View more from the
Why don’t customers do what they say they’ll do? Consider the experience of the home appliance manufacturer Electrolux. On the basis of its customer feedback, it once contemplated offering free washing machines and using smart technology inside them to charge customers by the wash.
In prelaunch research, consumers had welcomed the idea for several reasons: The washers involved no up-front purchase costs, used less energy, would be upgraded at no charge, and could be repaired faster and more accurately, thanks to diagnostic capabilities embedded in them. Surely this was the future of laundry? But when a trial was run in Sweden, there was simply no demand for the free washers, and the project was shelved.
Many marketers assume that cautionary tales like this just reflect what happens when you present people with an actual decision versus a hypothetical one. That’s part of the explanation, but another factor is at work here: social identity.
People are highly social animals, belonging to many social groups, each with a distinct identity. You can have an identity as a Catholic, a Jew, or a Hindu; as an American or a Russian; as a professor or a musician; and so on. People don’t identify with all their groups at the same time, of course. You probably won’t identify as a Red Sox fan when you’re in church any more than you would feel especially Catholic while taking in a game at Fenway Park.
Social identities are important for marketers because they guide people’s behavior at any given moment. Some behavior will bolster and support the group, and, equally important, some behavior will betray the group. It is no coincidence that people in the same profession—successful athletes, say, or chief executives—tend to buy similar cars and read similar magazines. When it comes to a purchase, the group you identify with at the time of the transaction is a very important factor in your decision.
But a customer’s social identity at such a moment can’t be easily captured through questions on surveys, whether before or after the purchase. Subtle shifts in social context can dramatically change what group we identify with at any instant. Waiting in the business lounge to board a plane, we might reach for Harvard Business Review, not just for its content but also, subconsciously, to reinforce our identity as a successful executive. A chance conversation about the background music with a neighbor in the lounge, however, might lead us instead to choose a music magazine to reinforce our identity as a rock fan.
This provides a plausible explanation for Electrolux’s failed experiment. Answering questions as part of a consumer study might have triggered a “market research respondent” identity in people, making them try to judge the proposed service dispassionately and with an open mind (behavior appropriate to that identity). But out in the real world, the washing machines ran up against another well-defined aspect of social identity: Middle class families don’t rent appliances—and they certainly do not need to pay per wash cycle, which would seem perilously close to feeding coins into the electricity meters found in some low-income homes. The target consumers didn’t want people to question their status as members of the middle class.
Over the past five years we have been studying how social identity shapes customers’ behavior, working with organizations in sectors as diverse as consumer packaged goods, retail, professional services, and philanthropy. As we’ll show below, companies can subtly influence which social identities customers will tap into and can even foster new identities altogether with very little effort. Let’s begin by examining the dynamics of social identity in more detail.
How Social Identities Shift
We all have an image of ourselves, or a sense of who we are; this is called our self-concept. A social identity is the part of our self-concept that results from our perceived membership in a group.
We can break this rather abstract-sounding notion into two components. First, we think of ourselves as members of some groups but not others. At any point in time, what we’re doing and where we are brings one of our groups front of mind. When exercising in the gym, for instance, we might see ourselves as members of the group of people who stay fit. We probably won’t be identifying with people who like drinking wine.
Second, there’s a set of behaviors that we recognize as appropriate to a given group. So in the gym we might buy a Gatorade not just for its taste and nutrition but also (subconsciously) to further demonstrate that we are one of the fitness-conscious. Our social identity helps us understand how to act in a context in a way that enhances our distinctiveness and status.
Recent research into why people put solar panels on their houses shows very nicely how this works. You might think that their main motivation is reducing energy costs or concern for the environment. But it turns out that the strongest factor is that other solar-powered homes are nearby. Your neighborhood represents a powerful social identity, and it colors how you think about decisions related to your home. If you see solar panels on houses around yours, you’re likely to feel that you should get them as well.
Different social identities can be triggered with relative ease and speed, as one striking experiment shows. Researchers at Lancaster University in the UK recruited a number of supporters of the team Manchester United, saying that they wanted to interview them about what it meant to be a soccer fan. Half the group was asked about being a Manchester United fan, and the others about being a soccer enthusiast in general.
The fans were then moved in smaller groups to a different venue, where they witnessed a man falling down some stairs in what they thought was an accident but, in reality, was an act by a stuntman. The stuntman in some cases wore a Manchester United shirt, in others a shirt from the team’s archrival Liverpool, and finally just a plain T-shirt. The researchers wanted to see whether the victim’s group affiliation affected his likelihood of being helped.
A supermarket shopper might have multıple social identıties—cook, family member who budgets responsibly, good host—that ınfluence his decisions.
