Social Networking Is Not a Business*
*But It Might Be Soon.
KickApps is an 80-person social-networking startup with its head office in a loftlike space just off Fifth Avenue in New York. In less than two years it has created the underlying structure for more than 20,000 social-networking sites–“mini-Facebooks” with an aggregate of 300 million page views per month. You’ve probably never heard of it.
KickApps gets a fraction of the press coverage of a giant like Facebook, but its growth has been sufficiently impressive that venture firms like Spark Capital and Prism VentureWorks have backed it with $18 million in startup financing, hoping for the payoff of a monster IPO. Its software allows companies to quickly roll out social networks with many of the features of Facebook or MySpace. Its clients–which include local radio stations and newspapers, national networks like NPR and ABC, and brands like AutoByTel, Harley-Davidson, and Kraft–want to offer fans a place to gather and share their love of a team, a product, or anything else.
KickApps’ CEO is Alex Blum, formerly of JumpTV, an online television service specializing in sports. “We have 35 programmers working in this office,” says Blum, leading a reporter through a sea of desks and flat-screen monitors, “and we only have two marketing people. We don’t really have to sell our product.”
Timeline of key events in the rise of social-networking sites.
See a comparison of Myspace and Facebook’s traffic and advertising growth.
View a graph on U.S. ad spending on social networking sites relative to U.S. online ad spending.
View a graph on worldwide online social-networking advertising spending.
Like most of the social-networking sites enjoying huge growth, KickApps is giving its product away, expecting that the communities built around it will generate ad revenues. It’s a model that stirs memories of the first Internet bubble: build the user base and hope the money comes–from an IPO, a buyout, or ads. At this point, KickApps does not reveal revenue figures, or even what kind of a cut it is taking from the ads. That, too, brings back memories: staying mum about revenue was always a sign that there wasn’t much to talk about.
Many Users, Few Dollars
Social networking is the fastest-growing activity on Web 2.0–the shorthand term for the new user-centered Internet, where everyone publicly modifies everyone else’s work, whether it’s an encyclopedia entry or a photo album. The growth of social networking is astonishing, and it has spread to sites of all sizes, which are increasingly intertwined as platforms open (see “Who Owns Your Friends?”). Even small players are soaring.
Ning, for example, is similar to KickApps but caters to individuals. Founded in 2004 by Netscape’s Marc Andreessen and former Goldman Sachs analyst Gina Bianchini, it has been backed with $104 million in venture capital by a variety of firms, including Legg Mason. “We’ve got 267,787 sites,” boasted Bianchini in May. “And we’re adding 1,500 to 2,000 a day.” ComScore, a firm that measures Internet usage, reports that the Ning domain, on which all the sites reside, sees three million unique visitors a month.
Meanwhile, Bebo, a social-networking site more popular abroad than in the United States, sees more than 22 million visitors a month. (AOL bought it for $850 million in March.) Club Penguin, a network for kids, sees five million. LinkedIn, a business-networking site, gets nearly five million unique visits.
But that’s just the smaller players. Facebook, according to ComScore’s latest research, saw 33.9 million unique U.S. visitors in January 2008, nearly double the number from the previous January (but down by about 2 percent from December 2007). MySpace doubled Facebook’s numbers again, with nearly 72 million unique visitors in the same month.
Nevertheless, the sites seem largely incapable of generating revenues commensurate with their popularity.
Last year, Microsoft bought a 1.6 percent stake in Facebook for $240 million, giving the company a dubious valuation of $15 billion. But Facebook is likely to lose $150 million this year, according to a January conference call heard by Kara Swisher of All Things Digital, a Wall Street Journal-affiliated site devoted to “news, analysis, and opinion about the digital revolution.” That’s based on projected earnings–before interest, taxes, depreciation, and amortization–of $50 million and an expected $200 million in capital expenses, including new servers.
Revenues for MySpace parent Fox Interactive Media fell $100 million short of predictions this year, apparently leading to the dismissal of the chief revenue officer. And Google met with disappointment after paying $900 million in 2006 to get a piece of MySpace’s traffic, buying the right to deliver ads for three years against keywords entered on the networking site. “I don’t think we have the killer best way to advertise and monetize social networks yet,” said Google cofounder Sergey Brin in a call with investors after Google announced its fourth-quarter 2007 results.
