Searching for Consumer Insights is Not Like Mining for Gold


There’s an interesting article by Matthew Kassel in The Wall Street Journal that outlines a trend in the advertising industry. Although agencies have used data to buy digital ads for years, they are now using data to shape the creative process. As Kessel writes:

Ad agencies’ increasing reliance on data is inverting the traditional creative model, where the teams dreaming up ad campaigns use their gut instincts to tap into the zeitgeist. Now, thanks to a preponderance of highly specific information about consumers collected from sources such as social media, surveys and search engines, more agencies are making creative choices based on what the data dictate.

One example from my agency involves a major car company we pitched last year. During the pitch our team discovered that, contrary to the overwhelming sentiment that polarization and partisanship are increasing while centrist views are decreasing, about 80% of Americans respond “yes” to the question, “Do you consider yourself politically moderate?” We used this data point to position the car company as a brand that represents moderate middle Americans, knowing that most people identify as such. Although we didn’t win the pitch, the data point provided a compelling perspective that inspired and justified the subsequent creative work.  

And yet, Kassel falls short on two points. The first is that agencies do not have “a preponderance of highly specific information about consumers.” Tech companies such as Facebook and Google do. Although agencies can access some of this data with a fee or subscription, these data giants know far more about who you are, where you live, what you buy, and how you tap and click. As the media commentator Ken Auletta says in an article published in New Yorker last year:

In the advertising world, Big Data is the Holy Grail, because it enables marketers to target messages to individuals rather than general groups, creating what’s called addressable advertising. And only the digital giants possess state-of-the-art Big Data.

Second, although agencies use data to shape the creative process, Kassel commits what I call the gold-mining fallacy. The gold-mining fallacy holds that data is “out there” waiting to be discovered. Take the previously mentioned statistic about 80 percent of Americans identifying as “politically moderate.” Instead of using scales that locked respondents into an existing framework—conservative-liberal or Republican-Democrat—the surveyor cleverly forced respondents to select “yes” or “no” to make a point: regardless of political ideology, Americans want to identify and be thought of as moderate. Was this statistic “out there” waiting to be discovered? Not really. More likely, the question spurred the attitude into existence.

The gold-mining fallacy stems in part from an orientation taken by social scientists such as Emile Durkheim, who famously studied differing suicide rates among Protestants and Catholics in 19th century Europe. What makes Durkheim famous—besides his insights into how stronger social control among Catholics explained lower suicide rates—was the subsequent methodology he helped to formalize.

According to Durkheim, observation must be as impartial and impersonal as possible, even though a "perfectly objective observation" in this sense may never be attained. A social fact must always be studied according to its relation with other social facts, never according to the individual who studies it. (Source)

From Durkheim’s perspective, the researcher who thought to ask the question “Do you consider yourself politically moderate?” and forced respondents to answer “yes” or “no” could be blamed for committing a methodological error (i.e., biasing the participants). But the goal is not to objectively collect information about age, gender, household income, or any other demographic that you’d see on a census or questionnaire. The goal is to bring a previously unexpressed attitude to life as a new discrete thought. From this perspective, a market researcher is more akin to a novelist: someone who relies on their subjectivity instead of avoiding it.

Recall that, compared to data giants Google and Facebook, advertising agencies are data poor. As a result, they can’t hire decent data talent, which perpetuates an already lopsided competition for client dollars. The large advertising holding companies are trying to compete directly. As Auletta reports, “W.P.P.’s GroupM, the largest media agency, has quietly assembled what it calls its ‘secret sauce,’ a collection of forty thousand personally identifiable attributes it plans to retain on two hundred million adult Americans.” Publicis Groupe—my employer—has taken a similar strategy.

It will be tough for them to win. An alternative approach would be to blend the data we do have with our core competency—creativity. That is, what if agencies applied their creative talent to change how they measure people? Elevating data and research to essential components of the creative process isn’t just a good idea. It’s our last competitive advantage.

Image credit:

Sam McNerney