Disprove Yourself

For all the talk on how marketing is as far removed from science as a discipline can be, marketing and advertising progress and evolve in much the same way as any field of science. Falsifiability.

I’m coming up to the ‘one year’ mark of being a planner, so I thought I’d impart as much wisdom as a 22-year-old psych grad can ever hope to, and talk about being wrong.

The single most useful thing I’ve learnt doing planning is just how necessary it is to become comfortable with being wrong, and more importantly, to accept it and move on when it happens.

This is where advertising and science are alike, because advertising isn’t a field where you learn by diving headlong into the murky waters of wrongness to try and ride some current of truth. Because that current is probably not truthful at all, and is rather a mix of cognitive biases shaped by identity and past experience.

Not that there’s anything wrong with that, but relying on being ‘right’ all the time is how bullshit spreads. Because when you invest too much in being right, and someone challenges you, you’re probably going to do everything you can to reaffirm the part of your identity that says you’re right, and deny the contrary.

It’s a pretty famous bias.

So I like to get comfortable with being wrong instead. I’ve found it easier to learn by being less wrong about things I was probably wrong about to begin with.

Marketing and advertising, like science, are more akin to a giant fucking boulder of bullshit. And the truth is not a nougat centre, but a smattering of gold dust spread haphazardly throughout it.

Incidentally, the best way to find any trace of gold dust is to gradually chip away at the boulder, removing excess bullshit, until you find a couple of measly grains, upon which you can proclaim: “oh look, there’s something. I’m now less wrong”.

All sciences, including the soft ones, operate by this principle. Hypotheses are developed and tested through falsifiability.

For instance, if I wanted to test the relationship between regurgitating bullshit and being a douchecanoe, I would first outline a null hypothesis, which would state that there is no relationship (correlation) between such variables.

The testable hypothesis would be the alternative hypothesis, which would state: yes, there is some degree of correlation (and maybe even causality). I would then construct an experimental design that would either reject the null hypothesis, or fail to reject it.

It’s a common misconception that science proves things. It doesn’t. It can only disprove things. Namely, it can only disprove that there is no relationship between whatever concepts, ideas, or variables you’re testing.

And if it’s good enough for science, it’s good enough for marketers and ad men too. Better to accept being wrong and strive toward finding avenues of less-wrongness, than to cling to pervasive concepts of absolute wrongness (no matter how attractive they may be).  

No one likes a bullshit artist.

a.ce