by David Burkus
But Ferriss had a plan for achieving optimal results from minimum effort there as well. He knew his target group specifically. “18–35 [-year-old] tech-savvy males,” he recalled. “Partially because I’m in that demographic.”2 And because he was in that group, he also knew that he didn’t need to spend time getting the attention of the big media players that most newbie authors salivate after. He didn’t need the New York Times to review his book, or to land an interview on Good Morning America. Instead, he went small. “I identified the primary channels through which I could achieve a surround sound effect,” he explained. “I identified that there were, let’s just say, 10 to 15 blogs and if I hit 50 percent of those on any given week, I would create the perception, and to some extent, the reality, that I was ubiquitous.”3 He believed he could make eighteen- to thirty-five-year-old tech-savvy males notice him as often as if he had chased down every possible publicity stunt, but without really doing so.
So Ferriss identified his precise targets, the ten to fifteen most popular niche websites in that demographic, and focused on developing relationships with them. He would go to conferences, meet writers from those publications for drinks, and ask them lots of questions about themselves and their work. “I never played the ‘I know what’s perfect for your audience’ card, ever,” he said. But eventually these reporters would ask Ferriss what he was working on. He would tell them he was working on a book and then give them a little about what it was about, then offer to send them a copy. “That was it,” he recalled. “That was the business plan.”4
It worked. Most of the journalists ended up writing about Ferriss or about an idea in his book, and most of them did it at the same time. Ferriss had created a media illusion among those eighteen- to thirty-five-year-old tech-savvy males that he was being talked about everywhere. Eventually, word began to spread. His book became a bestseller, which led to more publicity, which actually did lead to it being talked about in the New York Times and his being interviewed on Good Morning America. Ferriss had made himself look like he was everywhere without actually having to be almost anywhere.
And he is not the only one. Entrepreneur Andrew Davis used a similar strategy to launch not one but two businesses and grow them exponentially in a short period of time.
No matter where you begin, Davis started his career originally in television, either as a childhood commercial actor or as a producer and writer for the Jim Henson Company later on as an adult. But he quickly pivoted to working with Internet start-ups. “It was the first dot-com boom in 1998 and 1999,” he recalled. “All my friends who were in television were leaving to work in marketing, and they were making four times as much as me. It felt like everybody I knew.”5
So Davis left television and joined a start-up. “That’s where I learned how to apply all these things from television to marketing,” he said. He also learned quickly that many marketers were doing it poorly. As a result, it wasn’t long before he went out on his own. In 2001, he started a marketing agency called Tippingpoint Labs. They had great ideas about storytelling and content marketing, but they needed clients. Like a lot of agencies starting out, Davis and Tippingpoint would take whatever work came in the door and were simply working to grow the company. But after several years of this kind of haphazard growth, they felt something was off. “We decided we needed to get much more focused, because we were hiring more people and we were doing disparate work for people who we didn’t really love,” Davis said. “So we decided to take a much more strategic approach to growth and in growing the business.”6
That strategic approach came in the form of selecting specific industries and speaking just to them. Literally speaking to them at their conferences and events. Davis had already seen that speaking at conferences could be a useful tool for gaining more clients, but Tippingpoint’s scattershot approach had kept them running from niche to niche and never really establishing a foothold as the go-to agency for anyone.
The first industry they chose was construction and housing. “This was 2009; the housing market had kind of imploded,” said Davis, but therein lay the opportunity. “They were struggling to figure out what they were going to do from a marketing standpoint, and we thought it was a great opportunity for us to stand out.”7
So Davis looked up the biggest trade publisher in the construction space and reached out to the editor in chief of their different magazines. He then gave him his pitch: “Look, I know you have your annual event coming up, and I know the budgets are tight. I’d love to speak at your annual event. I speak a lot, but I’ve never spoken in this vertical, so to lower your risk you can just pay my travel.”8 This approach worked: Davis got the invitation to keynote the event. He was going to give the closing address of a three-day conference—the climax of the event.
To get up to speed, he showed up at the conference on the very first day. “I would basically sit in the audience of every session I could possibly attend for two days,” he laughed. “I’d learn as much as I could about the industry; I’d learn all the acronyms and vernacular of the industry. Then I’d take my normal domain expertise in marketing and apply it to their market.”9 It paid off. Davis’s talk was well received, and afterwards he was flooded with prospective clients . . . and invitations to speak at other industry events.
Davis even found that his strategy was working for more than just events. “If you speak at a publisher’s event, they invariably write about you on their website and in their magazine,” he explained. “And they distribute the videos and all of the sudden, even people who weren’t at the event start either inviting you to speak or are aware of the service you provide or of your expertise in the market.”10 Like Tim Ferriss with tech-savvy males, Davis’s small actions in targeting the major voices in an industry had turned him quickly into the hot topic.
