How Social Media May Undermine Profits

Research by Tuck professor Lauren Grewal shows consumers’ online identity can have negative effects on sales.

More recently, however, consumers have been offered a way to express themselves without actually buying anything. “Social media,” says Tuck assistant professor Lauren Grewal, “has made it easier to collect virtual products that can allow a consumer to curate a whole identity without physically owning anything.”

The common belief among companies is that the more they encourage consumers to engage with firms on social media, the more those consumers will buy their products. Grewal wanted to test that assumption. “I wanted to know if there were situations where a consumer could fulfill their identity-signaling needs virtually and negate the need of buying physical products,” says Grewal.

Grewal answers that question in an article titled “When Posting About Products in Social Media Backfires: The Negative Effects of Consumer Identity-Signaling on Product Interest,” forthcoming in the Journal of Marketing Research. Co-written with Andrew T. Stephen of the University of Oxford and Nicole Verrochi Coleman of the University of Pittsburgh, it warns marketers that they may want to consider other social media strategies or revise expectations for what these strategies are likely to achieve.

Across five experiments, Grewal and her co-authors established the main finding that posting identity-relevant products in social media can decrease purchase intentions for the same and similar identity-signaling goods. In addition, they found evidence that this occurs in situations when consumers feel as though their virtual actions sufficiently signaled their identity.

Posting identity-relevant products in social media can decrease purchase intentions for the same and similar identity-signaling goods.

For example, in one study using active Pinterest users and real posting behavior, Grewal manipulated the way that consumers thought about the products they posted. In some cases, consumers posted products that were framed as being useful and eco-friendly, or products that could showcase their environmentalist identity. Participants then posted those products on a private Pinterest board seen only by themselves, or a public board that would be seen by others. Afterwards, Grewal asked participants to rate how likely they would be to purchase an eco-friendly backpack. For those posting privately, participants primed to think about how their products signal their identity were significantly more likely to buy compared to those who thought about the functionality of the products they posted. For those posting publicly, however, there was a significant difference in the opposite direction. These findings indicate that when people cannot signal their identities to an audience, they are more motivated to purchase compared to when they are able to publicly signal identity.

In another study, this time using an incentive-compatible design where undergraduate students made decisions involving real money, Grewal found a similar effect as in the Pinterest experiment. In this experiment, the participants found products that were either seen as useful to them as college students or that showcased their college identity, and then they imagined posting them to social media. Some of the participants were primed to have a stronger college identity than others. Afterwards, the participants indicated how much money they would be willing to spend on a university tote bag that signaled their college identity.

Grewal found that for those who were not primed to have a strengthened student identity, individuals who imagined posting identity-relevant products were willing to pay significantly less for the tote ($.86) compared to those who posted functional products ($1.28). Interestingly, when students were primed to have an even stronger student identity, this difference went away. This pattern was found to be caused by the perceived strength of the identity signal, highlighting that a consumer needs to feel their identity is strongly signaled before the effect will occur.

Grewal cautions that she did not test the “network effect” of posting online across studies; that is, even if a consumer is less likely to purchase a product they post themselves, they may influence others to do so, offering a net benefit to sales. She is currently conducting more experiments to gauge those effects.

In a final study, she also found that the effects diminished if she waited a week, after which the salience of the virtual identity signal was lower. This may mean that companies should put some distance between the time of a purchase and when they market online to the same consumer. This could minimize the likelihood it will backfire and deter consumers from future purchases. “The immediate action of companies when someone interacts with something online is to bombard them with ads,” Grewal suggests. “That is not necessarily going to be successful.”  

“Social media matters,” Grewal concludes. “It has real-life ramifications not only on what products we buy, but on how we see ourselves.”