Ning does not release numbers about ad sales. All Bianchini will say is, “The number of networks we have is our leading indicator.” If Ning’s experience is anything like MySpace’s and Facebook’s, its leading indicator just indicates a lot of users.
Lookery, an advertising network that buys ad space on Facebook in bulk, has been reselling that space at 13 cents per thousand times an ad is served, or in ad industry jargon, at a $0.13 cost per mille (CPM). Facebook sets a minimum CPM of $0.15 for its “social ads,” which allow advertisers to target ads to Facebook users and groups according to characteristics like location and age. And over the last year, MySpace has lowered its banner-ad rate from a CPM of $3.25 to one of less than $2. By way of comparison, a banner on Mashable, a blog covering the world of social networking, has a CPM of $7 to $33, depending on its size. Websites with clearly defined audiences of executives and technologists who purchase corporate products and services, such as TechnologyReview.com, do best of all. Technology Review’s site boasts a CPM of $70.
But even low rates haven’t been enough to lure advertisers and media buyers to social networking. “A lot of advertisers are very hesitant to get into MySpace,” says Anthony Acquisti, who oversees strategy for emerging media at OMD, an advertising agency in New York. “We’ve even flat-out told interested brands, ‘You don’t want to be there.’”
Why not? The problems with social-network advertising revolve around three main issues: attention, privacy, and content.
In the last week of April, around 400 people who spend their days trying to figure out how to make money in social networking gathered at the Skirball Cultural Center in Los Angeles. The conference went by the not very catchy name of EconSM, short for Economics of Social Media.
The point of the conference was clear enough. As Kara Swisher, one of the panel moderators, joked on her blog: “I’m in search of the elusive profit, which I don’t think I’m going to find.”
Almost every player in the game was represented: smaller companies that sent their CEOs, like Alex Blum; investment banks that wanted to take them public; and companies like Yahoo, AOL, and Fox Interactive Media, which were in the market for acquisitions. (Facebook sent no one.) “This is a huge conference,” Blum said. “All the people we work with are here.”
But they didn’t seem very engaged. Audience members were jumpy, posting updates to the microblogging service Twitter, checking e-mail, reading blogs, dipping into the newspaper, and–occasionally–listening. Specific problems addressed in panel sessions quickly sorted themselves out into a general problem and a general response. People weren’t paying attention to the ads (as, indeed, people at the conference weren’t much paying attention to the panels). One panelist, Seth Goldstein, put it this way: “Right now, ‘social’ advertising is anything on a social-networking site that users are pretty good at ignoring.”
Goldstein should know, since his company, Social Media, sells advertising linked to the applications developed for Facebook and MySpace–products like Scrabulous and Compare People, which are hugely popular among the sites’ users. “The trouble,” says Goldstein, “is we’re putting ads up in front of users, where they can ignore them. We’ve got to get them between users.”
Goldstein’s comment had the air of a slightly worn sound bite, but it did acknowledge a problem that outside observers describe more bluntly. “It’s a really bad place to advertise,” Jason Calacanis, founder of Webblogs and Mahalo.com, says of social-networking sites. As he wrote in an e-mail, members of social networks “are busy in conversations and don’t want marketing messages.”
Compare the situation of social networks with that of Google, which manages to make money putting ads in front of users.
With about 140 million unique visitors per month, Google earned $16 billion in 2007, largely from ads that people did pay attention to. (It may bear mentioning that Facebook recently hired Sheryl Sandberg away from Google, whose phenomenally successful ad program she had led.) Google’s AdWords auction sells ads on a cost-per-click basis: advertisers pay not for a thousand viewings but for each individual click on a particular keyword. Advertisers choose how much to spend over any period of time, and they can influence the placement of their ads by paying more. Bids vary according to keyword, of course, but they were averaging around 70 cents per click in the first quarter of 2008.
