One Click: Jeff Bezos and the Rise of Amazon.com Part 5

Chapter 13.

Is Amazon Killing the Bookstore?

Amazon.com is not going to put bookstores out of business. Barnes & n.o.ble is opening a new superstore every four days. Borders is opening a new superstore every nine days.

-Jeff Bezos in 1998.

Barnes & n.o.ble is no longer opening a couple of new stores nearly every week. It has recently spent its time closing them. In early 2010 the world's biggest bookstore chain shut down the last of its B. Dalton bookstores, which it acquired in 1986 and once boasted eight hundred stores. Barnes & n.o.ble racked up $107 million in losses in the first nine months of its fiscal year in 2010.

Borders Group, Inc., the second largest bookstore chain, is suffering like a CEO with swine flu. It has lost a total of $605 million over the last four years. The company announced on December 30, 2010, that it would delay payments to some publishers while it tries to refinance its debt.

In order to remain compet.i.tive, the second-largest shareholder of Borders stock, William Ackman of Pers.h.i.+ng Square Capital Management, has proposed an unusual tactic: buying Barnes & n.o.ble, a company more than twice its size. In December 2010, Ackman proposed a price of nearly $960 million, and said he was willing to help finance the deal. The hope was that combining two struggling companies would create the economies of scale to remain compet.i.tive by cutting costs. A buyout seemed like a long shot to a.n.a.lysts when it was announced, but it shows the scope of the tactics the big chains are willing to consider in order to stay alive.

Compet.i.tors and publishers think Bezos is ruthless. He's considered to be a brutal tactician by others in the book publis.h.i.+ng industry-book publishers, retailers, and authors, for example-who feel their livelihood is being jeopardized by his cut-rate pricing and arm-twisting to get books at low cost. He seems willing to practically destroy the book publis.h.i.+ng industry that got him started if it keeps him ahead.

Is Amazon destroying bookstores? If the big chains are just ailing from a bad economy, Amazon is immune from the recessionary virus. While the chains shrink, Amazon grows. In 2010, Amazon's fourth-quarter revenues grew 36 percent to $12.95 billion, while profits were up 8 percent to $416 million. In that same time period, Barnes & n.o.ble's revenues increased just 7 percent, to $2.3 billion, while its profit fell 25 percent to $61 million. Borders reported a decline in revenues of nearly 18 percent, to $471 million for its third quarter in 2010 (which ended in November) and a loss of $74 million, nearly double its loss from the year earlier. Borders filed for bankruptcy in February 2011.

Of course, Amazon is selling much more than books these days. But it's still the biggest bookstore on earth. For the third quarter of 2010, it broke out its sales into electronics or media, the latter being primarily books and music. Since Borders and Barnes & n.o.ble also sell music, it's fair to compare their results to Amazon's media revenues in the third quarter. Amazon's media revenues were up 14 percent to $3.35 billion, while Barnes & n.o.ble's revenues dropped 18 percent to $1.9 billion.

The problem the bookstores face is a viciously compet.i.tive market. Once it was the big box retailers themselves causing trouble for the independent bookstores. Other retailers-such as Wal-Mart-jumping into the book business have also taken their toll. But everyone knows who's most responsible for a tough bookselling business now. "Today," says Oren Teicher, CEO of the American Booksellers a.s.sociation, "the compet.i.tion is clearly coming from Amazon."

And what about investors who decided Amazon would be killed by the chains? Between Amazon's initial stock offering in 1996 and the end of 2010, Barnes & n.o.ble's stock has dropped 29 percent (undulating in between like waves on a Waikiki beach), while Border's stock has fallen 96 percent. Amazon's stock has risen 10,320 percent. Based on stock prices at the end of 2010, Barnes & n.o.ble was worth $852 million, Borders was worth $65 million, and Amazon was worth $81 billion. That's some very pricey toast.

The main tactic Bezos employs is quite simple: low prices. It's hard to argue with that tactic. Who doesn't want lower prices? It certainly fits in with Bezos's promise to put his customers ahead of others.

But compet.i.tors are not happy with Bezos's tactics. "Amazon is a compet.i.tor that plays by a whole different set of rules," says the American Booksellers a.s.sociation's Teicher. Bezos, he says, "does not really care about books. He uses books as loss leaders to sell everything else. He's acquiring customers in order to sell whatever he can ultimately sell them. He has become the expert at marketing [other products] to customers once they've got them on the site."

