Adam Cohen
The Perfect Store
“I wanted to do something different, to give the individual the power to be a producer as well as a consumer.”
By including their e-mail addresses, Omidyar allowed users to communicate directly among themselves to solve each others’ problems. And Omidyar created a message board that allowed users to share information with the entire community without routing it through him. The more self-sufficient the users became,
But his interest in community was more than just pragmatic. At a time when the Internet was endlessly compared to the Wild West, Omidyar wanted his corner of cyberspace to be a place where people made real connections with each other, and where a social contract prevailed. He wanted it to operate according to the moral values he subscribed to in his own life: that people are basically good, and given the chance to do right, they generally will.
EBay had thrived because it fit this classic model: Omidyar had built both the commercial crossroads and the larger community that always came with it.
It is widely recognized today that, in purely business terms, such “many-to-many” sites are far more powerful than traditional “one-to-many” sites, in which a company like Amazon sells directly to consumers.
eBay increases “social utility,” as the economists put it, making people, as a whole, happier than they would be without those goods.
The implications of the eBay model—of Omidyar's original conception of a perfect, global marketplace that everyone comes to on an equal basis—are revolutionary. EBay gives individuals a degree of economic independence that was impossible before the Internet.
how off-base eBay's first business plan was about which part of the company would be profitable;
Part of eBay is in its California headquarters, but much more of it is not. EBay exists, like a religion or a social movement,
wherever its adherents happen to be. To get inside eBay, I had to travel to the places where it manifests itself —the office of a rare-autograph dealer who is gradually moving his business onto eBay; a convention of clothing-iron collectors in Kansas City, whose world has been transformed by eBay; and the London offices of eBay-U.K., where a webmaster labors to keep Americanisms off the site.
Omidyar used his skills to get his first paying job, computerizing his school library's card catalog for six dollars an hour.
finishing up his undergraduate degree at the University of California—Berkeley.
As it happened, Ink Development had also put together some software tools for online commerce, and this marginal project now seemed to be the most promising part of the business. The company relaunched as eShop, an electronic retailing company.
It was while Omidyar was at General Magic, working with both the Internet and with people, that he created AuctionWeb. It started, legend has it, with PEZ. In the summer of 1995, Pierre Omidyar was having dinner at home in Campbell with his fiancée, Pam Wesley. Wesley collected PEZ dispensers, and she mentioned that since they had moved from Boston to Silicon Valley, she was having trouble finding fellow collectors to trade with. It occurred to Omidyar that the still-fledgling Internet could provide the answer.
The PEZ dispenser story has been told and retold in countless popular accounts of eBay's history. But it is, Omidyar concedes, the “romantic” version of eBay's founding.
Omidyar kept hearing about company insiders, often friends and family of the founders, getting rich through stock purchases that were not available to average investors. This was standard practice for IPOs, but it struck him as unfair.
He was used to tinkering with Internet applications in his evenings and on weekends. He had already written a chess-by-mail program, which he was offering for free over the Internet. He had also completed the coding for a program he was calling WebMail Service, which allowed owners of small-screen computer devices like the Newton to get access to Internet pages through standard e-mail. More recently, he had created WebMail Watch Service, which monitored web pages users were interested in, and notified them when the pages had changed. With Labor Day approaching, Omidyar made the program for a perfect marketplace his project for the long weekend.
The computer code Omidyar wrote let users do only three things: list items, view items, and place bids. The name he chose was as utilitarian as the site itself: AuctionWeb. Since AuctionWeb was only a hobby, and he intended to offer its services for free, Omidyar tried to keep costs low.
Rather than create a new website, he added AuctionWeb to one he was already operating. That spring, Omidyar had formed a sole proprietorship for his web consulting and freelance technology work, which he had named Echo Bay Technology Group. The name was not a reference to Echo Bay, Nevada, the wilderness area near Lake Mead, or to any other real-world Echo Bay. “It just sounded cool,” he says.
Omidyar registered what he considered to be the next best thing: eBay.com.
After its traffic-free Labor Day launch, AuctionWeb started to attract a slow trickle of visitors.
The list was not a representative sample—it was every non-computer-related item on the site. A week later, Omidyar updated the list, which had grown from eighteen to thirty items, a 66 percent increase, in just seven days. Among the new listings: a 35,000-square-foot warehouse in Caldwell, Idaho, for which the bidding started at $325,000.
Throughout the fall, both listings and traffic on AuctionWeb increased steadily. While Omidyar was putting up his newsgroup posts, AuctionWeb was also starting to benefit from the marketing force that would drive its growth for years to come: word-of-mouth publicity. Computer geeks and tech-savvy bargain hunters were e-mailing one another the AuctionWeb URL, and inserting hyperlinks on their websites that took web surfers directly to the AuctionWeb home page.
Best's fee hike changed everything. “That's when I said, ‘You know, this is kind of a fun hobby, but two hundred fifty dollars a month is a lot of money,'” Omidyar says. To pay the bills, he started to charge AuctionWeb users—"basically out of necessity,” he says.
no market research, Omidyar decided he would not charge buyers at all, and that he would not charge sellers to list items. The only fees would be what he called final-value fees, which would be a percentage of the final sales price. The fees, he decided arbitrarily, would be 5 percent of the sale price for items below $25, and 2.5 percent for items above $25.
The amounts were not large, and the trappings were not fancy. Some of the envelopes contained dimes and nickels Scotch-taped to index cards.
That put his fledgling little website in a category almost by itself: it was one of the very few Internet companies to be profitable from its first month of operation.
AuctionWeb was to have any chance of taking hold, establishing trust and confidence was essential. Early on, Omidyar set out ethical guidelines for the AuctionWeb community to follow. In his experience, he said, people are generally good. He advised users to treat other people on the site the way they themselves wanted to be treated, and when disputes arose, to give the other person the benefit of the doubt. Omidyar's injunction was essentially the golden rule transported into cyberspace. It was the value system his mother had instilled in him, and one he tried to follow in his own life. “Some people say, ‘Isn't that trite, it's like a Hallmark card,'” he says. “But I think those are just good basic values to have in a crowded world.”