Advertising on Google works because visitors come to Google looking for specific information. If a user who types “scooter” in the site’s search field is hoping to buy a scooter, the keyword ads that appear at the right of the search results can be more useful than the results themselves. In social networks, on the other hand, users show up to find friends; ads are, at best, irrelevant to that goal. The click-through rates on social-networking sites bear this out. While around 2 percent of Google users actually click on a given ad (and the number is much higher when users are conducting searches for purchasing reasons), fewer than .04 percent of Facebook users do, according to a media buyer’s report obtained last year by the Silicon Valley blog Valleywag.
Users in the Crosshairs
When social-media insiders are asked how advertising could capture users’ attention, they always answer, “Targeting.”
Targeting is at the core of traditional advertising; apparel manufacturers advertise in Vogue, for example, to reach readers interested in fashion. In the case of social-networking sites, targeting means sifting through the data in your profile to get an idea of what you’re interested in. Social networks know more about you, your preferences, and your behavior than most businesses, and profiles are generally considered, in the words of former Fox Interactive Media executive Ross Levinsohn, “digital gold.” Mining that gold is the best way for a social-networking site to make money–but, given users’ attitudes toward privacy, the trickiest.
Startups that help advertisers and marketers better target the users of social-networking sites are fashionable investments for venture capitalists in North America and Europe. Such startups hope to sell advertisers detailed information about individual social networkers. They include the brand-new 33Across (which we profile in our list of 10 notable startups, which begins on page 50) and the more established Finnish company Xtract, which counts Vodafone, T-Mobile, and Blyk among its customers and has begun selling its software to advertising agencies and online marketers and publishers.
For social-networking sites, targeting will necessarily entail getting “between” users, as Seth Goldstein put it. You come to a social network because you are interested in your friends; ergo, the thinking goes, in order to get your attention, advertisers need to let you know what your friends are buying or thinking about buying, or they must somehow get you to send each other ads. It’s either a beautiful idea or a creepy one, depending on whether you’re an ad executive or the user of a social network.
In November 2007, Facebook tried to get between its users with its Beacon program. Announcing the program in New York, Facebook founder Mark Zuckerberg declared, “The next hundred years will be different for advertising, and it starts today.”
Beacon was an advertiser’s dream–and, like many things that are good for advertisers, very annoying to ordinary folks. Working with commercial websites like Blockbuster and eBay, Beacon tracked Facebook users’ purchases and displayed them to their friends.
The problem was that users were enrolled in the program automatically. If a user went to, say, the Blockbuster site and rented a movie, that information was automatically sent to everyone in her Facebook network. (That’s what happened to Cathryn Elaine Harris of Dallas; she is suing Blockbuster for violating the Video Privacy Protection Act.) Online petitions and negative press ensued, and the program was clumsily scaled back. On the company blog, Zuckerberg wrote, “We’ve made a lot of mistakes building this feature, but we’ve made even more with how we’ve handled them. We simply did a bad job with this release, and I apologize for it.”
Still, “Beacon is alive and well,” says Chamath Palihapitiya, Facebook’s vice president of product marketing. “What happened was unfortunate,” he says. “We took a step back and tried to figure out how to improve it.” Now it’s an opt-in system, and users can choose what information to share–or whether to participate at all. About 30 companies are still with the service, he says.
Most of the industry members at EconSM liked Beacon, wished it had worked better, and felt it would work eventually; Goldstein called it a “sign of things to come.” But maintaining the user’s trust in how data is used is paramount, says Roger McNamee, a venture capitalist who made early bets on companies like Electronic Arts and Intuit. “Facebook is so much more personal than Google,” says McNamee, who invested in Facebook and is a confidant of Zuckerberg. “It really matters to people how their information is used.”
Not every attempt at targeting has aroused as much protest as Beacon. In 2007, MySpace launched its HyperTargeting system, which scans users’ profiles for information about their interests and demographics. It sorts the profiles into 10 rough categories–such as sports and entertainment–that are subdivided into more than 1,000 narrower categories, such as baseball or a specific film. (E-mail and personal messages are currently not scanned at either Facebook or MySpace.) Says Adam Bain, president of the Fox Interactive Media Audience Network, “People are essentially hand-raising every single day on MySpace and other social-media sites. What we want to do is take that and put it into easy-to-buy segments.”