Furthermore, compet.i.tors claim, cheap prices please book lovers in the short term, but may hurt them in the long run. Discounting makes it more difficult for retailers, publishers, and, yes, book authors to make a profit. The more that publishers' profit margins are squeezed, the more focused they become on authors who can make the best-seller lists. The same goes for bookstores that need to move a lot of books to stay in business. "Clearly the deep, deep discounting of both physical and digital books is bad for consumers, publishers, retailers. You end up with fewer choices. The evidence is clear that if you keep discounting the product, there will be less product," says Teicher.

Bezos argues that low prices, combined with new technology, simply broaden the market, lower costs, and enable people to buy more books.

Well, yes and no. Data from Bowker, a company that tracks worldwide book sales from the Books in Print database, indicates that the number of book t.i.tles published continues to increase each year. In fact, the number of t.i.tles published took a big jump in 2009 (the last year for which it has statistics), increasing 87 percent from 2008, to 1.053 billion. However, a big part of that jump was due to print-on-demand (POD) books, produced by specialty printers that focus on public domain t.i.tles, self-publishers, and micro-niche publications. These t.i.tles jumped 181 percent over 2008. The problem is that actual sales of these t.i.tles are very low compared to books from the major publis.h.i.+ng houses.

And retailers say they have not seen the number of readers increase with lower prices. "The big box stores were supposed to increase the numbers of readers too," says Richard Howorth, the owner of Square Books in Oxford, Mississippi, who was one of the teachers of the cla.s.s on running a bookstore that Bezos took. "But they didn't."

The other complaint is the tactics Amazon uses to keep its prices low. It has enormous influence now, and publishers complain that Amazon puts huge pressure on them to lower wholesale prices, squeezing profit margins. "Amazon dictates terms to publishers," says Howorth.

That means other bookstores, who cannot pressure publishers into giving them similar price cuts, find it hard to compete. A 2003 study by the American Booksellers a.s.sociation shows that the average gross margins for physical bookstores was 30.8 percent. After payroll, facilities, and other costs were factored in, the result was a net loss of 1.7 percent. For 2009, Howorth's gross sales were up just 2 percent from the previous year.

And then there's the tax issue. Bookstores have to charge sales tax on books. But in 1992, the Supreme Court ruled that online retailers only had to charge sales tax on goods sold in states where they have a physical presence. To Bezos, that means charging tax to buyers in the state of Was.h.i.+ngton. The lack of taxes in most states means that Amazon's customers get a further price break. That gives Borders and Barnes & n.o.ble a distinct disadvantage for their own online sales, since they have stores in most states. Bookstore owners-and some states-say Amazon should charge tax in any state where it has a distribution center or any other kind of physical presence.

That has led to legal challenges from states and retaliation from Bezos. Texas is a big battleground. In October 2010, the Texas state comptroller sent Amazon a bill for $269 million, the amount the state feels Amazon owes in back taxes, penalties, and interest, because it has a distribution center in Irving. Bezos responded by saying he will shut down the site rather than pay the taxes. This is an unsettled issue that could still cause problems for Amazon in the future, although it will still have an advantage if it limits distribution centers to just a few states.

Retailers are also worried about the pressure Bezos puts on publishers. "The scary thing is that publishers are having such a difficult time," says Howorth. Big retail chains have always put pressure on publishers to discount their books, but Bezos has a very strong arm when it comes to exerting his influence. He has used tactics ranging from temporarily removing a publisher's books from the site to the subtler technique of simply removing the "1-Click" or "Add to shopping cart" icons from their books' listings if they don't give the discounts he requires.

The Authors Guild, an advocacy group for writers, has kept tabs on some of these tactics with a Web site called Who Moved My b.u.t.ton? According to the site, the guild decided to track the events because "publishers are often too fearful of antagonizing Amazon to say anything publicly about such incidents. Authors needn't have that fear (we can keep your ident.i.ty anonymous), however, and publicity is probably our best collective defense."

