That claim is not from the actual study, it's just from Fortune. All the results in the study point to sexualised and violent ads not being as brand effective as nonviolent/nonsexual (i.e. neutral) ads. "There were no significant effects of sexual or violent ads on memory or buying intentions".
...
Brand Attitudes:
Media content. Overall, attitudes were significantly less favorable for brands embedded in violent media than for the same brands embedded in neutral media, d 0.37; 95%CI 0.55, 0.18; k 9.
Only one study examined attitudes for brands in sexual media, so we were not able to conduct meta-analyses for sexual media (Bello et al., 1983). However, that studys effect size was in the same direction as the effect sizes for violent media, d 0.26, 95% CI 0.52, 0.01.
Ad content. Overall, attitudes were significantly less favorable for brands in sexual ads than for the same brands in neutral ads, d 0.32; 95% CI 0.55, 0.10; k 11.
Buying Intentions:
Media content. Overall, buying intentions were significantly lower for brands in media containing violence, sex, or both violence and sex than for the same brands in media containing no violence or sex, d 0.26; 95% CI 0.53, 0.004; k 8.
Effects did not significantly differ for programs containing violence (d 0.27; 95% CI 0.53, 0.013; k 7), sex (d 0.19; 95% CI 0.63, 0.26; k 3), and both violence and sex (d 0.24; 95% CI 0.91, 0.44; k 2), 2 (2) 0.10, p .95. However, only the confidence interval for programs containing violence excluded the value zero.
Ad content. Overall, buying intentions did not depend on whether the ad contained sex or violence, d 0.04; 95% CI 0.12, 0.20; k 13. Effect sizes did not significantly differ for sexual ads (d 0.10; 95% CI 0.03, 0.23; k 22) or violent ads (d 0.03; 95% CI 0.27, 0.32; k 5), 2 (1) 0.21, p .65. Both confidence intervals included the value zero.
Yes, and unless that cologne or vodka has people fucking in the bottle, I don't see how it is relevant.