Persuasive Advertising - J. Scott Armstrong [52]
• “plea plus information” group also received facts about recycling
• “household feedback” group received feedback about their own recycling behavior
• “neighborhood feedback” group received feedback about recycling in their residential area.
There was a significant increase in the frequency and amount of recycling by the household feedback group, and even more for the neighborhood-feedback group (Schultz 1998).
Field experiments demonstrated the importance of identification. They examined how signs affect the reuse of towels during their hotel stay. The standard sign, which asked people to participate in the reuse program to “help save the environment,” led to a 37 percent reuse rate. A second sign added a message that “In fall of 2003, 75 percent of the guests participated by reusing their towels at least once,” and increased the reuse to 44 percent. Another sign reporting on the reuse rate of guests in the particular room—“75 percent of the guests who stayed in this room”—increased the reuse rate to 49 percent (Goldstein, Cialdini and Griskevivius 2008).
A similar series of experiments used a sign stating that most guests accepted the conservation norm and that many similar people at the hotel had recycled their towels. This led to a 25 percent reduction in the hotel’s laundry bill in comparison with the older message requesting reuse for energy conservation (Schultz, Khazian, and Zaleski 2008).
Our analysis of quasi-experimental data provided additional support:
Print ads that showed products being used by people similar to a distinctive target market had better recall. Our WAPB analysis found 26 pairs of print ads in which one ad showed the products being used by people with distinctive features that were similar to the target market while the other ad did not. Recall for ads showing a distinctive target market was 1.21 times better than for the other ads.
Turning to non-experimental data, TV commercials rated above the median on being “for people like you,” had 16 percent better recall and 21 percent higher persuasion than did those below the median (Walker 2008).
TV commercials with background characters were substantially less persuasive, had lower recall, and poorer comprehension than those without background characters (Stewart and Furse 1986).
2.3. Scarcity
In 1934, B. Altman department store ran a print ad with the headline, “We believe that there are at least 500 men in New York who love their wives … and want to give them flowers for Easter.” It sold 500 bouquets of flowers in three hours, and the ad was selected as the best retail ad of that year (Watkins 1959).
If people perceive that a product is restricted or in short supply, they often value it more. This is partly due to a desire to impress people by having something others do not have. In that sense, it seems to conflict with “social proof.” However, scarcity sometimes serves as a signal of social proof in that “everyone wants this.”
Scarcity plays a role when it is perceived as a loss of freedom of choice. In 1985, Coca-Cola found that in blind taste tests, people preferred the taste of “New Coke” to the old formula. However, when the company removed the old formula from the market, customers reacted as if their freedom had been violated and demanded its return.
Much legal research supports these anecdotes on loss of freedom. For example, in a simulated jury trial involving an accident, when jurors were not told that the defendant had accident insurance, they awarded a damage claim of $33,000. When told that he did have insurance, the award rose to $37,000. However, when told that he had insurance and the judge instructed them to ignore that information, the award rose to $46,000 (Broeder 1959). Similarly, another lab experiment found that when weak information on guilt was ruled admissible in a law case, 26 percent of the subjects found the defendant guilty, and that the percentage rose to 35 percent when the same evidence was presented and then ruled inadmissible (Sue, Smith, and Caldwell 1973).
Still another experiment