Trust on the site was so high in the early days, and the feeling of community so strong, that it was common for sellers to ship items even before they had received bidders’ payments. Still, the harmony Omidyar hoped for did not always prevail. When buyers and sellers disagreed, they usually contacted Omidyar directly—easily enough done, since his e-mail address, Pierre@eBay.com,
Omidyar's routine when he received an e-mail with a complaint about another user was to respond to the author, send a copy of the e-mail to the other person in the dispute, and tell them both, “You guys work it out.” The parties usually resolved the matter on their own, but Omidyar realized he had to come up with a mechanism for enforcing good behavior. Unlike most companies, AuctionWeb was not able to control the quality of its service. “The brand experience” on AuctionWeb, Omidyar observed, was “defined by how one customer treats the other customer.”
This grand hope depends on your active participation. Become a registered user. Use our Feedback Forum. Give praise where it is due; make complaints where appropriate…. Deal with others the way you would have them deal with you. Remember that you are usually dealing with individuals, just like yourself. Subject to making mistakes. Well-meaning, but wrong on occasion. That's just human. Through the Feedback Forum, the complaints that landed in Omidyar's e-mail in box would be brought out into the open. The entire community would know about them and have an opportunity to deal with them appropriately.
He decided that when users’ Feedback Forum ratings got too low—negative four or less—they would be banned from the site. Omidyar arrived at the cutoff point of negative four without much deliberation—it just struck him as the point at which his assumption of goodness was sufficiently rebutted—and he did not reveal it to users. But even years later it would remain the number that caused eBay to “NARU” someone—to make him or her Not a Registered User.
Like the Feedback Forum, the Bulletin Board was designed to limit his role and place more of AuctionWeb's administration in the hands of the community. Omidyar did not have time to explain to each individual user how to write a listing in HTML, or to give advice on bidding strategy. The Bulletin Board was in the tradition of the Usenet newsgroups Omidyar had long used, a place for people to gather, share information, and ask for help. As soon as the Bulletin Board went up, the questions poured in. What was the best way to ship? What should a seller do when a high bidder disappeared? The answers came just as quickly. “If someone came on and said, ‘Please help me,’ there were twenty-five people who would rush to help,”
In June, when revenues doubled for the fourth consecutive month, topping $10,000, Omidyar decided it had become a real business. “I had a hobby that was making me more money than my day job,” he says. “So I decided it was time to quit my day job.”
“But I knew I wasn't going to be able to put together a business plan.” He started looking for someone who could. Omidyar thought immediately of Jeff Skoll, a Stanford MBA he had met through friends two years earlier. Skoll, a slightly built, hyperkinetic Jewish Canadian, was a born entrepreneur. His father sold industrial chemicals, and by age twelve Skoll himself was going door-to-door selling Amway products in Montreal.
Skoll eventually realized that “what Pierre was doing was a lot bigger than just a simple website.” In February 1996, Skoll had agreed to do consulting work for AuctionWeb. By August, the site was so successful that Skoll quit his job and signed on full-time. In Skoll, Omidyar found a yang to his yin. “It was the perfect balance,” says Omidyar. “I tended to think more intuitively, and he could say, ‘Okay, let's see how we can actually get that done.'”
Omidyar suggested expanding the search for permanent quarters to the city of Campbell. A sprawl of suburban homes and office parks, Campbell paid tribute to its long-lost agricultural heritage every May, when it played host to California's largest prune festival. Campbell was not as fashionable as Palo Alto, and it was certainly not the epicenter of the boom. But what Campbell lacked in hipness and frenetic activity, it made up for with more practical attributes. Rents were lower and, more important, there might actually be some offices to be had. From Omidyar's perspective, Campbell had another advantage: he lived there.
‘I'm wearing a lovely flower print dress and I just got through milking the cows,'” he says. “That's how it started about Uncle Griff actually being a cross-dressing bachelor dairy farmer who liked to answer questions.” The legend of Uncle Griff grew quickly. On the Bulletin Board, Griffith referred to his AuctionWeb persona in the third-person: Uncle suggests you do this; Uncle would never do that.
It was Jeff Skoll. He wanted to know why Uncle Griff had stopped posting on the boards. Griffith was stunned that his absence had been noticed at AuctionWeb headquarters. Skoll had an assignment for Griffith. AuctionWeb was receiving fifty to one hundred e-mails a day from users, and it had no customer-support staff. Skoll was prepared to pay Griffith to answer the e-mails on a regular basis, and to keep up his presence on the Bulletin Board. Griffith was up for it, but he wanted to make
Griffith became AuctionWeb's second part-time employee, at a salary of $100 a month, and its first official customer-support person. Skoll asked Griffith to select an alias to use as his AuctionWeb identity.
Ruby became eBay's second “remote,” as its employees outside of Silicon Valley came to be known. She started out part-time, but within two weeks of Skoll's call she quit her programming job and began working for AuctionWeb full-time. Skoll asked Ruby, as he had asked Griffith, to choose an AuctionWeb identity. She became Louise@eBay.com and remained Aunt Patti on the boards, both personas that would become famous in AuctionWeb's early days.
Mary Lou Song. Song, a stylish twenty-seven-year-old Korean American, was the perfect blend of eBay's two core values, commerce and community.
Song's family background was also a point in her favor—she would be a Korean American joining a Canadian Jew and a French Iranian American, trying to build a global marketplace.
