What's interesting about this story is that by every normal expectation McCann should have won the test. The gold box idea sounds like a really cheesy idea. Columbia was so skeptical of it that it took Wunderman several years to persuade them to let him try it. McCann, meanwhile, was one of the darlings of Madison Avenue, a firm renowned for its creativity and sophistication. Furthermore, McCann spent four times as much as Wunderman on media time. They bought prime-time slots for their space. Wunderman's ads were on in the wee hours of the morning. In the last chapter, I talked about how epidemics are, in part, a function of how many people a message reaches, and by that standard McCann was way ahead. McCann did all the big things right. But they didn't have that little final touch, that gold box that would make their message stick.

If you look closely at epidemic ideas or messages, as often as not the elements that make those sticky turn out to be as small and as seemingly trivial as Wunderman's gold box. Consider, for example, the so-called fear experiments conducted by the social psychologist Howard Levanthal in the 1960s. Levanthal wanted to see if he could persuade a group of college seniors at Yale University to get a tetanus shot. He divided them up into several groups, and gave all of them a seven-page booklet explaining the dangers of tetanus, the importance of inoculation, and the fact that the university was offering free tetanus shots at the campus health center to all interested students. The booklets came in several versions. Some of the students were given a "high fear" version, which described tetanus in dramatic terms and included color photographs of a child having a tetanus seizure and other tetanus victims with urinary catheters, tracheotomy wounds, and nasal tubes. In the "low fear" version, the language describing the risks of tetanus was toned down, and the photographs were omitted. Levanthal wanted to see what impact the different booklets had on the students' attitudes toward tetanus and their likelihood of getting a shot.

The results were, in part, quite predictable. When they were given a questionnaire later, all the students appeared to be well educated about the dangers of tetanus. But those who were given the high-fear booklet were more convinced of the dangers of tetanus, more convinced of the importance of shots, and were more likely to say that they intended to get inoculated. All of those differences evaporated, however, when Levanthal looked at how many of the students actually went and got a shot. One month after the experiments, almost none of the subjects — a mere 3 percent — had actually gone to the health center to get inoculated. For some reason, the students had forgotten everything they had learned about tetanus, and the lessons they had been told weren't translating into action. The experiment didn't stick. Why not?

If we didn't know about the Stickiness Factor, we probably would conclude that something was wrong with the way the booklet explained tetanus to the students. We might wonder whether trying to scare them was the appropriate direction to take, whether there was a social stigma surrounding tetanus that inhibited students from admitting that they were at risk, or perhaps that medical care itself was intimidating to students. In any case, that only 1 percent of students responded suggested that there was a long way to go to reach the goal. But the Stickiness Factor suggests something quite different. It suggests that the problem probably wasn't with the overall conception of the message at all, and that maybe all the campaign needed was a little gold box. Sure enough, when Levanthal redid the experiment, one small change was sufficient to tip the vaccination rate up to 28 percent. It was simply including a map of the campus, with the university health building circled and the times that shots were available clearly listed.

There are two interesting results of this study. The first is that of the 28 percent who got inoculated, an equal number were from the high-fear and the low-fear group. Whatever extra persuasive muscle was found in the high-fear booklet was clearly irrelevant. The students knew, without seeing gory pictures, what the dangers of tetanus were, and what they ought to be doing. The second interesting thing is that, of course, as seniors they must have already known where the health center was, and doubtless had visited it several times already. It is doubtful that any of them would ever actually have used the map. In other words, what the tetanus intervention needed in order to tip was not an avalanche of new or additional information. What it needed was a subtle but significant change in presentation. The students needed to know how to fit the tetanus stuff into their lives; the addition of the map and the times when the shots were available shifted the booklet from an abstract lesson in medical risk — a lesson no different from the countless other academic lessons they had received over their academic career — to a practical and personal piece of medical advice. And once the advice became practical and personal, it became memorable.

There are enormous implications in Levanthal's fear experiments and Wunderman's work for Columbia Records for the question of how to start and tip social epidemics. We have become, in our society, overwhelmed by people clamoring for our attention. In just the past decade, the time devoted to advertisements in a typical hour of network television has grown from six minutes to nine minutes, and it continues to climb every year. The New York-based firm Media Dynamics estimates that the average American is now exposed to 254 different commercial messages in a day, up nearly 25 percent since the mid-1970s. There are now millions of web sites on the Internet, cable systems routinely carry over 50 channels of programming, and a glance inside the magazine section of any bookstore will tell you that there are thousands of magazines coming out each week and month, chock-full of advertising and information. In the advertising business, this surfeit of information is called the "clutter" problem, and clutter has made it harder and harder to get any one message to stick. Coca-Cola paid $33 million for the rights to sponsor the 1992 Olympics, but despite a huge advertising push, only about 12 percent of TV viewers realized that they were the official Olympic soft drink, and another 5 percent thought that Pepsi was the real sponsor. According to a study done by one advertising research firm, whenever there are at least four different 15-second commercials in a two-and-a-half-minute commercial time-out, the effectiveness of any one 15-second ad sinks to almost zero. Much of what we are told or read or watch, we simply don't remember. The information age has created a stickiness problem. But Levanthal and Wunderman's examples suggest that there may be simple ways to enhance stickiness and systematically engineer stickiness into a message. This is a fact of obvious importance to marketers, teachers, and managers. Perhaps no one has done more to illustrate the potential of this kind of stickiness engineering, however, than children's educational television, in particular the creators of Sesame Street and, later, the show it inspired, Blue's Clues.

Sesame Street is best known for the creative geniuses it attracted, people like Jim Henson and Joe Raposo and Frank Oz, who intuitively grasped what it takes to get through to children. They were television's answer to Beatrix Potter or L. Frank Baum or Dr. Seuss. But it is a mistake to think of Sesame Street as a project conceived in a flash of insight. What made the show unusual, in fact, was the extent to which it was exactly the opposite of that — the extent to which the final product was deliberately and painstakingly engineered. Sesame Street was built about a single, breakthrough insight: that if you can hold the attention of children, you can educate them.