From what is known, Amazon's tactic seemed to start in 2008, when Amazon disabled the "Buy" icon from its British Web site for books published by Bloomsbury, and later for books from the British unit of Hachette Livre. The move meant that British buyers could only get the books through Amazon's third-party sellers, not from Amazon directly, and not with Amazon's discounted s.h.i.+pping options. "The buy b.u.t.ton is their weapon of choice and that's how they impose market discipline," said Paul Aiken, executive director of the Authors Guild. News of Amazon's disabling of the "Buy" icon went public when Hachette's CEO, Tim Hely Hutchinson, wrote a letter to his authors complaining of the "oddities" of the missing "Buy" icons after he resisted Amazon's demands for lower prices. "Amazon seems each year to go from one publisher to another, making increasing demands in order to achieve richer terms at our expense and sometimes at yours," Mr. Hely Hutchinson said in the letter. "If this continued, it would not be long before Amazon got virtually all of the revenue that is presently shared between author, publisher, retailer, printer and other parties."

Also in 2008, Amazon hit the fastest growing category of new books, print-on-demand (POD), with new restrictions. The self-published authors can sell their books through Amazon, but must print their books through the Amazon subsidiary Creates.p.a.ce or those books would lose their "Buy" icons. Bezos promoted it as a cost-saving approach for customers. An open statement on Amazon. com read: "It makes more sense to produce the books on site, saving transportation costs and transportation fuel, and significantly speeding the s.h.i.+pment to our customers and Amazon Prime members. We believe our customer-focused approach helps the entire industry in the long term by selling more books."

But one POD publisher, BookLocker.com, felt it was illegal for Amazon to demand a monopoly in printing POD books. (The majority of such books are sold online, giving Amazon enormous leverage.) BookLocker filed an ant.i.trust suit against Amazon. It was settled in December 2009, with Amazon agreeing not to retaliate against BookLocker for using another publisher, and agreeing to pay BookLocker's legal fees of $300,000. The settlement, however, only applied to BookLocker. Many other POD publishers had already agreed to Amazon's demands.

Perhaps the most surprising trend is that Amazon has not put the small, independent bookstores out of business as everyone once predicted. They were first battered by the big chains and other retailers selling discount books, and the rise of Amazon was supposed to swamp them like little boats. .h.i.t with the third leg of a perfect storm. But ABA's Teicher says that the small bookstores seem to be learning to weather that storm, although it looked tough for many years. "Between 1994 and 2005, no new stores opened," he says. "There's always churn, but the closing stores weren't being replaced." But things leveled off in 2009. "For the first time in more than 15 years the number of members in the ABA has stabilized," he says. "New stores have opened, and the stores that survived the onslaught are more compet.i.tive."

Howorth echoes that view. "What you don't know about Amazon is how much of your business they're taking," he concedes. But, he adds, "I think he's hurting the chains rather than the independents."

It also seems to fit the prediction that Bezos made in 1998, when he promised that Amazon would not destroy the local bookstore. "I still buy half of my books at bookstores," he said. "Sometimes I want the book right now, not tomorrow. Sometimes I just like to get out of the office and go to a nice environment. What you're going to see-and it's happening already-is that physical bookstores will become ever-nicer places to be. They are going to have more sofas, better lattes, nicer people working there. Good bookstores are the community centers of the late 20th century. That's the basis on which they're going to compete. There is plenty of room for everyone."

Well, saying there is "plenty" of room may be a bit of an overstatement. But if the numbers collected by ABA hold up, it may be possible to stay in business. They just have to work at it harder than they ever have before.

Chapter 14.

A Cool Guy with a Funny Laugh.

The thing about inventing is you have to be both stubborn and flexible, more or less simultaneously. Of course, the hard part is figuring out when to be which!

-Jeff Bezos.

Customers loved Amazon.com from the first day they started using it. Industry pundits, who like to demonstrate their expertise on executives and companies, insisted it would never last. In July 1997, Forrester Research President George F. Colony, a respected technology a.n.a.lyst and commentator, said, "Amazon's position is indefensible. They have some nice custom-built software, but that's about all they've got. They don't have a monopoly over the books, and their technology can be duplicated in six months. Forrester Research labeled the company Amazon. toast, while Barron's magazine called it Amazon.bomb. Bezos was just another overhyped dot-com rebel without a clue.