Omidyar was not as intense as Skoll, Song observed, but he was unusual in his own way. He was an “old soul,” Song says, and from the way he talked about AuctionWeb, he seemed to be operating on an almost spiritual plane. “What we're doing is building a place where people can come together,” Omidyar told her. “They just happen to be coming together around trading.”
relationship AuctionWeb had with its customers. For one thing, they were never called customers, always the “community.” And their views seemed to carry an extraordinary amount of weight. When Song started, Omidyar told her to read the Bulletin Board daily and to be responsive to the community's concerns.
But the AuctionWeb community did not respond to her proposal like any focus group she had ever heard of. Song was flooded with irate e-mail. “I got e-mail messages like ‘Are you insane? Do you know what you're doing?'” she says. Some of the criticism was aimed at her color choices—the shade of green she chose was, for some reason, especially unpopular. But most of the complaints concerned process. The community could not believe she would post a fully developed proposal for changing the site, even one involving something as inconsequential as Feedback Forum stars, without first soliciting their input. It was a level of entitlement Song was not prepared for.
Onsale, launched in May 1995, was the brainchild of Jerry Kaplan, a well-known Silicon Valley entrepreneur. Kaplan had previously tried to make his mark with Go Corporation, a start-up that sought to revolutionize computing through handheld, pen-operated computers, much the same dream Omidyar had chased in the first incarnation of eShop. Onsale was Kaplan's first venture since Go, and he was intent on getting it right this time.
In 1996, Onsale hosted more than $30 million in transactions, nearly four times as much as AuctionWeb. But there was a critical difference between the two companies. Onsale took possession of the goods it sold, much of it overstock or remaindered computers. It was a model that was superficially appealing. Onsale could control the consumer experience by examining the goods itself and making sure they were sent out to buyers as promised. This model also allowed Onsale potentially larger profits, since the company earned its revenue from the sale itself, not just by imposing a transaction fee.
Fortunately, AuctionWeb was growing at breakneck speed entirely by word of mouth. In October, the month Song joined, it hosted 28,000 auctions, and it was adding new auctions at the rate of nearly a thousand a day.
Some of history's first recorded auctions were held at the gates of Rome by soldiers selling off war plunder—used, foreign, and one-of-a-kind goods whose worth was subject to debate. In modern times, land-based auctions have been used mainly to sell art, antiques, and other unique items. Beanie Babies, whose value rose and fell daily based on popular whim, could take full advantage of the dynamic pricing mechanism that auctions provided.
percent of the site's total volume. Beanie Babies were the most visible, but hardly the only, collectible flooding AuctionWeb. In fact, in late 1996 all the fastest-growing categories on the site were collectibles of one kind or another—coins, stamps, baseball cards. The same factors that made Beanie Babies ideal for AuctionWeb applied equally to these more traditional collectibles.
Someone in a small town with an interest in Depression glass or southern folk art might have trouble finding like-minded people nearby. But on the Internet, thousands of collectors with the same fascination were only a few mouse clicks away.
Collecting communities arrived at AuctionWeb en masse in its early days. When one member of a Star Trek or an antique jewelry newsgroup discovered the site, word spread quickly. “Barbie collectors went out and told their Barbie-doll friends,” recalls Skoll, who watched happily as AuctionWeb's metrics soared. Sellers who had items listed on AuctionWeb had a particular interest in driving their fellow collectors to the site.
By the end of 1996, collectors were the driving force behind AuctionWeb's growth.
“Spend the money,” he told her, “like it's your own.” AuctionWeb's thriftiness, so rare in the heady days of the dot-com bubble, was largely a reflection of Omidyar's essential abstemiousness—another value he believed made sense in a “crowded world.”
In these early days, AuctionWeb raised thriftiness to an art form. Salaries were uniformly low.
practical—to help users get answers to their questions—or social.
EBay Café posters groused about AuctionWeb topics, like winning bidders who disappeared without paying up, but their chat could just as easily turn to boyfriends and husbands who had proven similarly unreliable. Instead of posting pictures of items for sale, they shared photos of their grandchildren and pets.
When a poster wrote the word “pop,” others would respond with “pop, pop” or “pop, pop, pop”—kicking off the eBay Café's hokey version of a popcorn party. The Café had regularly scheduled rounds of group “singing,” where one poster started off with a lyric and someone else posted the next line. Reflecting the Café's demo graphics, there was, Song recalls, “a lot of seventies music and Shania Twain.” But part of the board's appeal was that users never knew what they would find when they stopped by.
Personalities emerged and friendships formed. There was a breathless poster named Bubbles, whose messages had a flair for the dramatic. When she disappeared from the boards for a few days, which she often did, other posters clamored for her to return. Another poster named Darts could only use the board from work, because she did not own a computer. Members of the nascent eBay Café community chipped in parts of a computer, had them assembled, and delivered the full unit to Darts, so she could participate nights and weekends.
The AuctionWeb staff was also part of the eBay Café community. Omidyar, Skoll, and Song posted regularly. Their eBay IDs were their e-mail addresses, and users had no hesitation about writing to Pierre@eBay.com, Jeff@eBay.com, or MaryLou@eBay.com.
The boards also developed an informal “neighborhood watch.” If someone was being mistreated on AuctionWeb, board regulars often took matters into their own hands. “We used to band together and find the bad guys and make their lives miserable,” recalls Phillips.
What made the message boards so special, early users say, was that in the vast, cold expanse of cyberspace they were a convivial outpost, a place where users never felt alone. That feeling of connection was part of the overall appeal of AuctionWeb:
Omidyar liked to say that community was something that happened spontaneously—that it could not be imposed from the top down. Almost every city in the United States has a community center built by city hall, he used to point out, but most are empty, except for a janitor and someone working the front desk. Down the street, there could be a hole-in-the-wall diner where the easy rapport between the waitresses and the regulars makes the place feel like an extended family.
The community offered by the message boards, and by the site as a whole, was one of AuctionWeb's greatest assets—quite literally. In eBay's first business plan, Skoll wrote that the two key advantages the company had over its competitors were the number of users and the strength of the AuctionWeb community.
