In other words, people who truly know music (the kind of people who run record labels, go to clubs, and know the business well) love Kenna. They hear one of his songs, and, in the blink of an eye, they think, Wow! More precisely, they hear Kenna and their instinct is that he is the kind of artist whom other people—the mass audience of music buyers—are going to like. But this is where Kenna runs into a problem, because whenever attempts have been made to verify this instinct that other people are going to like him, other people haven’t liked him.

When Kenna’s album was making the rounds in New York, being considered by music industry executives, on three separate occasions it was given to an outside market-research firm. This is common practice in the industry. In order to be successful, an artist has to get played on the radio. And radio stations will play only a small number of songs that have been proven by market research to appeal—immediately and overwhelmingly—to their audience. So, before they commit millions of dollars to signing an artist, record companies will spend a few thousand dollars to test his or her music first, using the same techniques as the radio stations.

There are firms, for example, that post new songs on the Web and then collect and analyze the ratings of anyone who visits the Website and listens to the music. Other companies play songs over the phone or send sample CDs to a stable of raters. Hundreds of music listeners end up voting on particular songs, and over the years the rating systems have become extraordinarily sophisticated. Pick the Hits, for instance, a rating service outside Washington, D.C., has a base of two hundred thousand people who from time to time rate music, and they have learned that if a song aimed, say, at Top 40 radio (listeners 18 to 24) averages above 3.0 on a score of 1 to 4 (where 1 is “I dislike the song”), there’s roughly an 85 percent chance that it will be a hit.

These are the kinds of services that Kenna’s record was given to—and the results were dismal. Music Research, a California-based firm, sent Kenna’s CD to twelve hundred people preselected by age, gender, and ethnicity. They then called them up three days later and interviewed as many as they could about what they thought of Kenna’s music on a scale of 0 to 4. The response was, as the conclusion to the twenty-five-page “Kenna” report stated politely, “subdued.” One of his most promising songs, “Freetime,” came in at 1.3 among listeners to rock stations, and .8 among listeners to R&B stations. Pick the Hits rated every song on the album, with two scoring average ratings and eight scoring below average. The conclusion was even more blunt this time: “Kenna, as an artist, and his songs lack a core audience and have limited potential to gain significant radio airplay.”

Kenna once ran into Paul McGuinness, the manager of U2, backstage at a concert. “This man right here,”

McGuinness said, pointing at Kenna, “he’s going to change the world.” That was his instinctive feeling, and the manager of a band like U2 is a man who knows music. But the people whose world Kenna was supposed to be changing, it seemed, couldn’t disagree more, and when the results of all of the consumer research came in, Kenna’s once promising career suddenly stalled. To get on the radio, there had to be hard evidence that the public liked him—and the evidence just wasn’t there.

In Behind the Oval Office, his memoir of his years as a political pollster, Dick Morris writes about going to Arkansas in 1977 to meet with the state’s thirty-one-year-old attorney general, an ambitious young man by the name of Bill Clinton:

I explained that I got this idea from the polling my friend Dick Dresner had done for the movie industry. Before a new James Bond movie or a sequel to a film like Jaws came out, a film company would hire Dresner to summarize the plot and then ask people whether they wanted to see the movie. Dresner would read respondents proposed PR blurbs and slogans about the movie to find out which ones worked the best. Sometimes he even read them different endings or described different places where the same scenes were shot to see which they preferred.

“And you just apply these techniques to politics?” Clinton asked.

I explained how it could be done. “Why not do the same thing with political ads? Or speeches? Or arguments about the issues? And after each statement, ask them again whom they’re going to vote for. Then you can see which arguments move how many voters and which voters they move.”

We talked for almost four hours and ate lunch at his desk. I showed the attorney general sample polls I’d done.

He was fascinated by the process. Here was a tool he could use, a process that could reduce the mysterious ways of politics to scientific testing and evaluation.

Morris would go on to become a key advisor to Clinton when Clinton became President, and many people came to view his obsession with polling as deeply problematic—as a corruption of the obligation of elected officials to provide leadership and act upon principle. In truth, that’s a little harsh. Morris was simply bringing to the world of politics the very same notions that guide the business world. Everyone wants to capture the mysterious and powerful reactions we have to the world around us. The people who make movies or detergent or cars or music all want to know what we think of their products. That’s why it wasn’t enough for the people in the music business who loved Kenna to act on their gut feelings. Gut feelings about what the public wants are too mysterious and too iffy. Kenna was sent to the market researchers because it seems as though the most accurate way to find out how consumers feel about something is to ask them directly.

But is that really true? If we had asked the students in John Bargh’s experiment why they were standing in the hall so patiently after they had been primed to be polite, they wouldn’t have been able to tell us. If we had asked the Iowa gamblers why they were favoring cards from the blue decks, they wouldn’t have been able to say—at least not until they had drawn eighty cards. Sam Gosling and John Gottman found that we can learn a lot more about what people think by observing their body language or facial expressions or looking at their bookshelves and the pictures on their walls than by asking them directly. And Vic Braden discovered that while people are very willing and very good at volunteering information explaining their actions, those explanations, particularly when it comes to the kinds of spontaneous opinions and decisions that arise out of the unconscious, aren’t necessarily correct. In fact, it sometimes seems as if they are just plucked out of thin air. So, when marketers ask consumers to give them their reactions to something—to explain whether they liked a song that was just played or a movie they just saw or a politician they just heard—how much trust should be placed in their answers? Finding out what people think of a rock song sounds as if it should be easy. But the truth is that it isn’t, and the people who run focus groups and opinion polls haven’t always been sensitive to this fact. Getting to the bottom of the question of how good Kenna really is requires a more searching exploration of the intricacies of our snap judgments.

In the early 1980s, the Coca-Cola Company was profoundly nervous about its future. Once, Coke had been far and away the dominant soft drink in the world. But Pepsi had been steadily chipping away at Coke’s lead. In 1972, 18 percent of soft drink users said they drank Coke exclusively, compared with 4 percent who called themselves exclusive Pepsi drinkers. By the early 1980s, Coke had dropped to 12 percent and Pepsi had risen to 11 percent—and this despite the fact that Coke was much more widely available than Pepsi and spending at least $100 million more on advertising per year.