Bain says MySpace did extensive research before the launch. “The users said, ‘I understand I have to live with ads, and I don’t mind them,’” he says. “The concept of relevance really resonated with users.” The algorithm is constantly modified by a 150-person team; it is already on its 12th revision. Although the program has not yet led to riches, “it has led to an unprecedented amount of advertisers coming to MySpace,” Bain says. “We’re getting blue-chip brand names like Adidas, Schwab, and Electronic Arts, Frito-Lay, Kraft, General Mills, and McDonald’s.”
Are those advertisers as excited as Bain is? “We’ve bought a little bit,” says Marc Ruxin, director of digital strategy and innovation at the advertising firm McCann. “It’s been okay.” Well, that’s better than nothing. Still, it doesn’t exactly settle the question of whether targeting, even if it avoids the worst of users’ privacy concerns, will ever be able to punch through the attention barrier.
Another problem that targeting may not be able to solve is the one posed by what advertisers call “content adjacency.”
Unlike a newspaper or television show, social networking is a medium whose content is deeply unpredictable. In the sports pages of a newspaper, an advertiser knows roughly what kind of material its ads will be running next to. But an enormous, highly visible brand may not want to risk seeing its ad wind up on a page such as that run by the actual Facebook group “I’ve Had Sex with Someone on Facebook,” which at press time had 59,353 members. Or consider the MySpace profile (turned up after about two minutes on the site) of 18-year-old “Nikki AKA Death Angel!,” which is adorned with the motto “Don’t fuckin fuck with ninjette bitch we’ll cut ur fuckin head off an give it to ur momma.”
This is not content that commands high rates, although certain buyers mind less. “Right now, the low-hanging fruit is entertainment, because they’re agnostic about content adjacency,” says Goldstein. Indeed, Nikki’s badass profile features an ad for the Warner Bros. film Get Smart. But even entertainment companies are steering clear of the user-generated communities offered by Ning and KickApps. “It’s not a controllable universe right now, with the porn sites and such,” Ruxin says. “It’s a blind buy.”
Not everyone is so pessimistic. Andrew Braccia, a partner at Accel, one of Facebook’s early investors, thinks advertisers will eventually become more accepting of the “breathing, dynamic” nature of social networking and grow to understand that its unpredictability is part of its allure. And Facebook’s Palihapitiya, perhaps naively, doesn’t seem to think the adjacency problem will arise much on his site; Facebook, he says, has “a tremendous amount of user content moderation, with a very simple mechanism for flagging inappropriate material.”
Users’ ideas of what’s appropriate are hardly the same as advertisers’, though. Such arguments may not be enough to sway the enormous, image-conscious brands that drive the majority of the advertising market. And Palihapitiya, deliberately or otherwise, may be missing the point: advertisers dislike rude content not merely because it might reflect badly on their brands, but because people reading such stuff are probably not thinking about buying many things that advertisers are selling.
Still, backers of social networking feel strongly that so many eyeballs must have value. Braccia points out that while more than 6 percent of advertising dollars are spent online, 20 percent of media consumption now happens there. “It’s a significant opportunity,” he says. “We’re so young, so in our infancy here.”
“These sites are no different from traditional media properties,” says Paul Kedrosky, who writes Infectious Greed, a much-read blog on venture capital and the Internet. “We’re holding these sites to an absurd standard. The advertising allocations will follow the consumer, and right now they’re badly out of whack.”
Roger McNamee remains convinced that Facebook is too alluring, too useful, and too established not to be profitable somehow. The answer is out there, even if he doesn’t have it. “Someone,” says McNamee, “is going to have to get creative. I take it on faith that it will emerge. After all, I’m an investor. I’m hopelessly biased.”
Marc Canter has a few ideas. Canter, who cofounded MacroMedia, is now CEO of the company that produces the social-networking tool PeopleAggregator, which aims to allow communities, tools, search engines, and the rest of Web 2.0 to interconnect in one giant open mesh. He imagines ads of all kinds making up only about a third of revenue, with profits coming from a “long tail” of sources–from Craig’s List-style marketplaces to on-demand music downloads to branded apparel to ad-free premium services.