But in January, listings soared as sellers returned to the site, making up for lost time. There was another factor AuctionWeb's staff suspected might be behind the heavy January listings: unwanted presents. “A lot of people wanted to unload those fruitcakes they found under their trees this Christmas,” Skoll figured. There was one more reason for the Great eBay Flood: in early January, the company launched its first promotion of a specific auction.
Until the system overhaul was complete, Omidyar and Skoll decided there was only one thing to do: actively drive AuctionWeb's overly enthusiastic users away from the site. The problem was, they refused to be discouraged. “It was like holding back a hurricane,”
The first group AuctionWeb went after was deadbeat sellers, who were abusing Omidyar's trusting policy of granting unconditional credit. In February, AuctionWeb instituted a “credit approval” process that required sellers without strong payment records to provide either a credit card or a ten-dollar prepayment.
Time-based listing limits had an unintended effect. Sellers, desperate to get their items up on the site, sat at their computers with multiple web browsers open, ready to start listing the moment the ten-minute window began. The rest of the time, many of them went to the message boards to pass the time until they could try to list again. “For fifty minutes of every hour, they were talking, bonding, and really getting to know each other,” says Song. Once again, community on eBay was forming in a completely unplanned way.
Notably, since mid-1996, AuctionWeb had been charging a fee for listing items for sale, in addition to the final value fees. Omidyar insists that the purpose was not to bring in more money. Users had demanded the listing fees, he says, in e-mails and message board posts, because there was simply too much junk cluttering the site. “No one wanted to list their fine antiques,” Omidyar says, “next to a pair of dirty socks.” Omidyar decided, once again without benefit of market research or expert opinion, what the fee should be: $1 per listing. The community responded right away. One dollar was too much, they e-mailed him in frightening numbers, for listing low-cost items. Within a week, Omidyar had switched to a graduated scale starting at ten cents, based on the cost of the item.
When AuctionWeb hired Mike Wilson, it also got “Skippy,” the most legendary figure ever to post on the boards. The basic outline of Skippy's life was well known. Always referred to in a genderless third-person, Skippy lived in a cave, had a pet vole named Marta, and trolled the beach looking for dead things to eat. Skippy was a technology genius who always seemed to have inside information about what was going on with the computer system at AuctionWeb.
Skippy became a growing presence on the boards at the same time that Omidyar stopped posting, and quickly became the voice—albeit a slightly twisted voice—of AuctionWeb. When the community asked for new features, Skippy would explain why they could or could not be added. When users asked about outages, or glitches in the system, Skippy gave them the view from headquarters. “The neatest thing about Skippy was that he was very honest,” says Keith Antognini, an early AuctionWeb customer-service manager. “Skippy never tried to pull the wool over the user's eyes.”
Del Vecchio and Schwartz tried to explain that AuctionWeb had a community and a brand, and that they were fast building a prohibitive lead in the Internet auction space. And that not having buildings and trucks was actually an advantage: it was because AuctionWeb was a virtual business that it had gross margins above 80 percent. Most of these arguments were met, Del Vecchio recalls, with blank stares.
Coming forward and meeting that need was what made Pongo such a revered member of the AuctionWeb community. Pongo—the only name she was ever known by on the site—was a message board regular who knew a lot about computers and image hosting, and she was generous about answering questions. At first, Pongo gave advice exclusively by e-mail. But eventually she wrote up tip sheets, which she made freely available to AuctionWeb users, on topics like how to upload pictures, how to resize them, and how to convert images from proprietary formats to standard ones.
They just wanted Pongo to handle their photos, and they were willing to pay her to do it. Arrangements like these were the genesis of Pongo.com, a small image-hosting business Pongo started running out of her home. Pongo began by charging fifty cents a photo, and trusted her customers to keep track of what they owed, paying her when their balances reached the point where it was worth sending a check.
In addition to hosting photos, Pongo.com became an informal AuctionWeb drop-in center.
Note: could we create a way to self fumd the community thruthese kimdsof interactions?
Posting as Pongo, the name of her black-and-white spotted cat, Dee quickly became a fixture on the message boards. Like Uncle Griff and Aunt Patti, she made a name for herself by sharing her expertise.
Like Griffith, Dee was often depressed in the off-line world, while she was shining on AuctionWeb's message boards. “It really gave me a reason to get up and face the world,” she says. AuctionWeb also allowed Pongo to build relationships that would have been difficult for her to sustain off-line. “In my real life I'm an extreme hermit,” Dee says.
But somehow, on the AuctionWeb message boards, Dee felt comfortable expressing, through Pongo, her Jody Roberts persona—highly sociable, hyper-competent, a talented writer, and the willing center of attention. Dee's father was captivated by Pongo and loved going to Pongo.com and AuctionWeb to see the world she had created, and the positive feedback ratings and comments people left for her. His daughter Jody, as he once knew her, was lost to him in the real world. But when he missed her, he could visit her online.
In early 1997, Skoll finally emerged from his office with AuctionWeb's first business plan. As a blueprint for eBay's success, it still holds up well in many respects.
In a section entitled “Philosophy,” Skoll sketched out a vision of economic transformation that read more like an essay on economic history—a Fabian tract, even—than a mere appeal for financing. Before the industrial revolution, Skoll wrote, goods like chairs were custom-made for individual buyers. The craftsman and the customer had a personal relationship. The quality and service were excellent, but the personal attention also meant that goods were expensive, beyond the reach of most people. The industrial revolution ushered in a new era of mass production, in which goods were cheaper. But these efficiencies came at a cost: with large factories and elaborate distribution channels, buyers and sellers no longer formed personal relationships. The computer, Skoll argued, would make it possible to reclaim the old-style relationships. Internet auctions represented nothing less, he wrote, than “the opportunity for mankind to recapture the lost ambiance of the town market, when personal interaction and personal attention was the key to a trade and to life in general.”