And Tippingpoint Labs didn’t stop there. The next year they repeated the formula, but this time did so in financial services. Then the following year they targeted fast food and quick service restaurants. Then travel and tourism. Each year they extended their expertise into a new vertical and quickly became the most talked-about option. And each time they did this by targeting the biggest, most connected trade publisher in the space and appearing at its event, then letting speaking invitations and referrals push them easily into the lead.
By 2012, Davis was spending most of his time during the workweek traveling to events in these different verticals and generating new leads for the agency. That was when he pivoted again and started a new business. Davis sold his interest back to the agency, wrote a book on branding and marketing, and started out on his own with a company of one. One at a time, Davis started revisiting the same contacts he had first made at trade magazines and pitched his speaking. “I basically said, ‘I have a book and I already know your industry, and you know me well; you’ve seen me speak,’” Davis said. “‘Except I’m not free anymore.’”11
To his pleasant surprise, the strategy worked. More and more referrals for speaking started coming in, and it was largely because people were spreading the word in between events about this new expert they were hearing from. Sometimes he would hear from prospective clients inviting him to speak, and they would say he was recommended by three different people. “They’re like, ‘Wow, I just heard about this guy yesterday from another person. I’ve never even heard of this person, but now everybody’s telling me they’re reading this book,’” Davis said.12
Like Tim Ferriss, Andrew Davis has had an experience that’s a testament to a weird phenomenon: appearing to be everywhere without actually being everywhere. Even if you’re not selling a book, or a speech, or a long-term marketing contract, Ferriss’s and Davis’s experiences suggest that you can create the impression with an individual or a group that you are much better connected or better known than you really are. And the research into social networks seems to support this. The truth is, we routinely make judgments about other people, products, or trends based on what the social networks around us appear to be
supporting.
But it is also possible that those same networks are tricking us into judging something as more popular than it is. It works a bit like an optical illusion; in fact, researchers refer to it as the majority illusion.
When Majorities Aren’t
To understand how the majority illusion works, we first need to revisit a concept we have already covered: super-connectors. Recall that super-connectors are people on the highest point of a power law in terms of connections. The number of people in their network dramatically exceeds the number in the average person’s network. Most people have a relatively small number of connections, and only a few people (such as super-connectors) have huge numbers of friends. In addition to super-connectors perhaps making you feel like you have a small network compared to theirs, the very existence of super-connectors skews the averages and makes almost everyone feel like they have a smaller network than their peers. The reason is quite simple: their network is smaller than average. In a network, most people have fewer friends or business contacts than the average.
This phenomenon is what those who study networks refer to as the friendship paradox. First asserted by Scott Feld, a sociologist at Purdue University, in his paper “Why Your Friends Have More Friends Than You Do,” the friendship paradox stems from a contradiction to our optimism (or narcissism).13 When surveyed, most people believe that they have more friends than their friends do. But when you sample a community or a cluster inside a social network, you find that most people have fewer friends than the average.
To better explain the paradox, consider an analogy to height. If you are a male and you measure the height of all of your male friends, you will find that the average is about 5 feet 9½ inches. If you plot the heights of all your friends on a graph, you will find this 5 feet 9½ inches to be pretty much the center, with a fairly even distribution on either side (the bell curve or inverted U shape we have discussed before). But now suppose that one of your friends is a lot taller than you. Not just a foot taller, but closer to 100 feet taller than you. This one person’s height would dramatically skew the average. His height would ruin the average, pushing it much higher than 5 feet 9½ inches. Thus, you could confidently say that, on average, your friends are a lot taller than you—your hypothetical giant friend skews the average to guarantee it.
This is what is happening in social networks. Because a small group of people are capable of maintaining an especially large number of contacts, when you look across your network, you will find that, on average, your connections have more connections than you do. You might still have more connections than most, but the presence of those connectional giants will make your network appear smaller than average.
This dilemma became even worse with the invention of social media. The researchers Naghmeh Momeni and Michael Rabbit, both of McGill University, studied millions of Twitter users and hundreds of millions of tweets.14 They found that the same power law applies inside online social networks. A small band of elite Twitter users have millions of followers; everyone else has far fewer. Thus, these Twitter “millionaires” skew the average enough that we can confidently say that the average Twitter user that you follow has more followers than you do (unless you happen to have 155,657 followers).
The presence of super-connectors (and the fact that connections follow a power law instead of a normal distribution) skews our perception of the network around us. But as Tim Ferriss and Andrew Davis learned, it is possible to play to this skew and appear more popular or more in demand than you really are.
This possibility was first proven only recently, by researchers at the University of Southern California led by Kristina Lerman.15 Lerman and her colleagues were playing around with the friendship paradox and its connection to the idea that since we as humans can’t see what our entire network is up to, we have to process interactions as they come and assess what the entire group is doing based on those people we interact with. Given this, it stands to reason that well-connected members of our network might skew our perception of how popular an idea or person actually is.
