Does Sex Really Sell?

Society drives people crazy with lust and calls it advertising,” John Lahr once wrote. Of course, whether that craziness translates into sales is debatable.

MediaAnalyzer Software & Research in Somerville, Mass., recently set out to explore how men and women look at sexually themed ads and what effect, if any, that visual behavior might have on the ads’ effectiveness. In September, the company had 200 men and 200 women take an online test. The first part of the test solicited general opinions about sex in advertising. (Those answers are summarized in charts on page 17.) The second part involved a visual test in which MediaAnalyzer used its AttentionTracking software to follow the visual behavior of respondents as they looked at 10 print ads. (The software has users move the mouse over each ad to indicate where he or she is looking.) The ad sample consisted of two U.S. print ads, one sexual and one nonsexual, from each of five product categories—cigarettes, credit cards, jeans, shoes and alcohol. MediaAnalyzer used the data from the visual test to create versions of the ads that show viewing patterns (with arrows) and time spent in each ad region (with percentages). Those versions are reproduced over the next three pages. (The study also produced “heatmaps,” not featured in this story, which show the emergence of each ad’s “hot spots” over five seconds of viewing.)

Responses to the general questions in the survey revealed that sex in ads is a polarizing issue. While almost half of men (48 percent) said they like sexual ads, few women did (8 percent). Most men (63 percent) said sexual ads have a high stopping power for them; fewer women thought so (28 percent). Also, most women (58 percent) said there is too much sex in advertising; only 29 percent of men said so. Women were also much more likely than men to say that sexual ads promote a deterioration of moral and social values and that they are demeaning for the models used in them.

The visual test exposed a similar polarization. Men tend to focus on an ad’s sexual imagery (breasts, legs, skin, etc.), which draws their attention away from other elements of the ad (logo, product shot, headline). This may be why men’s brand recall was worse for the sexual ads than for the nonsexual ones. An average of 19.8 percent recalled the correct brand/product for the nonsexual ads; for the sexual ads, 9.8 percent did. MediaAnalyzer calls this the “vampire effect,” with a too-strong visual sucking up a lot of the attention that would have otherwise been spent on an ad’s actual communication.

Women, meanwhile, tend to avoid looking at the sexual imagery, but curiously, their brand recall was worse with the sexual ads, too. An average of 22.3 percent recalled the correct brand/product for the nonsexual ads; only 10.8 percent correctly recalled the sexual ads. MediaAnalyzer hypothesizes that this might be the result of a general numbing effect that sexual stimuli has on the brain.

In trying to determine the effectiveness of each ad, the survey measured three other criteria besides brand recall: ad like, product like and purchase intent. Men said they liked the sexual ads more, liked the products advertised in them more and would be more likely to buy those products. Women scored the sexual ads lower than the nonsexual ones on all three of those criteria.

MediaAnalyzer also uses the data to offer advice for agencies considering using sexual images. For info, visit

Over the next three pages, we focus on how men and women looked at each pair of ads.