It is sadly ironic that The Simpsons predicted the outcome of the presidential election 16 years ago, while almost all of the polls and predictions run by the media just before the 8th November 2016 were wrong. Again, I might add, having experienced a very similar phenomenon just a few months ago following the Brexit referendum.
Working in our industry, especially as a planner, I wonder how much we can actually still listen to the polls, the research groups, the quantitative studies? In a more unpredictable and uncertain world, is there still a role for forecasting and foresight? Can data ever be trusted?
Last Tuesday, on the day of the election, I attended the Future Foundation’s Trending 2017 event. On the day they revealed their rebranding to the Foresight Factory – a day when millions of Americans defied all the foresight. In hindsight, this doesn’t just feel like a bad coincidence. It almost seems symptomatic of the state our industry is in.
Don’t get me wrong. I have been working with the Future Foundation for years and intend to continue to do so. The event was a really interesting one to attend, with lots of food for thought around the evolution of conversational commerce, personality pressures in our social media driven world and the latest stuff on biohacking. Definitely enough material for another blog post and a testament to the work from companies like the Foresight Factory to inspire us all to think beyond the present and keeping an open mind for the future. Gazing into the future and thinking about what’s next is critical to what we do, and will always be something I enjoy most about my job.
However, there seems to be a more urgent question we need to ask ourselves at this point: has the way we handle ‘foresight’, research and ultimately data, put us out of touch with what actually moves the majority, or at least a big part of our society?
This might be a surprising question to ask for BBH Labs, but an important one nevertheless. In her article ‘Reality check: I blame the media’, Danah Boyd reflects on the role the media played in the election outcome and demands that “all of us who work in the production and dissemination of information need to engage in a serious reality check”. I would include the advertising and wider marketing industry, so see this as our reality check.
Here are three observations on what we can learn from the data flaws in predicting the US election and what has gone wrong when it comes to ‘data’ in our industry. As always, we are interested in hearing your thoughts.
Data itself has become the spectacle
Going back to the #TrendingFF17 conference. On the day there was an Amazon Echo inconspicuously sitting on the podium. Throughout the program, everyone in the auditorium giggled at Alexa giving us the latest polls and predictions when asked to do so. The source and content of the reassuring predictions of Hillary Clinton having a clear lead in the election almost seemed to be secondary, as everyone was still quite confident about the outcome and the technology took centrestage, or as Boyd puts it: “I believe in data, but data itself has become spectacle.”
Apart from the fact that our industry has a certain obsession with the latest gadgets, data itself and the way it is presented (in this case by a hands free, first generation AI, voice controlled speaker) tends to become more important than the actual facts it represents.
No question, Amazon Echo is a fascinating device and we love exploring what the future might hold. In this instance it was just another symptomatic reminder of how ‘the medium really is the message’ and that it is easy to overlook the validity of the data being presented through all those shiny devices. “This abuse of data has to stop. We need data to be responsible, not entertainment.” Which leads me to my next point.
Predictions aren’t properly scrutinised
No, this election might not have marked the ‘Death of data’ but in this article on filter bubbles and analysis, Kalev Letaru points out that “the mass availability of data today means we are increasingly grabbing at data and using it to produce findings without spending the time to think about the limitations and biases of the views it may provide us and especially issues like self-censorship.”
Everyone working with various forms of research and data inputs knows this. Methodologies, the size of the sample, the ways consumers respond in different environments, and the way conclusions are derived, are critical yet often overlooked or at least easily forgotten once the results are in.
This is the main reason why the media “weren’t paying attention to the various structural forces that made their sample flawed, the various reasons why a disgusted nation wasn’t going to contribute useful information to inform a media spectacle.”
The points about ‘self-censorship’ and a ‘disgusted nation’ are really important ones. It suggests that a big part of society is disenfranchised with everything that represents the establishment and the system that is working against them, including the media, corporate America and maybe the world of Marketing, brands and advertising.
In her Guardian article, Mona Chalabi points at the same phenomenon when she shared her observations working with Nate Silver’s website FiveThirtyEight and that ‘there was also a certain arrogance that comes from being part of an elite that “gets the numbers”, and an entrenched hierarchy meant that predictions weren’t properly scrutinised.’
This should make marketers uncomfortable and question their data and the research that drives their decisions. Maybe people don’t actually want your business to succeed. Maybe they don’t want to engage with a brand. Maybe they are fed up answering endless questionnaires on their attitudes and purchase behaviours. True, no one is forcing them to, but the same can be said for the polls.
The only thing I am saying is, let’s be more rigorous with our data and let’s not fall in the arrogance trap and scrutinise every prediction and conclusion. It is critical for our business to have an eye on the future and continue to ask what’s next, but we should always question trends, future forecasts and the data that lies beneath.
Mistaking foresight with insight
The most important point though is to truly listen. I know it sounds like a cliche but with all the sophisticated data sets and tools we have at our disposal, they still don’t make up for truly understanding of how people think and feel right now.
I was fascinated when I saw this Michael Moore talk. He predicted Trump’s election months ago. Not only is he from a white, middle-class background, he spent a lot of time travelling around the country, talking and listening to people and trying to understand why they would support Trump as a candidate.
The truth is, real insight into people’s behaviours has always led to the more impactful solution, whether it is an ad campaign, a newly designed service or in this case, one of the most surprising and effective (as sad as that might be) campaigns ever. Donald Trump’s campaign succeeded because it tapped into an insight – that a big part of society felt disgusted, left behind and neglected by the whole system.
So let’s not forget that data is only as useful as the insight you can gather from it. If insight trumps foresight, maybe the Simpsons are the best way to predict the future. By holding up a mirror to society, Matt Groening and the writers behind the show have predicted many things to come true over the past few years.
In hindsight it all seems so obvious, we should have listened to the Simpsons. Or as someone once said “An insight is an insight, when it is obvious in hindsight.”
Author: Achim Schauerte, Strategy Director, BBH London & BBH Labs