The Fallacy of Virality, or Why It’s Hard to Make Change From A Like

June 18 2013
Bayard Brewin
Bayard Brewin

A big rationale for using social media is virality. It’s the belief one can change the wisdom of a crowd by changing its shared experience, without having some third party drive that change. The idea is simple: make that crowd change their beliefs by giving them a piece of content so compelling, that crowd in turn will widely share it in their own words across their own networks.

There’s a big problem with this belief: you can’t make something go viral. When content gets adopted quickly across many audiences, it’s a happy accident at the intersection of content, interest, and opportunity – not some calculated plan with clear probabilities. And when you look at everything that has to happen for content to become viral, you quickly see why shooting for virality just doesn’t make sense:

  • Most arguments have limited appeal. The idea of explosive sharing assumes lots of people will love your content and be willing to put their own brand on it. In reality, most people only share content within a small range of interests; that content represents a tiny fraction of all the content they see; only a tiny fraction of their followers will share it; and the vast majority of sharing takes place on just one network. That’s a fraction cubed! Since most messages have really limited appeal to begin with, these numbers make virality unlikely.
  • The share window is short. Email usage is rapidly declining among younger audiences. Far more sharing now happens across social networks in every age group. Memes now come and go in hours (even minutes on Twitter). If your content is lucky enough to go viral, you probably don’t have the ability to respond at that pace. And if you can’t convert exposure quickly into something sustained – say, inbound texts or clicks – you’ve lost momentum you can’t repeat.
  • People share stuff that proves their beliefs, not challenges them. Most people would share content that challenges their beliefs – provided they’d been unexpectedly exposed to it. Unfortunately, people tend to share content from a common set of frequently-used sources, which tends to focus shares around ideas they already talk about (confirmation bias). Sometimes this sharing is favorable (they agree with your content), sometimes not (they ridicule it). For others in their network, the net effect is no change: with no new perspective or argument being offered, there’s no reason to change how (or even if!) the recipient will pay attention.
  • Sharing happens once. Even if a recipient wasn’t aware of a viewpoint, their awareness only spikes once with a share. It quickly fades because the share’s source has no way to drive frequency with individual recipients, or give them an easy way to rehearse a new belief or behavior. You need something other than the viral content – something you can directly control – that can continue this dialogue (PR, advertising, etc.). Thing is, once you’ve generated enough frequency through other means to make a difference, it’s almost impossible to tell what the original share contributed to that belief change (if anything).
  • You can’t control the meta-conversation. Once content’s out there, you can try to influence how people share it, when and with whom; there are lots of (usually expensive) tools that give readers easy ways to share, and communicators easy ways to document sharing. But you can’t actually control what people do with content. For one thing, it doesn’t take long for widely-shared content to drive conversation about the content itself. Call this the Snopes Rule: people start questioning the content’s validity, the intentions or credibility of its source or promoters, etc. Plus, cross-commentary between influencers (especially hashtag hijacking on Twitter) tends to create its own traction among followers – and that can drown out the voice of your original post.
  • Many viral appeals are extreme – but readers aren’t. Lots of folks try to cut through the social clutter with a viral appeal based on high-arrest language or visuals. But unless you’ve got a matching strong action, all you’ve done is yell. Also, many of those appeals tend to be negative, simply because it gives the appeal a sense of urgency. Thing is, positive messages tend to have higher click rates, but they’re usually not very urgent.
  • Your content’s power to drive change may not be what you think. Take, well, climate change. A lot of K Street-based “green” groups use the same strategy to create viral campaigns: paint the Arctic as some sort of frozen ASPCA where seal pups, polar cubs, etc. need to be rescued with a Like or Share. But this strategy only really validates a recipient’s love of cute animals. It doesn’t actually commit the reader to change public policy, or even their own climate-changing behaviors. There are just too many degrees of distance between the share and the change. By contrast, look at Greenpeace’s highly-focused campaign against Royal Dutch Shell (the company that fields the lion’s share of Arctic oil exploration). Greenpeace starts with polar bears too, but shows them as homeless mammals wandering UK Shell gas stations and offices, promoting imaginary and absurd corporate initiatives, and brand-jacking Shell’s advertising and corporate identity. This completely on-brand street theatre creates intentional conflict with authorities at Shell-branded places, which Greenpeace in turn captures on video and shares as a call to pressing action. Under this strategy, the enemy has a face you can see, the conflict feels right-around-the-corner, and the action they’re seeking is human-scale and urgent.

So is virality even worth it? And should you use it?

If you think of virality as just one more way to reach a larger goal, and treat it as more process than product, then virality can help your communications program. But again, it’s an unexpected consequence of doing the right job, not the key performance indicator of that job.

Begin with a clear understanding of your campaign’s Win – the big hairy game-change you want to achieve (don’t worry about little things like Likes or clicks just yet). Then, draw a tactical map from start to finish of what people will need to hear and see, how they should respond, what you want them to believe, and ultimately what the market’s belief (and actions) will look like when you’re done. At every point where there’s an action, insert a matching metric for just that tactic and time window (so you never lose track of what tactic created what change). Give each metric a base value, add a way to accumulate these metrics, and set up alerts in case any one of those tactics goes way beyond its base value.

Now, build a matching fast-path around the heart of that diagram. This is the path a viral event will take; it’ll move faster than anything you’ve just drawn – in fact, it’ll move faster than anything you can even manage. That’s the reason you just drew this path: to figure out what extra resources you’ll need to support this spurt, so you can line them up and set them loose when a viral condition arises.

Obviously, good program planning like this requires prior experience in running the viral rapids, along with access to monitoring resources that are beyond most folks’ reach. Sage Communications can help you design a program plan that makes sense for what you do, how you work, where you’re headed, and who you’re taking along. We let you offload monitoring and response tasks, so you can stay focused on your mission – especially for times you get too much success. More importantly, we can help you build those successes in future campaigns while staying aligned with your strategic focus.


I am agreed that if a

I am agreed that if a recipient wasn’t aware of a viewpoint, their awareness only spikes once with a share. It quickly fades because the share’s source has no way to drive frequency with individual recipients


So many good points, but the very useful advice in here is to be prepared if the happy accident of viral content happens to your campaign. Thanks!

Post new comment

The content of this field is kept private and will not be shown publicly.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Allowed HTML tags: <a> <em> <strong> <cite> <code> <ul> <ol> <li> <dl> <dt> <dd>
  • Lines and paragraphs break automatically.

More information about formatting options

Comments are welcome, since the web is a collaborative medium, but spam, profanity, and politically-charged comments will be removed.

By submitting this form, you accept the Mollom privacy policy.