Chamath Palihapitiya expects Facebook to generate revenue by selling a variety of such services to users. The site has rolled out a “gift” program, in which friends spend real money to “give” friends virtual items, such as an image of a box of tissues with a get-well note. He also suggests that Facebook may at some point see revenue from ads served through applications on its site, a growing and potentially major source of income from which it currently gets nothing.
Perhaps most optimistic of all is venture capitalist Ron Conway, the subject of the book The Godfather of Silicon Valley, who has invested in Google, PayPal, and dozens of Web 2.0 companies. “MySpace projected it would do a billion dollars’ worth of revenue this year. They came up short and did $800 million,” he says. “Rupert Murdoch only paid $570 million for the whole thing. It’s been called the best acquisition of all time. I think Facebook is a couple of years behind MySpace but on the same trajectory. It’s a hugely monetizable business. I think it’s a slam dunk.”
A GLOOMY FORECAST
Can social-networking sites continue to make significant inroads into the U.S. online advertising market? The outlook is uncertain. A shaky economy and setbacks in targeted-advertising initiatives have caused leading online marketing research firm eMarketer to project more modest revenue growth for social-networking sites over the next four years than it had previously predicted.
THE GLOBAL VIEW
Social networking is a global phenomenon, and reaching users outside the United States will become increasingly important as advertising dollars flow to Western Europe, Asia, and beyond.
Things Fall Apart
The ghosts of vanished giants haunt social networking. So many formerly great Internet companies are struggling or dead. Consider CompuServe, AOL, Netscape, Napster–even Yahoo. Lycos, a search engine that was sold to Terra Networks in 2000 for $12.5 billion, was sold to a Korean firm for $95 million four years later.
What CompuServe and many of the others have in common is that they were portals: gateways to the Web. Facebook wants to be something similar: more than just a useful and fun social tool but the first page people open on the Web, and the platform they use for all their other communication on the Internet.
As would-be portals, however, social-networking sites are vulnerable to one of the problems that brought down those earlier Internet businesses. The portals were “walled gardens” where inexperienced Internet users congregated for a time but where they became restless at last–leaving for the wider, wilder Web. Facebook and MySpace understand this and are now struggling to achieve an appropriate balance between openness and control.
They’re also struggling with faddishness. Danah Boyd, a doctoral candidate at the University of California, Berkeley, studies social networking as a cultural phenomenon. She describes online hot spots as though they were popular pubs. “It’s supercool when all of your friends go there,” she says. “Then all sorts of other people come in. Even if the pub doesn’t start feeling physically crowded, it starts feeling socially crowded when your ex is at the other end of the bar talking to some creep who brought his fellow gang members. How long until you say, ‘Enough–I’m outta here’?”
Several attendees at EconSM took the same flight home, and anyone paying attention on that red-eye from Los Angeles to New York got a lesson on social networking’s place in modern life.
Just before the plane began its descent, a 28-year-old woman named Erin fainted on the way to the bathroom. She was possibly overtired, or maybe weirded out by the inhumane crush of economy class. Even she didn’t really know what happened. By the time we were on the runway, she had regained her senses. Her first question to the flight attendant was, “Did anyone get my phone?”
As soon as the attendant handed her her iPhone, she opened it up and went right to her Facebook account. She wasn’t looking for ads and she wouldn’t have noticed one, unless it annoyed her by getting in the way. She wanted to reach her friends, and that was all.
Bryant Urstadt has written for Rolling Stone and Harper’s.
The inside story of how ChatGPT was built from the people who made it
Exclusive conversations that take us behind the scenes of a cultural phenomenon.
Sam Altman invested $180 million into a company trying to delay death
Can anti-aging breakthroughs add 10 healthy years to the human life span? The CEO of OpenAI is paying to find out.
ChatGPT is about to revolutionize the economy. We need to decide what that looks like.
New large language models will transform many jobs. Whether they will lead to widespread prosperity or not is up to us.
GPT-4 is bigger and better than ChatGPT—but OpenAI won’t say why
We got a first look at the much-anticipated big new language model from OpenAI. But this time how it works is even more deeply under wraps.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.