Note: how can we find this doc?
EBay, the corporate entity that owned AuctionWeb, was actually engaged in two distinct businesses. Its AuctionWeb auction site was getting all the attention. But Omidyar and Skoll had quietly put together a second business, called SmartMarket Technology,
AuctionWeb's founders did not realize how valuable their fast-growing site was. But Skoll insists that their strategy was a reasonable one, given how small and vulnerable the company was. The focus on software licensing meant that even if bigger players moved into the online auction space, eBay could still survive by selling them software. Like the stores that prospered during the Gold Rush not by digging for gold but by selling equipment to the prospectors, eBay wanted to be, Skoll says, “in a position to provide the picks and shovels.” Now that they had a business plan, Omidyar and Skoll were ready to hunt for venture capital.
Throughout 1996, Omidyar had been leaving Dunlevie monthly voice-mail messages keeping him apprised of eBay's progress. In early 1997, Omidyar called Dunlevie and said eBay was starting to look for VC money, and that he wanted to talk to Benchmark. Dunlevie did not immediately jump at the opportunity. EBay was “this sort of online flea market type of thing,” says Bob Kagle, a Benchmark partner. “I think it was a little challenging to get Bruce's interest.”
It struck him, he says, as “a gathering place for Main Street America.” Kagle, an avid collector of hand-carved fishing decoys, typed “fishing decoys” into the eBay search engine, and was stunned by the vast array that appeared, including one by Bud Stewart, a renowned Michigan decoy carver. “I'd been looking for this stuff at trade shows for probably five years and hadn't come across it,” Kagle says. Kagle bid on the Stewart decoy.
Skoll, never one to leave things to chance, was sending Kagle regular e-mails highlighting every upswing in eBay's metrics. “People say to me now, ‘How did you know that an online flea market would be successful,'” Kagle says. “But the proof was in the pudding—it was already working at the time that we invested.” By June 1997, Benchmark was in. It paid $5 million for 21.5 percent of the company.
“It's unfair to ask whether it added up to four billion dollars of guidance because no one could ever have anticipated just how valuable eBay would become,” he says. “But when I think about whether their contribution was worth one-quarter of the company—a real partnership stake—the answer is absolutely.”
Song developed what she thought was a good list and posted it. The community's response was a repeat of the Feedback Forum stars fiasco. What astonished Song this time was not only the vehemence of the objections, but the level of detail. Button collectors, a group whose existence had escaped her until now, excoriated her for being so ignorant about buttons. “Did you know there are vintage buttons, antique buttons, and modern buttons?” one irate button-seller lectured her.
To help with the transition, eBay opened the eBay Beta board, eBay's first live customer-support board. Instead of throwing out questions to the community on the Q&A Board, users could post their questions on the Beta Board and get answers in real time from Mike Wilson, Jim Griffith, Patti Ruby, or another member of the eBay staff. After the transition, Ruby successfully lobbied for the board to be made permanent.
When Mike Wilson's newly scalable site debuted on September 1, 1997, the name AuctionWeb was retired. Both the auction site and the domain name were now eBay.
Every successful start-up has a Paradise Lost moment, when it casts innocence aside and embraces a more complicated adult identity. For eBay, that moment came in August 1997 when it hired Steve Westly as vice president for marketing and business development. Westly, a Stanford MBA, had the buttoned-down look and careful speaking style of an old-economy executive. His title alone mapped eBay's coming of age.
“They wanted to know why this whole community thing was so important,” he says. “They did not buy into it at all.” In biblical times, the Gileadites and the Ephraimites could tell each other apart by how they pronounced the word shibboleth, the Hebrew word for stream. The Westlyites, it soon became clear, had no end of shibboleths that set them apart from eBay's indigenous tribe. In a T-shirt-and-shorts office, they came to work in jackets, crisp blue shirts, and dark dress pants. While old-timers like Omidyar and Mike Wilson talked endlessly about “community” and “empowerment,” the Westly team packed their conversations with phrases like “cash flow” and “click-through rates.” If the Westlyites had an idea, they didn't pop their heads into Omidyar's office, as Song or Wilson would have. They developed a mathematical model that made their case and put together a PowerPoint presentation. “They definitely had their slides together,” Sandra Gaeta says with a sigh. That Halloween, some of eBay's more bohemian early employees came to work in boring blue shirts and black slacks—and went as Westly, Rock, and Adams. If the arrival of the Westly team was a shock to eBay's ecosystem, it also proved critical to eBay's success. In the fall of 1997, eBay was ready to grow.
Westly, Rock, and Adams were all well versed in the arcana of banner ads and pricing clicks per thousand, and they got right to work. Within days of starting, Rock had hammered out eBay's first major business-development contract, a pact with Netscape for ads that would drive traffic to eBay's site.
In November 1997, as part of its new effort to raise its profile, eBay launched a public-relations offensive that its PR firm was calling, somewhat grandly, the Fall Vision Tour. The idea was to unleash Omidyar on reporters and industry analysts on the East and West coasts to spread the eBay gospel. The fact was that outside of its passionate user base, eBay was still hardly known. There had been few press stories about the company, and since it was still private, no analysts followed it. As he had for his Benchmark appearance, Omidyar dressed up in a suit and tried his best to transform himself into a salesman. Omidyar had prepared himself for tough questions, but on many of his stops there were hardly any questions at all. “People didn't really know what to ask,” says Johnny Wong, a PR agent who went along on the tour. “It was such a totally different animal, they didn't know what to make of it.”
The eBay Vision Tour was at best a limited success. It introduced Omidyar to East Coast opinion makers, and prepared them for the day when eBay would become too big to ignore. But as a vehicle for generating immediate coverage and for driving more traffic to the site, it was a disappointment. Fox Television abruptly canceled its meeting with Omidyar.
when things were slow, someone would run out to the nearby Century Theater to buy tickets, and the whole office would spend the afternoon watching a bad movie.