It turned out that Amazon's position was highly defensible. Colony was absolutely correct about what Amazon did not have. But he didn't understand the weapon Amazon did have: Jeff Bezos himself. It's true that, in theory, anybody could have copied Amazon's strategy and reproduced its software. Several executives tried. But imitations of the original are never quite as good, and in the online retailing business, Bezos is a true original.

By now, n.o.body can dispute that Bezos is a great entrepreneur. His original vision of how the Internet could provide a unique service to customers, as opposed to being simply an online book ordering system, kept Amazon ahead of the compet.i.tion. Only an entrepreneur with the right vision could see the need for all the features necessary to keep the company ahead, whether he dreamed them up, or his team thought of them first, or he borrowed good ideas from other companies.

Unlike many of the dot-com executives who came later, creating a great company was more important to him than becoming wealthy. Even after the company went public and Bezos was suddenly worth half a billion dollars, he made it a point to let people know he lived in a small apartment in Seattle and drove a Honda (although he later moved into a more impressive house on the border of Lake Was.h.i.+ngton near the mansion built by Bill Gates and bought three adjacent apartments in Manhattan's art deco Century Building from Sony Music executive Tommy Mottola for $7.7 million).

Perhaps it's his goofy laugh and silly grin that made people underestimate him; certainly his childlike playfulness contributed to that perception. At their wedding reception, Jeff and MacKenzie provided an outdoor adult play area that included water balloons. Even that playfulness served the secondary purpose of helping to draw attention to his company. In 2003, he played tennis with Anna Kournikova at Grand Central Terminal's Vanderbilt Hall in order to publicize the fact that the Anna Kournikova sports bra (designed specially for the tennis star) was making its debut in Amazon's Apparel & Accessories store.

But mostly the skepticism came about because outsiders believed that the much more seasoned executives at the giant bookstore chains were going to outwit him and his "Who needs profits?" strategy. In reality, people just didn't understand him.

That's his own fault. Bezos seems to thrive in the limelight, but he has always carefully controlled his public image. That includes elaborate publicity stunts. In June 1999, in order to celebrate Amazon's ten-millionth customer, Bezos personally delivered a set of golf clubs to a construction worker in Boston while he was at the construction site, with the press in tow. The customer picked, however, was only somewhere around the ten million mark. Bezos chose him because he had to be in Boston anyway at that time. And the clubs he handed to the guy weren't even the ones he ordered. They were just there for the photo opportunity. The customer had to give them back and wait for the real clubs to arrive by mail. These days, Bezos virtually never gives interviews to the press unless he has a specific agenda to promote on the talk show circuit (such as hawking the latest sports bra or Kindle). A CEO who controls his image so carefully may rightly be considered disingenuous, and Bezos is no exception. But he's also farsighted and clever, and manages to get reporters starved for interviews to hang on his every stunt-studded word.

In fact, Bezos personifies a new breed of executive that arose with the emergence of the game-changing technology companies in the 1980s and 1990s. In a January/February 2000 Harvard Business Review article by anthropologist and psychoa.n.a.lyst Michael Maccoby, t.i.tled "Narcissistic Leaders: The Incredible Pros, the Inevitable Cons," Bezos was one of the business leaders singled out as a "productive narcissist." (Bill Gates and Larry Ellison also made the list.) These executives have big enough egos to make up seemingly random rules of business leaders.h.i.+p. However, unlike other narcissists, they get the job done.

Bezos also has a hugely infectious enthusiasm for his company. One of his incredible talents has been to convince employees, from the highest manager to the lowest customer service rep stuck to her phone ten hours a day, that working at Amazon was not just a job-it was part of a visionary quest, something to give higher meaning to their lives.

To Andreas Weigend, a computer science professor whom Bezos hired in 2002, Bezos is a rock star, partly because of his highly contagious enthusiasm. Weigend describes Bezos as "a cool guy with a funny laugh," and was most impressed with his upbeat and inspiring att.i.tude. "I'm a very positive guy," says Weigend. "And I left every meeting with Jeff more happy, more energetic than when I went in. I had never met anybody who is consistently so empowering."

