5.6 The Benefits and Drawbacks of Targeting: Why is My Feed Full of Otters?

When John Wanamaker made that comment about half of advertising money being wasted but not knowing which half, he didn’t have the benefit of modern audience analysis techniques. He would probably be impressed with the current world of targeted advertising, in which video streaming services, online retailers, and news sources know a lot about what you like to watch, buy, and read. Theoretically, this should lead to a world in which no one is bothered by ads for things they don’t want, or wastes their time checking out television shows that don’t suit them. Certainly the system is not perfect (a Facebook friend recently complained, “I am getting so tired of the way I have to scroll through literally dozens of stupid ‘Otter lovers’ or ‘Sloth lovers’ or ‘Giraffe lovers’ pages every time I sign on to Facebook”), but it is nice to get alerts about things that are tailored to you, or hear news you might not otherwise have heard about an obscure musician you like.

So what is the downside of all this data collection? Marketing expert Jason Voiovich has thoughts on the issue:

Despite the geometric rise in the amount of data available about all aspects of consumer behavior; despite the increase in the immediate accessibility of that data; and despite increasingly sophisticated tools to (supposedly) help make sense of it all, marketing remains the least improved of all the business disciplines.

In his view, it has actually become too easy to collect data on audiences, lulling researchers into thinking that they understand audiences, and making them forget the role of conscious thought and freedom of choice in purchasing decisions. “More data has pushed us further away from our audience, not moved us closer to them. Paradoxically, the more we know about audiences, the less we know them.”^[1]

If more information about your audience improves communication, why do so many people feel like the music world is artistically dead, movies are formulaic, and the divisiveness in the political realm is unsolvable? Social media has been accused of fostering an “echo chamber,” where people just reinforce pre-existing beliefs and no one listens to new ideas. Internet activist Eli Pariser has warned of “filter bubbles” that result in different people getting markedly different results when they search for the same things, because search engines and news sources already know what the user likes to hear, leading to “intellectual isolation and social fragmentation.”^[2] Algorithms also rest on the fundamental assumption that people already know exactly what they like and what they believe, and are not interested in anything new.

One clue can be found in the 1994 film Shawshank Redemption. Ellis “Red” Redding (played by Morgan Freeman) is serving a life sentence, and gets the chance to appeal to the parole board. When they ask him if he’s been rehabilitated, Red’s answer is “Oh yes sir, absolutely sir. I’ve learned my lesson. I can honestly say that I’m a changed man. I’m no longer a danger to society. That’s God’s honest truth.” His appeal is rejected. For years I have shown students that scene and asked why they think his message failed; many students say it’s because Red is only telling them what they want to hear. Why is that a bad thing? The main theme of this chapter is that tailoring your message to your audience is the key to effective communication, so what is wrong with “telling them what they want to hear”?

For one thing, it’s sometimes too obvious, which is made clear when Red gets another parole board hearing 10 years later and gives the identical speech. For another thing, it leads to predictability and uniformity: if prisoners figure out what the parole board is listening for, that’s all the board will hear from every prisoner. But the biggest problem is that “telling what they want to hear” lacks the other key ingredient they are looking for: sincerity. Over-tailoring the message means that what the source wants to say is no longer part of the equation and, paradoxically, that is part of what audiences want to hear.^[3] All of these problems apply to advertising, music, and political messages as well: if the techniques are too obvious, the results too predictable and formulaic, and the audience doesn’t sense any sincerity, the message will fail.

Data collection tends to make people risk-averse. As mentioned, Coca Cola, Microsoft, Google, and Facebook don’t like to lose money, and would have preferred to avoid product launches that ended up failing. I suggested that some thorough audience analysis in advance would have spared them huge losses. But if a company becomes too risk-averse, it doesn’t take chances on new things or create anything truly original, and that can kill a company, too. Steve Jobs was well known for taking leaps with new things such as iTunes, Pixar Animation, and the iPhone, and ignoring warnings that all of these things would fail. And in Shawshank Redemption, when Red finally dared to tell the parole board things they were not accustomed to hearing, and made it clear that he was not tailoring his message for them (“Rehabilitated? It’s just a bulls**t word. So you go on and stamp your forms, sonny, and stop wasting my time, because to tell you the truth, I don’t give a s**t”), that’s when they finally did release him, because they finally recognized that he was telling the truth.

In other words, sometimes too much audience research can be a bad thing, and there is something to be said for flying blind.