Creative Analytica

Cambridge Analytica made headlines after whistle-blower Christopher Wylie revealed that the company had manipulated millions of voters in the 2016 U.S. election. They had exploited Facebook data “And built models to exploit what we knew about them and target their inner demons. That was the basis the entire company was built on.” according to Wylie. Their ‘secret sauce’ was using illegally obtained Facebook likes and interest data to predict where 87 million people fell on the ‘Big Five’ personality traits scale: openness, conscientiousness, extraversion, agreeableness, and neuroticism. This targeting data would then be used in targeted advertising campaigns to sway voter’s affiliations. Take the issue of gun rights. Extroverts may prefer a pro-gun ad that talks about hunting as a group activity, whereas neurotic people might respond to a message emphasizing that the Second Amendment will protect us. “You say the same thing but with two very different messages,” says Tom Dobber, and expert in microtargeting at the University of Amsterdam. The consultancy had been active in over 200 elections around the world, including the shock Brexit result in the UK. The revelations resulted in a $43 billion decline in the value of Facebook stock for their role in the data breach, and led many to question whether democracy was under attack, and how to defend it.

The public discourse on Cambridge Analytica has been focused on the data breach, and micro-targeting, because a “psychological warfare mindfuck tool” that knows “more about you than any other person in your life, even your wife” is rightly terrifying. However the notion that micro-targeting gave Cambridge Analytica some unholy advantage has at this point largely been discredited. Researchers maintain that while psychographic profiling is based on real science, the effect is small. In a study targeting 1.5 million people; the result was about 100 additional purchases of beauty products compared to no psychographic targeting. Another study found that the ‘Big Five’ personality traits predicted as little as 5% of the variation in individuals’ political orientations. In March 2017, a New York Times article said psychographics weren’t actually used by Cambridge Analytica in the 2016 election. Before the election, Republican strategists were already expressing less-than-stellar opinions of the company. Former Ted Cruz aide Rick Tyler told The New York Times that the psychographic models proved unreliable. The British Government concluded from a 3 year enquiry that Cambridge Analytica was “not involved” in the Brexit campaign, "beyond some initial enquiries made... in the early stages”.

If micro-targeting was mostly inconsequential, and Cambridge Analytica was over-selling its targeting capabilities, then what really drove outcomes in the 2016 U.S. election and U.K. Brexit vote a year prior? The answer has been hidden in plain site. It can be found in an obscure complaint by Chris Wylie, Cambridge Analytica whistle-blower: "If you work in a creative team, you shouldn’t have to do something because an algorithm said so.”. Why is the former Director of Research for Cambridge Analytica complaining about creative freedom? In a talk given by Alexander Nix, former CEO Cambridge Analytica, he said “today we can use Big Data to understand exactly what messages each specific group within a target audience need to hear”. We’ve been focused on micro-targeting, when in reality creative testing was doing the heavy lifting. The Trump campaign tested tens of thousands of variations of ad creative on Facebook, an approach Facebook executives “showered praise” on. Memes such as "crooked hillary", "drain the swamp”, “make America great again” were tested first on social media, with those that resonated amplified by news media outlets hungry for controversy. Dominic Cummings, who led the Brexit campaign, also credits creative: “It is hard to overstate the relative importance in campaigns of message over resources”. We’ve been worried about targeting, when the real danger comes from memes.

About the Book 📕

← Back to Essays 📝