It took her months of quiet lobbying to get them to produce a list of eBay's top sellers. When she got it, she began calling them personally.
The biggest complaint she heard was a minor one: that, not infrequently, eBay's credit card verification company wrongly rejected their fee payments, and they were temporarily prevented from listing. The
In the end, Stahl came up with a name herself: Powersellers. It suggested, she felt, that the top sellers had been empowered by eBay, but not that they were superior to those who sold less.
formative era for eBay's many baby boomer users. The four letters of the logo overlapped with each other, suggestive of the ties binding the eBay community together. The letters also had what graphic artists call “baseline shifts,” meaning the letters were not on a straight line. That quirky lineup gave the logo an offbeat feel, reflecting that eBay was not Nieman-Marcus or Tiffany's. Then there was the issue of capitalization. The Cleary designers first tried the logo with a capital B, the way eBay was spelled out in print. But as one of the designers says, they found that the B was like an enormous roadblock in the middle of the name. Instead, they capitalized the Y, another way of being quirky and different. Branders like to say that branding is about personality, and the new logo was an attempt to capture eBay's
The libertarian Omidyar, who had founded AuctionWeb to keep the Internet open and accessible to all, was just not the sort to lock up intellectual property rights by patenting key technology. Handler was too late to fight the patent war, but he was in time for another battle, which would ultimately prove more important for eBay's long-term success. It was critical that eBay establish the principle that it was “only a venue,” and therefore not legally responsible for items sold on the site. This was, at the time, far from a settled question of law. The Telecommunications Act of 1996 gave ISPs like AOL safe haven, recognizing as a matter of law that they were like telephone lines, mere conduits for the information they transmitted. But eBay was in a legal netherworld. If it was functioning like newspaper classifieds, as eBay argued, it should not have been responsible for illegal and misrepresented listings. But if it was more like a land-based store, as its critics maintained, it might have had an obligation to investigate and stand behind the items it sold.
Handler's response to the watchmaker was that eBay had no duty to look for infringing items itself, but if the watch company notified it of a specific illegal item being sold, eBay would end the auction. The watchmaker agreed.
It was the start of eBay's “Legal Buddy” program. Eventually, companies like Louis Vuitton, Disney, and Planet Hollywood, and individuals like the rock group Hanson and actress Jennifer Love Hewitt would sign “Buddy Agreements,” and conduct their own searches of the eBay site, looking for infringing items. EBay, for its part, kept its promise to remove illegal items that it was notified about. Microsoft would eventually become one of the most active members of the program, fighting aggressively to prevent the sale of pirated software on the site. EBay worked with law enforcement to track down buyers and sellers who engaged in fraud
Onsale approached eBay to ask about buying the e-mail addresses of its registered users. EBay refused. But after it did, Mike Wilson noticed “bots,” or automated data harvesting devices, crawling across the site and collecting the same users’e-mail addresses eBay had declined to sell. Wilson traced these electronic intruders back to Onsale. EBay sent a letter to Onsale's cofounder and CEO, Jerry Kaplan, asking him to desist. But the bots kept coming.
As commander of the eBay army, Skoll set up a war room in a spare office at eBay headquarters. The walls were covered with army camouflage; mosquito netting dangled from the ceiling. Old bazookas were casually strewn around. The troops who reported for combat—Skoll, Omidyar, Song, Handler, and Wilson—were issued military helmets and dog tags. On a more practical note, Skoll had set up two computers in the room, one to monitor Onsale, the other to track Auction Universe.
In Omidyar's model, eBay was no more than a middleman. It had no reason to favor buyers or sellers; its only economic interest was encouraging listings and completed transactions.
“second-mover” or “fast-follower” advantage. In the case of online auctions, however, the first-mover advantage was real. In part it was because “switching costs” were so high on eBay. Consumers will stay with a given provider, economists explain, when the cost of switching is more than the gains they are likely to achieve from the switch. People rarely switch primary-care doctors, for example, because of the time and effort involved in locating a good doctor,
But with eBay, registered users had a significant investment in remaining on the site. They had reputations, in the form of Feedback Forum ratings reflecting months, even years, of online activity. If they switched to Onsale Exchange or Auction Universe, they would have to leave those ratings behind.
The strength of eBay's online community also made switching costs high.
The other reason eBay's first-mover advantage was so real was that its rise was a product, as Omidyar and Skoll liked to say, of a “virtuous cycle.” Buyers came to eBay because it was where all the sellers were; sellers came because it was where all the buyers were. Once eBay achieved critical mass, which it did early on, it would have made no sense for users to go to any other site.
EBay's virtuous cycle was a specific instance of an Internet phenomenon called “network effects.” An interactive technology has value only if many people use it: when one person has a telephone, it is worthless; the more people who have phone service, the more useful phones become.
Metcalfe's Law, which holds that the utility of a network equals the number of users squared. A network that has twice the number of users as another network is actually four times as valuable. When a traditional e-commerce site adds a new user, it is adding only one new relationship: the site can sell products to that new user. But every new member added to a network like eBay represents not just a single networked relationship, but a relationship with all other eBay users. Those eBay relationships work in both directions: anyone in the network can buy from or sell to the new user.
Note: this is why i want to see alum recruitimg grow
When Westly came back with the AOL deal, Wilson led the charge against it. Wilson saw eBay as a unique cybercommunity—like his former employer, Well Engaged—that had always been allowed to grow organically. But the business-development department wanted eBay to start paying corporate partners enormous amounts of money for new traffic.
“[Bob] Pittman always said, ‘Write a contract, but put it in a drawer,'” Berlow says of his boss, who oversaw AOL's online service at the time.
