IN DEFENSE OF CREATIVITY
The discipline of marketing has changed dramatically over the course of the last few decades. With the rise of powerful computing, data aggregation, reporting and visualization tools, and finally AI-driven action, we’ve reached a point where the SCIENCE of marketing has overtaken the ART of marketing. Virtually every conversation I can recall over the last few years with CEO’s, investors, and even fresh-from-college marketing interns begins and ends with a focus on metrics-driven marketing. For good reason, we have access to data algorythms that can not only ‘accurately’ measure the performance of marketing campaigns, but in some cases predict the outcome of future efforts - when variables hold steady.
That caveat is important, however: “when variables hold steady”. Turns out most marketers are not the best scientists, and don’t apply a rigirous scientific method to their performance metrics. A/B testing of website content often fails to control for the traffic source, content of the source ad/link, time of day, external news/influences in sentiment, geography, platform, audience demographic and more. Accounting for and controlling all variables is what makes an experiment sound, and while it’s virtually impossible to control for all variables (especially without the assistance of big data computing), many marketers often control for only one: results. They are pressured by the business and the current scientific focus of marketing to justify all decisions with metrics - hard numbers that show that A outperforms B and therefore we are optimized to perform best.
While I could go on for days about all of the faults in the ‘scientific method’ applied by 99% of the marketing programs out there, and I’m not denying that metrics matter and should be used to govern decisions, there’s a key point that gets lost in all of this focus on (often faulty) science.
In order to perform an A/B test, you must first have an A, and a B.
If it’s ad copy, someone first has to write A, and write B, and then test the comparative performance of each.
With the focus of marketing shifting to scientific performance, many businesses have forgotten to acquire talent that can create A, B, and subsequently C, D, E etc. Meanwhile, as the tools to automate measurement and testing improve, the creative talent is actually more important, because no computer can yet simulate the spontaneous synthesis of one random concept to another concept to CREATE something new. There are no truly NEW ideas, just new combinations, and right now humans are infinitely better at fusing disparate concepts into new things. A book from over a decade ago can make this argument far better than I can; Daniel Pink’s “A Whole New Mind” from 2006 makes the case for why creative thinking is more valuable than analytical thinking as we automate more and more via technology.
“Today, the defining skills of the previous era - the “left brain” capabilities that powered the Information Age - are necessary but no longer sufficient. And the capabilities we once disdained or thought frivolous - the “right brain” qualities of inventiveness, empaythy, joyfulness, and meaning - increasingly will determine who flourishes and who flounders.
— Daniel Pink - A Whole New Mind
Computers can perform quantitative analysis and tell us whether A or B is more effective far better than any human, but right now only a human can conjure A and B from the synthesis of the random experiences of business, music, art, sports, social interaction and more.