Bezos is also one of those rare executives who can understand the nuances of technology as well as the big picture of how to succeed. Weigend is a computer scientist, an expert in data mining. He loved the fact that Bezos could understand the technology, and if he didn't, could pick it up easily. Weigend recalls showing Bezos a power-log graph of customers' buying habits on Amazon. It's the kind of graph that turns a trend line showing exponential growth into a straight line for easier a.n.a.lysis. Weigend explained the nuances of the graph to Bezos one day, and a week later Bezos was explaining the graph in a meeting with authority. "After one week, he was talking about it as if he knew it from kindergarten," says Weigend. "That's the wonderful thing about him. He gets it. There's a reason why he's the only CEO from a dot-com company who's still here." That may be a bit of an exaggeration (especially if one considers Google as one of the dot-com companies started in 1999)-but not by much.

Certainly not everybody bought into the Bezos hyperenthusi-asm. Former Amazon customer care rep Richard Howard's 1998 article "How I 'escaped' from Amazon.cult" described a cultlike environment in which employees wors.h.i.+pped Bezos as a visionary. They kept talking about Amazon's goal, not to be a gigantic online retailer, but to "change the world." Since Howard had to provide three letters of reference, two writing samples, SAT scores, and college transcripts just in order to get that $10 per hour entry-level job, he a.s.sumed there would be ample opportunities for advancement. It turned out to be a job that lasted just a few weeks.

Besides, there was so much enthusiasm from almost everyone around Howard. He noted that he was surrounded by employees who saw Amazon as a "life-changing experience." But Howard never caught the "Amazon Is Great" disease. So many people kept trying to convince him that Amazon was the greatest place on earth, he started wondering if he was, for some reason, being singled out for conversion to the Amazon cult, until he talked to another beginner at the company. Her response: "I've gotten a couple of those too, only my friend warned me in advance that working here was sort of like being in The Stepford Wives, so I've more or less taken it in stride."

An article in The Was.h.i.+ngton Post quoted another customer service employee who described the environment as a socialist collective. "It's like Communist China under Mao," the rep said. "You're constantly being pushed to help the collective. If you fail to do this, you're going against your family. But if this is a family, then it belongs on Jerry Springer."

In a 1999 article in The New York Times, Peter de Jonge explored some of the quirks at Amazon, visiting junior employees and senior managers alike. He also noted the cultlike atmosphere that Bezos inspires-or, at least, projects-about his company. Every manager at the company refused to discuss the company's soaring stock, as though it were an offensive question compared to the company's great vision, and yet all of them seemed to know the current price. "Bezos never serves up e-commerce as naked capitalism," noted de Jonge. "It's 'helping people find and discover the things they want,' 'helping folks make better purchase decisions' and so on and so on. When it comes to seducing his employees, he offers not just a low paying job with a handful of stock options, but also a life's work."

Even in the December 1999 Time magazine article that named him "Person of the Year," Bezos was described as "pathologically happy and infectiously enthusiastic." The writer, Joshua Quittner, noted that the company was sprinkled with banners listing Amazon's Six Core Values: "customer obsession, owners.h.i.+p, bias for action, frugality, high hiring bar and innovation"-and dubbed the company culture as "the Cultural Revolution meets Sam Walton," or, more succinctly, "dotcommunism!"

Clearly, the Amazon cult is not for everyone, and those employees who didn't buy it (such as Richard Howard) didn't last long at the company. But many others did, noting that Bezos's enthusiasm was genuine-and got results. Peri Hartman says that it kept people going when things were difficult. "He was very positive-optimistic is probably a better word. He would say that people will tell you all the time that something can't be done, and if you listen to them, you will fail. He said we will make it work. He saw through all the negativism. It made him fun to deal with."

In Silicon Valley, this cultlike belief in the company is dubbed "the vision thing," and the best CEOs have it (think Steve Jobs). Bezos definitely has it, even if he has been sometimes difficult to deal with on a personal basis. And some of Amazon's most important employees have found that, like Steve Jobs, he can sometimes be difficult indeed.

Chapter 15.

But What Kind of Manager Is He?

Different executives who have worked with Bezos in the past have strongly different views of his strengths and weaknesses as a manager. He's not always a "nice" CEO. He can inspire and cajole, but can also irritate and berate. He can see the big picture, and micromanage to distraction. He's quirky, brilliant, and demanding. Some of his former employees adore him. Some have found him to have serious shortcomings. All seem to feel that he's a great visionary who knows how to build a lasting company.