The results were unequivocal. If the interviewees had been asked about being a Manchester United fan, they were much more likely to help a victim wearing a Manchester United shirt or a plain shirt than one wearing a Liverpool shirt. If they had been talking about soccer in general, however, the team on the victim’s shirt was less important than whether he was wearing a soccer shirt or a plain one.
This experiment clearly demonstrated how much our social identity depends on context, such as who’s around us and what’s being told to us. We can think of our array of social identities as radio stations; each of us subconsciously moves the dial until we land on one that is in tune with us and our surroundings. That station helps us understand the social situation most clearly. We can select only one at a time, even though they’re all broadcasting. But within moments, we can change stations and assume another identity.
Managing Social Identities
The implications for marketers are obvious. If social identity shapes decisions, then a company’s marketing strategy should encourage customers to tune in to an identity that inspires behaviors like visiting a website, going into a store, buying the product or service, getting more value from it, telling others about it, and helping design a better product.
The first step is to shift from the traditional focus on an individual’s attitudes to a focus on the individual’s social self. Once you make that shift, it becomes easier to see what identities the consumer might be selecting at the moment he or she encounters a brand. It’s intuitive, for example, that a supermarket shopper might have multiple social identities—as a cook, a family member who budgets responsibly, a good host, and so on. Interviews and other research techniques will help you discover the various identities that might be influencing your target consumer’s decision.
But retrospective interviews won’t reveal which identity was selected when. To find that out, it’s necessary to observe consumers over time. This can be done by literally tagging along with them during their customer journeys or through technology. In some sectors insights into social context can be teased out from careful analysis of social media postings. Another good option is real-time experience tracking, a research approach wherein consumers are asked to report in by text message whenever they encounter a certain brand. (See “Better Customer Insight—in Real Time,” HBR, September 2012.)
Unilever has used real-time tracking in its marketing campaigns for the deodorant Axe (known as Lynx in some countries). Its ads focus on how the product makes its target market of young men more attractive to women. Executives were initially baffled about why the campaigns, which were so successful in other countries, did not work well in Italy. Real-time research revealed the problem: Young men in Italy were often still living with their parents, so when they saw the ads on TV at home, their relevant social identity was “dutiful son” rather than “available man.” As a result they rejected the ads’ message, since it focused on behavior that seemed uncomfortable in front of mom. The answer was to market to the target customers in the evening, when they were out with friends, through billboards and the “Axe Police,” attractive young women spraying passing men with Axe. Same target consumer and same message but delivered in the context of a different social identity.
Once the range of possible social identities has been surfaced, the marketer’s strategy should be to achieve one of the following goals:
Boost the signal strength.
When consumers identify with a social group that has a well-defined, positive image, they tend to select products that most clearly broadcast membership in it. Toyota’s marketing for its Prius hybrid car provides an example of how to do this well. By September 2014 the Prius accounted for over half the hybrids ever sold in the United States, and its cumulative U.S. sales were seven times greater than those of its nearest competitor, Honda’s Civic.
The principal difference between the two models is that the Prius is available only as a hybrid and looks distinctively different from Toyota’s conventional gasoline-powered cars. Honda’s hybrid, in contrast, came out under a product line that included similar-looking conventional cars. Driving a Prius could only mean that you were driving a hybrid, but other people couldn’t tell if your Civic was a hybrid without squinting at its rear to see if it carried a hybrid badge. In other words, the Prius gave consumers who identified themselves as environmental supporters a far better opportunity to demonstrate their green credentials to others, which in turn strengthened this group.
Help customers tune in better.
In some cases, a company’s communications may inadvertently suggest that using the product will clash with behaviors required of members in a relevant social group. This can often be fixed by simply reframing the messaging.
Nescafé discovered this early on in the launch of its first instant coffee in the 1950s. The product was initially positioned as a time-saver; no longer would you have to spend 10 minutes brewing a pot of coffee. It did not catch on. The reason, it turned out, was that saving time and effort in this way clashed with the identity of most women, who during that decade felt pressured to be perfect homemakers for hardworking husbands. So the company changed its approach. Instead of claiming that the new product would reduce the time it took to make your husband coffee, the ads suggested it would help you serve him better coffee and leave you extra time to do more for him. Dated and offensive as this may sound today, the reframing was effective. Sales tripled during the decade and had increased 12-fold by the mid-1970s.
Add a song to the station’s playlist.
Social identities aren’t supported by just one behavior but by a collection of behaviors (or playlist, if you will). Another option for marketers, therefore, is to add a new behavior to the recognized set. That can be done by suggesting a new goal to the group.
Take Jeep’s all-terrain vehicles, which give their owners privileged access to hard-to-reach places. As a long-standing supporter of the Tread Lightly initiative, which urges people to respect, protect, and enjoy the outdoors, Jeep encourages owners to drive responsibly in these environments—leaving minimal tracks in wilderness areas, for instance. Notably, Jeep organizes courses in off-road driving skills at Jeep weekends, which are attended by the most enthusiastic Jeep owners.
