But little Mouse, you are not alone,
In proving foresight may be vain:
The best laid schemes of mice and men
Go often askew,
And leave us nothing but grief and pain,
For promised joy!
- Robert Burns, “To a Mouse”

It’s indisputable that social media has enabled effective viral marketing. There has never been greater opportunity to generate positive word of mouth about your product or service. Done properly, some level of success is more or less guaranteed. Where things get a little less certain is when the goal is to develop a viral promotion that is intended to become a huge hit - one that multiplies geometrically across the internet.

Purportedly sophisticated agencies spend an enormous amount of thought, time and money trying to come up with the next big viral hit. And then stand by puzzled why theirs fizzles while some kid uses his phone to record a one minute video that becomes a hit viewed by millions on YouTube (can you say Star Wars Kid?). Why is it such a hit and miss endeavor?

Much of what we do successfully comes from experimentation, analysis and trial and error. In many situations we are able over time to identify the things that work and the things that don’t work. This is especially true on the internet as we are able to gather so much analytical data. But not when it comes to trying to manufacture a viral marketing top hit.

With viral marketing it’s almost impossible to know what will take the ‘net by storm (or storm the ‘net) and what will provide a lower level of success and then fade away. The reason? We’re not able to gather meaningful data from failures or successes that can then be used to design the next successful viral promotion - thanks to the ways in which nonlinearity and its first cousin chaos theory play a role.

Nonlinearity

Nonlinearity (sometimes called disambiguation) describes a situation that has a disproportionate cause and effect - fancy terms for something familiar to us all. The classic example used to illustrate nonlinearity is a pile of sand. I’m going to use kid’s blocks as my example.

Lilly, my 4-year-old granddaughter, and I like to build towers out of a set of cherry wooden blocks (I relate extraordinarily well to 4-year-olds as part of my brain seems to be stuck at that age level - something I’m grateful for as you can learn a lot from a 4-year-old). We build a base with a number of blocks and then start stacking them one on top of the other. Lilly is good at this - she lines them up and then places them carefully. As a result we are usually able to build a pretty tall tower. We know we’re done when one last block manages to crash the entire tower. The action of just one component manages to affect them all - to mix metaphors “the straw that broke the camel’s back.” That’s nonlinearity.

If we go the other way, where something gains momentum on the upside as opposed to collapsing, we might call it going viral. What Malcolm Gladwell in his eponymous book refers to as a tipping point: “the levels at which the momentum for change becomes unstoppable.” Gladwell defines a tipping point as a sociological term: “the moment of critical mass, the threshold, the boiling point.” That’s also nonlinearity

The challenge in trying to design a viral hit (let’s say a viral video) is that not only is there no way to predict whether or not this point will be reached, it’s impossible to predict or identify the (boiling) point at which it would have been reached if it failed - or the (boiling) point at which it took off if it succeeded. Which means there are no meaningful data points to use as reference when trying to design your next hope-to-be viral video.

Chaos Theory

Chaos theory (sometimes more intuitively called nonlinear dynamics - see note at end of post) compounds the challenge substantially. Chaos theory describes the phenomena whereby a small input can lead to a disproportionate response. Which makes it very difficult to calibrate the initial small input given that very small differences in the beginning can lead to huge differences further down the line.

Chaos theory was first discovered by a meteorologist by the name of Edward Lorenz. In the early ’60s he was working on a computer program designed to help predict the weather. With the modest computer power available at that time it took along time to run his programs. To save time, he decided to run one sequence from the middle rather than the beginning. Here are the two curves:

A Starting Difference of .000127 Produces Totally Different Results!

The original curve was produced by a value that had been carried out to six decimal places. For the second curve Lorenz had entered a value carried out to three decimal places. In the original sequence the initial number was .506127 and in the second it was .506. A difference of .000127 in the beginning value produced a totally different result for the two sequences.

You have probably heard of chaos theory described as the “butterfly effect.” Lorenz once described it as “… a butterfly flapping its wings in Brazil might cause a tornado in Texas.”

The phenomena described by chaos theory makes it exceedingly difficult to isolate and identify which variables will have what effect - and to what magnitude - as you distribute your hoped for hit around the internet. An expression on an actor’s face, a word of dialog, a prop on the set, little things can, and will, have enormous effect when multiplied many times.

And again, if your video fails to achieve hit status, you are not able to determine why it failed in order to design a “better” one next time. There are just too many small variables with potentially huge effects for you to identify the key ones to change, which means you have not been able to acquire any meaningful data points that might help you.

Go Ahead - Try for a Hit - But Don’t Bet the Farm on It

Using social media for viral marketing is effective. But maintain reasonable expectations. Having a viral marketing production achieve hit status is largely due to luck. Sure, take a shot now and then for fun, but direct the majority of your viral (and other) marketing dollars into tactics that have predictable, quantifiable and repeatable results.

Addendum: A Bit More About Chaos Theory

You might be thinking, Why should I care about about chaos theory? Because understanding it can be very useful when you are trying to analyze a marketing campaign tactic. Being aware that seemingly insignificant differences in a variable can have enormous effects on the degree of effectiveness on a campaign gives a perspective on how important things can be on a very granular level.

Sometimes the name chaos theory makes it counter-intuitive to understand. In everyday usage the word chaos means “a condition or place of great disorder or confusion”, which seems similar to randomness, which is defined as “having no specific pattern”. But the word has a different meaning when used in a mathematical or scientific sense. In mathematics and science chaos theory is also called non-linear dynamics, which does a better job of separating the everyday usage of the word chaos from the more technical usage.

One aspect of a chaotic system that is really counter-intuitive is that it is not a random system. If the inputs are exactly the same then the outputs should be exactly the same. For example, a roulette wheel is not a random device. If the input parameters were exactly the same every time, i.e. the speed of the wheel, when and where the ball is entered, the force with which the ball is entered, etc., then the output would be exactly the same - the ball would always land on the same number. But the input always varies, if only minutely, which causes the result to vary making it look random.

Chaos theory describes the phenomena where a tiny difference in input will create a huge difference in output in complex systems (i.e the Internet) with many iterations. I know it all seems kind of abstract, but an understanding of chaotic theory can be very useful when trying to measure results in a project as it sheds light on the impact of variables.

Another reason to spend some time with chaos theory is that it will lead you to fractals:

By the way, this one was a mild hit with over 550,000 views.

---