Scholarly Insights: AMA’s digest of the latest findings from marketing’s top researchers
The Problem with Chasing “Likes” in Social Media
Companies spend billions of dollars each year on social media to establish and maintain a presence on social networking sites. Given the substantial investment of time, money, and firm resources spent to accumulate “likes” on Facebook, marketers are increasingly asking themselves, “What is the value of a Facebook fan?”
And the punchline: Joining a brand’s social network induces no change in consumer attitude or behavior; it is simply a symptom of preexisting fondness for the brand.
In a recent article published in the Journal of Marketing Research, my colleagues and I address this question by assessing whether joining a brand’s social network changes consumer behavior. We look at both the consumers who join (i.e., first-order effects) and their friends (i.e., second-order effects).
There is an inherent appeal to the idea that recruiting new followers in social media translates into positive marketing outcomes such as improved brand attitudes and sales. However, evidence to support this assumption has been elusive. A recent survey shows that 87% of Fortune 500 CMOs have not been able to document whether social media creates new customers (for more information, see the CMO Survey’s latest results).
Adding further to the confusion, red herrings abound: associations between social media engagement and offline behavior are often taken as evidence that acquiring “likes” – or more broadly, obtaining social media followers – boosts the bottom line. Indeed, an influential Starbucks study published in the Journal of Advertising Research seemed to “confirm” this—people who had subscribed to the brand’s Facebook page spent more and transacted more frequently than those who had not. Case closed, right? … Not by a longshot.
Disentangling Cause and Consequence in Social Media Engagement
These observations are plagued by a fatal logical flaw: They confuse cause and consequence. When we see the association – e.g., fans spend more than nonfans – it’s tempting to chalk it up to causality and conclude that social media engagement causes positive marketing outcomes, such as improved brand attitudes and increased purchasing. In other words, becoming a fan makes a person buy more than she otherwise would have.
But surely those with a positive predisposition toward a brand are more likely to follow it relative to those without such feelings. In other words, both purchasing and following a brand on social media are, at least to some degree, symptoms of a person’s fondness for that brand. In this case, following a brand wouldn’t cause the person to buy any more than she otherwise would have. So how much, if any, of the positive associations between following a brand on social media and offline behavior is truly causal?
This distinction is not trivial. For marketers and social media, disentangling cause from consequence is important because it points to different best practices for social media use. If there is a causal element to this association – if following a brand causes a person to buy more from that brand – then it’s reasonable to expect that recruiting followers will have a positive effect on sales. To capitalize on it, a marketer would want to proactively get people to join the brand’s social media channels – for example, by offering them incentives to join. However, if there’s no causal relationship in there, it would be unrealistic to expect recruiting followers to boost sales, at least without any supporting marketing activities. In this case, the marketer may be better off using social media as an organic way of finding their best customers. In other words, dangle the brand out on social media and let your best customers find you.
The Research: Does “Liking” Lead to Loving?
To assess whether “liking” changes consumers’ attitudes and behaviors toward a brand, we conducted a series of experiments with more than 14,000 participants. We induced some people to join a brand’s social network (i.e., to “like” it on Facebook) and subsequently measured their attitudes toward the brand and propensity to buy it. Others – our control group – were not induced to join the social network. This control group therefore provided us with a crucial counterfactual: it told us how people would have behaved toward the brand had they not “liked” it on Facebook.
Will Consumers Buy More of a Brand if They “Like” It?
Using a treatment and control approach enabled us to pinpoint whether “liking” induces changes in consumers’ attitudes and behaviors – changes that would not have occurred had they not “liked” it. In subsequent studies, to provide a benchmark against which to compare the (in)effectiveness of “liking,” we also tested the effect of advertising on brand attitudes and purchasing propensity. These studies show that in contrast to traditional advertising, which we found to have a modest but positive effect on consumer attitudes, “liking” had no effect. Consumers attitudes toward the brand and their likelihood of purchasing the brand did not change.
Will “Liking” a Brand Influence Consumers’ Social Networks?
What about possible second-order effects? In other words, does “liking” brands cause consumers’ friends to view that brand any more favorably? We find that when consumers see that a friend has “liked” a brand, they are less likely to buy the brand compared with more meaningful (“offline”) forms of social endorsement – for example, when they see a friend using the brand or when a positive brand experience comes up in conversation with the friend.
Turning “liking” into improved brand attitudes and increased purchasing by either consumers or their friends requires more than just the click of a button. In fact, new research by Daniel Mochon, Karen Johnson, Janet Schwartz, and Dan Ariely (forthcoming in the Journal of Marketing Research) suggests that when marketers support ‘likes” with advertising to these new followers, it can influence impact consumer behavior for the better. In other words, Facebook can be effective as a platform for paid advertising.
Summary of the Findings
We find that the mere act of joining a brand’s social network (i.e., “liking” a brand on Facebook) does not cause consumers to view the brand more favorably. The people we encouraged to “like” the brand acted no more positively toward the brand relative to people in our control group, who were not induced to “like” it.
So what explains the positive associations between brand social media engagement and improved marketing outcomes? We conclude that it’s because such engagement is a symptom of being fond of a brand. In other words, social media engagement is a consequence, not a cause, of fondness for a brand.
Should Marketers Continue to Invest in Accumulating “Likes”?
We find no evidence of direct (first-order) benefits from inducing consumers to “like” brands on Facebook. So from this perspective, consumers’ willingness to “like” brands can give marketers the illusion that their social media efforts are having an impact; after all, “number of members” is both a salient and readily available metric.
However, this is not to say that acquiring “likes” on Facebook, or building a social media presence more broadly, is ineffective for brands. In addition to being an effective platform for paid advertising, social media may also have a host of brand benefits that should be explored; for example, what potential does social media hold as a relationship-building, involvement-raising, and knowledge-enriching tool? While there are still many unresolved questions about the true benefits of social media engagement, one finding seems to loom large: mere “liking” does not lead to loving.
Article Citation
Leslie K. John, Oliver Emrich, Sunil Gupta, and Michael I. Norton (2017), “Does ‘Liking’ Lead to Loving? The Impact of Joining a Brand’s Social Network on Marketing Outcomes,” Journal of Marketing Research, 54 (1), 144-155.
doi: https://doi.org/10.1509%2Fjmr.14.0237
Leslie K. John is Associate Professor of Business Administration, Harvard Business School, Harvard University.