In this way the carmaker has given the social group of Jeep owners the goal of protecting nature and pointed its members to behaviors in line with that goal. The training is done by more-experienced Jeep customers on a volunteer basis and raises their status in the group. Newer drivers engage in the training and adopt the new practices not just out of environmental respect but to embed themselves more deeply in the group. With minimal expenditure, the program significantly increases appreciation for the Jeep brand, which translates into recommendations and repeat purchases.
The study results were clear: The members of the groups with the stronger social identities engaged far more in the behaviors the company desired.
Find a different station playing your song.
Sometimes customers encounter a product when they’ve adopted a social identity that promotes behaviors at odds with a product’s value proposition. In these cases smart companies look for ways to trigger, or prime, a more constructive identity.
As part of its initiative to reduce global water usage, Unilever markets a fabric conditioner, Comfort One Rinse, that requires far less water than most conditioners do. Objectively, this is a plus in water-stressed parts of Asia, such as India, Thailand, Vietnam, and Indonesia. Yet despite a strong positive response in early trials, the team behind the brand saw neither the expected sales nor a change in water consumption; most customers still used the product with copious amounts of rinsing water.
The reason was that in many of the product’s markets, women do their laundry in public spaces. While women were engaged in this activity, their primed social identity was “diligent mother,” and the appearance of cutting corners by washing clothes in one bucket of water rather than three did not reinforce it.
To get around this problem, Unilever has sought to prime an identity of “smart, savvy homemaker” in customers through a variety of initiatives. One distributes good-housekeeping guides with tips on saving money and time. Another provides a variety of education sessions to local groups of women while they do their laundry. Also, TV ads focusing less on the formulation of the product and more on smart women who do washes together as a group reinforce the low water usage as acceptable. These efforts seem to be having an effect, because Comfort One Rinse sales have increased 66% over the past three years.
Sometimes customers using a product or service may not be strongly tuned in to any particular group. Or the group may be defined by a negative experience. Airline travel is a case in point: Most travelers see themselves as sufferers of what is a rather boring experience at best, even in the relative comfort of business class. The answer here is to associate the offering with another, more positive identity.
Delta’s Innovation Class program tunes passengers in to a different, sharper identity. It targets budding entrepreneurs by offering them the chance to win a business-class seat next to a business guru as they travel with Delta. This associates the brand closely with high-performing business players, a social identity both the gurus and the entrepreneurs easily take on. It also transforms expectations about travel. No longer is a trip a dull episode. Instead it becomes a learning experience.
Social identities are more than a lens for understanding customers’ current social behavior. Marketers can actually create new social identities to both deepen the relationship with existing customers and attract new ones. We’ll turn now to what that involves.
Launching a Brand-New Station
To be sure, the idea that companies can profit from building a community around a social identity is not new. The Harley Owners Group, organized by the motorcycle manufacturer, is a textbook example of how a company can support a base of enthusiasts. Then there’s McKinsey’s famous alumni group, which plays an important role in winning business for the consulting firm. Companies have long been aware that it’s possible to create a shared identity around their products, services, and brands, and have made significant investments in doing so.
All those approaches, however, focus on people who are already customers. In effect, they’re tools for managing customer loyalty and getting more out of the current base. In most cases the communities tap into an existing social identity. The idea of going out to create a user group of not-yet customers for a new product that hasn’t been branded or launched probably doesn’t seem very practical.
But that assumption ignores a key lesson of most of the social psychology research on identity that has taken place outside the commercial sphere. Experiments around group formation suggest that social identities can be created almost at the drop of a hat and with very little effort. Some of these experiments date back to the 1960s and 1970s. Perhaps the most famous was conducted by Philip Zimbardo at Stanford in 1971. Creating new social identities, therefore, need not involve a big commitment.
Lessons of the Stanford Prison Experiment
In 1971, Philip Zimbardo ran what has become one of the most famous and controversial experiments in psychology. In it he recruited university students to take part in a role-playing exercise underwritten by the U.S. military, which wanted to understand sources of conflict between military guards and prisoners.
Zimbardo randomly assigned the participants to the role of either guard or prisoner. To his surprise, both groups took to their roles with gusto; not only did the guards rapidly become highly authoritarian, but the prisoners (after an initial rebellion) adopted passive, almost helpless attitudes at the same speed. In fact, the guard behavior became so extreme so quickly that the experiment was abruptly halted, after only six days, for fear of serious harm to participants.
Initially, commentators argued that the experiment demonstrated the pernicious effects of mob rule (it should be remembered that the civil rights movement was happening during this time). But social identity research since suggests a subtler interpretation: Those involved were behaving in accordance with their newly primed social identities of guard or prisoner. More than that, group members were trying to excel in these new identities. Hence, in the case of the guards, a rapidly increasing risk of brutality, torture, and psychological harm forced the experiment to halt.
