there isn’t one. And you are one of the few that has clearly understood that.

    That’s why you’ll do well at Cannes this year. As per usual.

  2. March 5, 2008 at 11:24 pm

    I tend to follow your great analysis, claiming upfront that you are producing a viral is presuming of the potential impact of your creative based on outdated criteria. The internet is such an evolving environment that what was a success sometimes ago is likely to be a turkey today.

    Now having said that, there are some pillars on which virals have succeeded over time: secrets, stunts, sex and stupidity. Those are ingredients which can be blended (Trojan = stupid sex) and are likely to drive your success online. But to do so, they will have to be unique. The Bravia Balls stunt was such a success because it really took everybody by surprise; Paint and Bunnies are just sequels that are surfing on the success of the first opus, and as a result they buzz around them are less obvious, and certainly declining.

    I like the fact that you mentioned that some pieces have turned into virals without having been conceived as such. I think this is one of the great thing about internet, there are people behind the displays, and they are still shaping their environment. You might achieve a certain effect by pushing them some appealing content through wise channel strategy, they might end up turn their attention to something else. Think about the success of the Pinguin Baseball Game. What a success!! And no brand behind that, not even BirdsEye or Haagen Dazs. And what to say about all the viral activity which followed the Worldcup Final and Zidane’s headbutting: games, spoof videos, cartoons, etc. UGC at its greatest glory!

  3. JacobW
    March 6, 2008 at 4:22 am

    See, I’d go further – this point is exactly the reason why every blogger and theorist out there who is basically proposing that viral mechanics are the future of marketing is unbelievably, colossally wrong. And it’s remarkable, when you think about it, just how much of the theorising rests on the idea that agencies (or amorphous masses of consumers for that matter) will be able to reliably be able to produce pass-around content that generates word of mouth.

    Success rate, as I understand it, for generating ANY positive WOM is currently averaging around 15%. I don’t think it will improve.

    What will change, however, due to inevitable commercial pressures, is that digital media will evolve to deliver reliable, engaging interruptive formats e.g. 30″ spots in IPTV delivered TV shows. Consumers want this type of content, are willing to put up with ads to get it, and it is far better suited to client needs (which are predictable results) than viral will ever be.

    So if you believe this to be true, what viral is for is not the basis of future marketing theory, but instead an opportunistic tactic that can be deployed if you’re lucky/talented enough to come up with something interesting or fresh enough to go viral.

    What we learn from gorilla and balls (and from Duncan Watts) is that the model should be to do something interesting and engaging, put it in interruptive media, then if it goes viral, so much the better.

  4. Digicynic
    March 6, 2008 at 5:54 pm

    Great response J, I agree with you. Although I don’t totally agree with Watts.

  5. March 6, 2008 at 10:21 pm

    how does the whopper freakout rate on your success meter?

  6. JacobW
    March 7, 2008 at 8:04 am

    Success Meter says…


  7. Digicynic
    March 7, 2008 at 10:26 am

    I thought it was very good. Although I think it got us (the industry) a lot more excited than it did with the audience. But again, for what it might have cost to produce, it’s a massive success and one I wish I was involved in.
    Anyway, it will win a Cyber Lion at Cannes.

  8. March 8, 2008 at 7:03 am

    I blogged about this back in May last year.


    And then wrote a bit about how it can be part of the solution for an evolved agency model over here too. Likeminds?


  9. Digicynic
    March 9, 2008 at 5:52 pm

    Hi Charles,

    Thanks for that.

    When you say, “i blogged about this’, I’m not sure what you mean though?

    I carefully read the two posts you mention.

    The first one is a great review of Taleb’s book.

    The second about an evolved agency model based on the Black Swan model. I’m not sure what your point is on that one. I’d be interested to know it, though.

    Black Swan’s an international best-seller that’s about a year old and I have no pretention to be the first one to talk about it or mention it in an advertising context.

    However, I think the points I was making are different from the ones you made. The only common thing being that we were inspired by the book. (which I read on holiday a few weeks ago). Like you, I really enjoyed it and I was inspired to write this article as I was growing tired of having people asking me what the rules of successful viral are, and always obsessing over past success to try to recreate new ones.

    I always give credit when credit is due. But in this case, I do not read your blog and I think our posts were quite different?

    Does that sound fair? I might have missed your point. I’m happy to talk about it.

  10. March 16, 2008 at 7:43 pm

    Sorry for all the analogies, just couldn’t help it…I think virals are like lightning, (as you say one offs) cracking ideas that strike once and are usually impossible to replicate so they don’t usually conform to any long term strategic planning. They’re a good idea with a contextual brand bolted on (sorry) or should I say, neatly meshed into the idea? That’s why we have ‘idea banks’ to store away our random ‘black swan’ ideas for those perfect storm days when you synthase a beautiful idea with the ideal brand that then hopefully goes on to cause electrifying engagement and wildfire WOM!

  1. March 19, 2008 at 11:05 pm
  2. April 16, 2008 at 8:43 am
  3. June 9, 2008 at 10:03 am
  4. September 10, 2008 at 4:15 pm
  5. July 22, 2009 at 4:38 pm

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: