Scary Health Warnings Can Boost Sales

by Ziv Carmon, Yael Steinhart, and Yaacov Trope

Highlighting the risks of a behavior—smoking or taking a medication, say—ought to make people think twice about using a product. But a series of experiments we conducted reveals that sometimes just the opposite is true: A warning label canincrease a product’s appeal. This happens when there’s a lag between delivery of the message and a decision about buying, consuming, or assessing the product. Why? The mere inclusion of a warning builds trust, because consumers feel that the seller is being honest—and over time that trust becomes more prominent, while the substance of the warning fades.

In one experiment, we showed a cigarette ad to 71 smokers and asked if they wanted to place an order for the brand. Half the participants saw a version of the ad that included a warning about health hazards; half saw a version with no warning. Half the people in each group were then told that any cigarettes they bought would arrive in a day—a “near future” condition. The other half were told that the cigarettes would come three months later—a “distant future” condition. In the near-future condition, the warning was effective—participants who had seen it ordered 75% fewer packs of cigarettes, on average, than those who hadn’t. In the distant-future condition, however, the warning backfired—participants who had seen it ordered 493% more packs than those who hadn’t.

To test the impact of a different kind of time lag, we created two versions of an ad for an artificial sweetener—one that warned of immune-system damage and other adverse side effects, and one that didn’t—and tested them on a group of 74 women. Again, half the participants saw each version of the ad, but this time we set up immediate-choice and delayed-choice conditions: Half the women in each group could order the sweetener right away, and the other half were given the opportunity to order two weeks later. Here, too, the lag caused the warning to backfire. In the immediate-choice condition, women who had seen the warning ordered 94% fewer packages of sweetener than those who hadn’t; in the delayed-choice condition, however, they ordered 265% more.

Subsequent experiments involving erectile-dysfunction and hair-loss medications showed that delays have similar effects when people rate the desirability of a risky product after seeing an ad for it: Warnings led to higher ratings when participants were asked to evaluate the product two weeks after they saw the ad and when they were informed that the product would become available in a year.

Source: Three warnings from the authors’ study

These findings have important implications for regulators and for managers in fields such as consumer products, health care, and finance. Those who want to minimize the deterrent effects of a warning would do better to build in a delay of some sort (almost any kind of delay will do) than to bury the warning in fine print. In fact, the latter course might hurt sales by forgoing the credibility that warning labels can confer and arousing suspicion instead. Those who genuinely wish to warn consumers should ensure that the message is conveyed—or repeated—shortly before the relevant event, whether that’s product use, a medical procedure, or a high-stakes investment.

Ziv Carmon is the Chaired Professor of Marketing in Memory of Erin Anderson at Insead. Yael Steinhart is a senior lecturer in marketing and consumer behavior at Tel Aviv University. Yaacov Trope is a professor of psychology at New York University.