Irons had been disintermediated. At the height of the dot-com bubble, disintermediation was a favorite buzzword of business consultants and business journalists. The Internet, they predicted confidently, would wipe out middlemen of all kinds. Businesses and individuals who relied for their livelihood on buying undervalued merchandise in one place and selling it for a markup in another would be, as the title of an influential book put it, “blown to bits.” The intellectual forebear of disintermediation was Joseph Schumpeter, the Austrian economist who argued that capitalism is inherently a force for “creative destruction” of existing institutions and distribution patterns. Disintermediation was wildly oversold: dot-coms were projecting that Americans would soon be buying almost everything—groceries, drugs, pet food—online, and that bricks-and-mortar retailers would wither away. But disintermediation was also, to a limited extent, real.
Flea markets and antiques-and-collectibles shows were also starting to be hurt by disintermediation.
EBay was not driving Irons out of business. He was fortunate to have a picturesque antiques store in a heavily touristed area; the experience his customers had buying from him could not be replicated over the Internet.
Founders who will not step aside in favor of more seasoned management are a familiar Silicon Valley story. But Omidyar and Skoll were the rare exception: founders who were clear-eyed about their own limitations. “We were entrepreneurs and that was good up to a certain stage,” says Omidyar. “But we didn't have the experience to take the company to the next level.”
Whitman asked a question that impressed Kagle. “If this is something I decide to do,” she said, “Pierre isn't going anywhere, is he?” Just as Omidyar and Skoll were not the typical founders, intent on leading the company no matter how large it grew, Whitman was not the usual new CEO, eager to push the original management out. “She was very cognizant of the fact that this was a special thing and that Pierre had been the father of it,” says Kagle. There was no real debate—a consensus formed quickly around Whitman.
Whitman spent two hours reviewing eBay's financial records. The 30 percent month-to-month growth rate was astonishing, far greater than anything she had ever seen in the old economy. But what really amazed her was eBay's cost structure: its gross margins were 85 percent. Omidyar, for his part, was struck by Whitman's ability to make sense of eBay's books. When she pointed out eBay's gross margins, he remembers asking himself, “What are gross margins?” Whitman left this time “pretty sold on the idea,” she says, of joining eBay.
begun hiring specialists—Jill Finlayson in Toys and Dolls, and Eric Moriarty in Antiques, Sports, Coins/ Stamps, and Pottery and Glass were two of the first—to organize and actively promote specific categories. Category managers were expected to build relationships with experts, clubs, dealers, and major collectors in their areas, and to encourage them to think of eBay as a place to buy and sell. They were also told to monitor eBay's boards, to track what the community was thinking about their categories, and to follow trends in their areas elsewhere on the Internet and off-line.
Rosaaen helped make the Elvis chat room one of the closest communities on eBay, posters regularly sharing their personal recollections of Elvis. One woman, in a fairly typical post, recalled the time she chased Elvis through a parking lot. Another told about talking the cleaning crew on Elvis's private plane into giving her the trash, only to have her mother throw it out while she was at work. Members of the Elvis chat group met off-line several times.
Omidyar, accepting the road show as a necessary evil, put on a business suit and prepared to make the rounds. But he refused to capitulate completely. He insisted that Goldman, Sachs pick up the cost of the lavish travel—private jets, first-class hotels, and stretch limousines—that was standard road show fare. It was a rebellion against un-eBaysian extravagance, but Omidyar also had a more practical objection to footing the bill.
unheard-of gross profit margins—88 percent, compared to 22 percent for Amazon, 16 percent for CD Now, and 9 percent for Onsale.
While the road show was under way, Omidyar, who could repeat his sixty-second spiel in his sleep, had the least to do. He decided to use his free time to send a series of e-mails back home, to give the eBay staff who could not come along a feel for what it was like. These “Pierregrams,” virtual postcards from the road, were filled with his reflections on the absurdities of the process, and personal glimpses of the management team in action.
“Low-key is far better than arrogance and extravagance,” she says.
With its large, spare rooms filled with simple cubicles, and its vending-machine-stocked kitchen areas, the new headquarters looked, at first glance, like the back office of a midwestern insurance company.
Young had become one of thousands of Americans—the New York Times would later put the number at 75,000—running businesses entirely on eBay. EBay had empowered her in just the way Omidyar had hoped when he created AuctionWeb.
On February 19, 1999, eBay banned guns and ammunition from the site. It was simply too difficult, eBay explained, to ensure that buyers on eBay met the legal requirements for purchasing a gun.
“You've literally put them out of business.” Song, who was hearing from many of these small businessmen, agreed. The gun ban was “the first time in eBay's history that we'd come off the mountaintop and ever said unilaterally how it's going to be,” she says.
EBay responded in a way that stunned the DNF Posse. With a few keystrokes back at headquarters, Pinkhamr and thirty-six other members of the Posse were thrown off eBay. EBay's official reason for “NARUing” the Posse was feedback shilling—giving positive feedback ratings to each other in a collusive way. There was a kernel of truth to the charge.
EBay restored Pinkhamr and thirty-three of his confederates. In explaining the situation to one of the wrongly suspended Posse members, an eBay staff member told him he had been a “dolphin caught in the net.” From that offhand comment, the incident entered eBay lore as “the dolphins in the net.” The incident was a small free-speech milestone that made eBay rethink its message-board policies and give posters more leeway in criticizing the company.
The gun controversy was transformative in another way: it made eBay more respectful of the community's role in bringing change to the site. Management realized it was a mistake to announce the new policy and then say, “Oh, by the way, let's have a discussion on it,” Burke says. But there was one part of the gun-ban blowup eBay never regretted: the decision itself.
“The support boards had always been our direct link to the community.” The live-support boards were fast-paced and untamed—actual eBay staff used them to post responses in real time. In the glory days, Omidyar, Skoll, Song, and Skippy had all shown up
The number of participants grew over time. Every night, as 10:22:22 P.M. approached, DNFers announced that it was almost time for TUZ, and postings on other topics stopped. It was a trick to time a posting just right: the X factor was the speed of the user's Internet connection.