To test this, the researchers first created a model. The model showed a small-scale network of just fourteen people, with connections between people determined randomly. But what wasn’t random was how many connections each person had. In the model, they were deliberately made to vary wildly—from just two connections to half the network. They then chose three people inside the model to be “active”—their term for any attribute to be represented, such as having red hair, owning an iPhone, or recommending Andrew Davis to speak.
When they chose the lowest-connected members of the network to be active, the number of people connected to other active persons through those connections was low . . . less than half of the total population. But when they chose the three highest-connected members, it was pretty easy to give off the appearance of popularity—everyone was connected to at least one active person, most were connected to two, and a few were even connected to all three active people. If you are sitting in this model and trying to judge the popularity of an idea, it is highly likely that you are hearing multiple people talk about it. Which can make it look like everyone is talking about it—even if only three people really are. Like tech blogs and Tim Ferriss, suddenly a minority of voices become the loudest sounds echoing throughout an entire network.
Having shown the phenomenon in small models, the researchers then increased the scale. Using computer simulations, the researchers built models as big as 10,000 people, all with varying degrees of connectedness. Not only did the majority illusion hold true, but in some models it was enhanced by bigger networks (since bigger networks meant a greater degree of variance when it came to the number of connections that each person had).
Building models inside a computer was one thing, but testing it in the real world was another. The researchers needed to know that the same phenomenon held true when the network scaled from 14 people to 14,000 or to 14 million. To do these tests, they collected data from three readily available networks. The first was the network of coauthorship of academic papers among high-energy physicists (which, as we’ve seen, is a great source of network data reflecting connections and collaborations). The second was the follower network of the social news aggregator Digg.com, and the third was the network of links among political blogs.
Among all three networks, they studied only mutual connections. In each case, the majority illusion held. Perhaps the most drastic example was in the world of political writing. “The effect is largest in the political blogs network,” they wrote. “As many as 60%–70% of nodes will have a majority of active neighbors, even when only 20% of the nodes are active.”16 Just 20 percent of the political blogs could put forth an idea and the perception of any given node in the network would be that the majority of the network shared that belief. In other words, the majority illusion explains how easy it is to trick a population into believing something is true and widely believed, when the reality is just the opposite.
Just like the friendship paradox, the majority illusion happens because of the vast range of connections between different people in a network. Those with a lower number of connections, observing their more-connected colleagues put forth an idea or trend, conclude that the idea must be more popular than it really is. This can lead them to feel peer pressure (even though it’s not a majority of their peers), and the idea or trend gets adopted even further. The majority illusion can become a self-fulfilling prophecy. And while we’ve seen how individuals like Tim Ferriss and Andrew Davis used it to create a self-fulfilling illusion of popularity, it turns out that whole companies have used the majority illusion, albeit unknowingly, to rapidly scale their business. It’s appropriate that perhaps the most drastic example of the majority illusion comes from the world’s largest online social network—Facebook.
The World’s Largest Small Community
Facebook wasn’t supposed to become the world’s largest online community, at least not at first. At first, it was just another
dorm room side project of Harvard undergraduate Mark Zuckerberg and a few of his friends. Zuckerberg arrived on Harvard’s campus in Cambridge, Massachusetts, in September 2003 and quickly got to work turning intriguing ideas into software and websites.17 Within the first week, Zuckerberg had built a program called “Course Match.” The idea behind it was to give his fellow students a fun way to pick classes based on who else was taking them. Students could click on a course to see the entire list of who had signed up or click on a specific person to see what other courses he or she was enrolled in. Did you want to know what other courses that cute girl in American History was taking? No problem: Course Match could tell you. It was a program that worked particularly well for the status-aware students of Harvard, who judged classes not on the content or the professors, but on who else in the community was enrolled. Almost immediately, hundreds of students had started using Course Match. The program was a hit. Zuckerberg’s next project, however, wouldn’t go over so well.
For his encore performance, Zuckerberg started a project he called Facemash. At the time, for each of its dormitories, Harvard had online “facebooks”—most of them no better in quality than a middle school yearbook—featuring student photos to make it easier for students to get to know each other. Zuckerberg decided to use these photos to present a more pressing question, especially among eighteen- and nineteen-year-old college students: who was the hottest person on campus? To do this, Zuckerberg created a computer program similar to how the worldwide ranking of chess players was determined—that is, each time one player played another one, the result affected the rankings—to rank the photos in as many online facebooks as he could find. The security and privacy for each dormitory facebook varied. With some of them, the photos were publicly available, with others he hacked into the website remotely, and with others he sneaked into the dormitories in order to use the private network and access the photos.
