Persuasive Advertising - J. Scott Armstrong [167]
8.9.1. Use sex only when it has relevance to the product
In an ad on the jewelry.com website, a woman wearing pearls is shown in a steamy horizontal embrace with a man. The caption is: “Actual results may vary.” A 1997 Singapore campaign on breast cancer showed men ogling a woman’s breasts as she walked to work; the caption read, “If only women examined their breasts as much as men do.”
On the other hand, if the product is not related to sex, advertisers should avoid sexual appeals as it distracts customers from the arguments. This is especially so for high-involvement products that have benefits. The same applies for the use of violence.
Evidence on the effects of using sex in ads
Subjects, 169 males and 172 females, viewed ads for a fictitious brand of body oils, some with dressed male models, some with suggestive ads, and others with nude models. Sexual content was relevant here, and nudity led to higher purchase intentions (Simpson, Horton, and Brown 1996).
A review of six empirical studies concluded that while sexual appeals attracted attention, they reduced the customer’s thinking about the product. A follow-on lab experiment compared pairs of matched print ads in which one used sexual appeals and the other did not. The experiment covered 13 pro-social topics (e.g., healthy eating, museum attendance, HIV prevention). Sex was not relevant to the products. For example, an ad for libraries had a suggestive photo with the headline, “Find out what’s hot between the covers at your local library.” The 658 subjects were asked to view each ad for 30 seconds. Responses to a questionnaire showed that while the sexual appeals attracted more attention, they did not increase intentions to comply with the ads’ appeals (Reichert, Heckler, and Jackson 2001).
Further support was provided by a lab experiment in which 12 print advertisements were shown to 60 male subjects. Six contained photographs of neutral subjects (e.g., a landscape, a house, or a motorboat), while six showed females in various stages of undress. A well-advertised household-product brand name was provided below each photo. After seven days, the ads were again shown to the subjects; they correctly recalled 61 percent of ads with non-sexual photos, but only 49 percent of those with sexual photos (Steadman 1969).
Here is an experiment about advertising when the TV program involves sex and violence. Subjects saw nine commercials for supermarket items as they watched a TV show that was either neutral, had sexual content, or had violent content. Recall immediately after exposure and again after 24 hours was highest for the subjects who viewed the ads in the neutral program (Bushman and Bonacci 2002).
A lab experiment used 12 ads for unfamiliar brands of inexpensive consumer products that were advertised on TV programs. The 336 subjects were equally assigned to three programs with explicit sexual content, three with violent content, three with both, and three with neither. The products had little relationship to sex. Those viewing the ads in programs with sexual or violent content had lower scores for brand recall, brand recognition, buying intentions, and coupon redemptions. For example, the brand recall was 68 percent higher for people who saw a program without violence or sex than for people who saw a program with violence or sex (Bushman 2005).
In their analysis of non-experimental data, Stewart and Furse (1986) found that TV commercials using sex scored lower on recall, comprehension, and persuasion. These commercials were directed primarily for viewing on major channels, so it was expected that few products advertised in this sample would relate to sex.
8.10. Models
The use of models, typically human but also animals, has a long history in advertising. For example, in 1905, ads using drawings of models by the artist J. C. Leyendecker had a strong impact on college men. They followed every detail of dress in his Arrow Collar shirt ads.
8.10.1.