Another lesson to be drawn is that these powerful behavioral changes were triggered very rapidly and required little investment in identity building. This suggests that people are quick to find and relate to new social identities. Creating a new social identity, therefore, is not necessarily an exercise in nation building.
Reflecting on all of this, we began to wonder: Could marketers inspire specific behaviors by quickly creating temporary groups with new and distinctive social identities? If so, marketers would have a powerful tool at their disposal. And if this could be done with new brands that had no previous customer relationships to draw on, any marketer, not just those with established businesses and deep pockets, could use it.
Creating New Identities (in Just 20 Minutes)
To test this idea, we invented a fruit smoothie brand professing to be both healthful and convenient, and invited consumers to help design launch initiatives for it. The participants were asked to comment on some draft marketing ideas.
The consumers were assigned at random to three groups. Those in the first group were told it was made up of people with superior creative skills, and those in the second that they were simply market research participants. Our aim was to create a stronger social identity for the first group, which was constantly reinforced throughout the experiment. For example, we gave the group a name, “20/20 Creative Vision Group,” which was continually on display to members. The idea was that this identity would encourage them to engage in high levels of creative behaviors, in the name of looking good compared with those outside the group.
The third group was given a strong identity in a different way. Its members were told that the firm’s goal was to launch a “pro-sustainability” brand, which would be socially and environmentally responsible, while the first and second groups were told that the goal was simply a successful launch. In the end we created two social groups with specific social identities, one tied to a quality (creativity) and one to a goal (sustainability).
The brand team requested from all participants certain behaviors, such as carefully reviewing the marketing materials, committing to buy the product, and promising to volunteer time and pledge donations to charities the brand was working with. We tracked those behaviors and also measured the strength of the consumer-brand relationship at the end of the study.
The results were clear: The participants in the groups with the stronger social identities engaged far more in the behaviors the company requested of them.
What’s more, those behaviors appeared very quickly. Participants were allowed to spend as much time as they wanted evaluating the supposed marketing materials for the product launch; on average, it took them 27 minutes to complete the entire exercise, which included reviewing two marketing initiatives and then working through a battery of questions and postreview checks. But among the groups primed with a clear social identity we usually began to pick up evidence of the desired behaviors just 15 minutes in. Overall, about 20 minutes was all it took to build a social identity powerful enough to prompt specific behaviors. Moreover, this effect occurred regardless of the participants’ prior attitudes—for instance, in the case of the pro-sustainability group members, whether or not they’d been concerned about the environment. The social identity alone was the driving factor. And these were behaviors that typically are not easily encouraged among consumers.
To put it another way: We didn’t need to find consumers with certain attitudes as a precursor to specific behaviors. Our “manufactured” social identities led them straight to those behaviors. And because we got these results with a brand that had not existed in consumers’ minds 20 minutes earlier, we believe these effects are within the reach of every marketer.
A few companies have already undertaken initiatives that resemble our experiment. Frito-Lay’s innovative Crash the Super Bowl contest, held since 2007, asks consumers to develop their own TV ads for Doritos and post them on a website. The best, as decided by voters on the site, are broadcast during the football game. Presenting a clear goal to the group participants—scoring the top spot on USA Today’s Ad Meter assessment of ad effectiveness for the evening—has helped the contest efficiently crowdsource commercials that keep up with expensively produced competition. Crash the Super Bowl spots have won first place on the Ad Meter on three occasions.
It’s surprisingly easy to make consumers switch identities and even to give them new ones. That presents marketers with a rich opportunity.
The contest is a good real-world example of marketing through the creation of a new social identity. It generates online engagement with the brand over the course of several weeks, as consumers post thousands of submissions and vote on them. It garners a lot of positive media coverage as well. Frito-Lay has inspired a group with a clear social identity tied to creativity, with which the company frequently interacts. Direct competition with ad agencies strengthens the group identity as members strive to secure status and distinctiveness. Who wouldn’t want to say they’re more creative than the world’s leading ad agencies? And that identity transforms the way those people relate to the Doritos brand.
When marketers want to get consumers to identify with a product, they typically think of building a community—a long-term commitment that may require a big investment. That can certainly be a valuable strategy. But a richer opportunity may lie in a less obvious lesson offered by the experiments we’ve just described: It’s surprisingly easy to make consumers switch identities and even to give them new ones.
Much of the noise that disrupts customer communications and brand experiences is, we believe, caused by the constant turning of the dial on a social identity radio. And if that is the case, then by engaging with the tuning process on a day-to-day basis and offering new identities, marketers can clarify and adjust those communications, to the benefit of customers as well as firms. That should be music to all marketers’ ears.