It was also evidence that even after the IPO, the increased emphasis on profits, and the dustups between management and users, the heart of eBay—the community—was still beating.
A refrain emerged at eBay: Yahoo! Auctions had community without commerce, while Amazon Auctions had commerce without community.
Once again, eBay had been saved by Metcalfe's Law. EBay's network had so many buyers, it was difficult to lure sellers away, and so many sellers, it was difficult to lure away buyers. Before the launch, Amazon approached Jeff Fisher,* one of the biggest stamp sellers on eBay, and tried to persuade him to transfer some of his auctions to its site. Fisher's mailbox quickly filled with Post-it notes, pens, and personalized business cards, all gifts from Amazon. When those didn't work, a manager of the Amazon Auction stamps category called him personally and asked what it would take to get him to list on the site. Fisher said he had always wanted a poster of Buffy the Vampire Slayer.
Fisher kept his word and listed some stamps, but he did it in the form of an experiment. He placed fifty identical stamps on each of the two sites. On Amazon, eight of them sold, two for more than the starting price. On eBay, thirty-seven sold, twenty above the starting price. “I got my answer right there,” he says.
Collectors liked to blame eBay, but their complaints were actually just a new version of a classic complaint about auctions—what economists call the “winner's curse.” First described by three petroleum engineers who wrote about oil lease auctions in the Journal of Petroleum Engineering in 1971, it holds that the high bidder in an auction almost always pays too much.
EBay was now a price-setting mechanism—a stock market for everyday goods. Buyers and sellers had actually begun using eBay's thirty-day search function, which let users search final prices in completed auctions going back thirty days, as an authoritative reference source.
But what Lanier found surprised him: the price curve looked more like a two-humped camel. One hump represented a group of items that sold above the ideal price; the other was items that sold below it. The overpriced hump was due to what Lanier called smart sellers and dumb buyers. Smart sellers were the ones who worked the eBay system well, whether within the rules or not. One way smart sellers got high prices was by being creative about selecting the category they listed in.
Dumb buyers, the other force driving prices above the ideal, often simply misjudged an item's worth. Another kind of dumb buyer fell prey to “auction fever.”
The underpriced hump was primarily the result, Lanier found, of dumb sellers. EBay sellers often made mistakes that drove down the prices of their items. In some cases, the errors were as simple as poor spelling.
A Canon Hansa, a pre-World War II Japanese camera that is among the world's rarest, sold on eBay for $6,000, well below its $25,000 value. The seller, unaware of what he had, had listed it simply as “Old Japanese Camera.”
Bowen had expected that returning the T-shirt would end his relationship with eBay. But instead he got a call from Matt Bannick, eBay's vice president for product and community. The two men talked for two hours about the gun ban and what eBay was doing wrong in its relations with the community. When the conversation was over, Bannick decided to invite eight buyers and sellers to visit eBay and talk with management about what was on their minds. Bannick wanted a mix of fans and foes for Voices 1, as the group would later be called. He talked to customer service, and found someone who complained that eBay was not doing enough about fraud. At the other extreme, he chose Dottie Sucara, a Powerseller from Florida, who sold art and jewelry from her husband's homeland of Thailand and who spent a couple of hours a day on the Q&A Board helping other users.
Voices groups weighed in on matters as large as the upcoming alcohol and tobacco ban, and as small as the text of Whitman's “Millennium Letter to the Community,” which Voices 1 members read and edited. The Voices program also allowed eBay to get its own message out to the community. Randy Pinkham, leader of the DNF Posse, was invited to be a member of Voices 3. After visiting eBay headquarters and talking with management, Pinkham said he came to understand eBay's decision to suspend him and the rest of the DNF Posse. “If I had been in their position,” he said later, “I could see myself doing the same thing.”
EBay invited several of its remote employees—including Uncle Griff, Aunt Patti, and Sonny Wagner—to move to Draper and become supervisors. The remotes appreciated the chance to join the eBay fold, but they were not excited about the location.
Whitman had a simple question, which she repeated constantly throughout the Friday morning staff meeting:“How are we going to make it up to them?” It was not rhetorical. She and Omidyar were about to write an open letter of apology to the community, and she wanted ideas. Whitman decided on the spot that eBay would automatically refund the fees for any items listed during the outage—even if the auction had been completed, and even if the item had sold for a good price. Bengier told her the offer would cost about $4 million, a significant amount of money to eBay at the time. But Whitman decided that not doing it would ultimately cost the company more.
there was no excuse for not having spent more money on technology. Failing to do so, Whitman would later concede, was perhaps the biggest mistake of her eBay career—and one that could have led to the company's demise. The June outage was a rite of passage for eBay. It revealed just how inadequate and disorganized eBay's technology operations were.
Whitman installed Webb in a cubicle adjoining her own. Webb had never worked in such physical proximity to a CEO before, but then he had never had a job where he had been as mission-critical as he would be at eBay. “Maynard in many ways saved the company,” Whitman would say later.
Lucking-Reiley auctioned off fifty pairs of identical Pokémon cards on eBay. He put one card from each of the pairs up for auction with a secret reserve price of 30 percent of its value and no minimum bid. He put the other card up with a minimum bid of 30 percent of its value and no reserve. When the auctions ended, 72 percent of the cards with minimum bids ended up selling, but only 52 percent of the cards with reserves did.
What the conventional wisdom did not recognize was that buyers were so put off by secret reserve prices that many simply refused to participate in auctions that used them.
If the Feedback Forum did what eBay claimed, buyers should be paying more for mint Indian head pennies from sellers with high feedback ratings than from sellers with low ratings.
The U.K. site also had a significant legal obstacle eBay.com did not. England had no First Amendment, and its defamation laws were far stricter than in the United States. EBay-U.K. found that when its users received negative feedback, they often called up and threatened to sue both the writer of the feedback and eBay.