"There is no doubt that Jeff is a brilliant businessman with a strong vision for where he wants to take his company," wrote Shel Kaphan, the company's first employee, in an email to me. "However, he is also a demanding micro-manager who is extremely hard to work for. This is one reason there are very few people from the early days of Amazon still at the company. He is also very temperamental, and has (or at least had, when I was there) a bad habit of severely chewing out his subordinates in front of others." Another former executive, for example, said Bezos had an annoying habit of waving his hand in front of subordinates' noses when he didn't want to hear any more of their conversations.

That's part of the hyperactive side of Bezos's personality, one that he sometimes displayed in his previous jobs as well. Although Graciela Chichilnisky, his first employer, considered him an excellent manager, he has always had an enormous amount of energy that might intrude on most a.s.sociates' comfort zone. "He jumped out at you, his eyes practically jumped out of his sockets," recalls Chichilnisky.

He has always been an unorthodox manager. One former executive recalled that, at an offsite retreat where other managers said the company employees should start communicating more, Bezos stood up and declared, "No, communication is terrible!" He wanted a decentralized, even disorganized company where people could come up with independent ideas rather than subscribe to groupthink. He ruled the company with the "two-pizza team" concept, that dictated any team should be small enough to feed with two pizzas.

Empathy is not something that comes to him naturally. When he was ten years old, on a trip with his grandparents, he decided to try and get his grandmother to quit smoking. For that he relied more on his geekiness than on sensitivity to a sore subject. He calculated that the amount she smoked would reduce her lifespan by nine years. It made her cry. His grandfather had to teach him to be more sympathetic. "My grandfather looked at me, and after a bit of silence he gently and calmly said, 'Jeff, one day you'll understand that it's harder to be kind than clever,'" Bezos said.

He has no empathy for employees who complain about working long hours in pursuit of his quest. Bezos often pushed his people with the finesse of a galley slave driver. One customer service manager recalled that, when the staff got a week and a half behind answering emails despite putting in twelve-hour days, seven days a week, Bezos called her to complain about the lapse. When she told him they couldn't work any harder, he came up with a solution. They dedicated one weekend competing with each other to see who could get through the most emails in the backlogs. During that forty-eight-hour period, everyone worked at least ten hours past their regular s.h.i.+fts. Each person was given a cash bonus of $200 for every thousand messages he or she could answer. It cleared off the backlog.

At other times, Bezos also had a rather geeky way of showing his appreciation for his programmers' hard work, which was often done on their own time. Rather than a raise, he started rewarding exceptional accomplishments with awards named after the Nike slogan "Just Do It." The prize? An old, used Nike shoe. Greg Linden, for example, a programmer who joined Amazon in February 1997, found a better way to pa.r.s.e out similar tastes in books in order to make recommendations to customers. Amazon would recommend books to a customer that were bought by other customers with similar buying patterns. On his blog about his days at Amazon, he recalls that when his program went live on the site, "Jeff Bezos walked into my office and literally bowed before me. On his knees, he chanted, 'I am not worthy, I am not worthy.'" Linden won "Just Do It" awards for this and other accomplishments, and long after losing the shoes, he wrote on his blog, "What was not lost was the sense of pride. I was proud to have gotten that c.r.a.ppy old shoe."

Some former employees also say that Bezos's reputation as a ruthless business executive is wrong. In an email correspondence with me, Linden describes Bezos as a "geeky, ambitious tinkerer with a focus on doing right for customers (and people in general)." Linden adds that the public perception of Bezos as a hard-charging, take-no-prisoners CEO are incorrect. "I'd disagree with the characterizations of him as compet.i.tive (which I think was just misinterpretation of his ambition) or secretive (which I think is more about wanting to protect his team and his customers). Jeff could much more accurately be described as a naively optimistic geek than a calculating megalomaniac."

Bezos also has a lighter side, which, if sometimes goofy, helped to reduce stress. In the very early days, he had employees pick out the twenty strangest t.i.tles sold every week, and awarded a prize for the strangest. Some of the winners: Training Goldfish Using Dolphin Training Techniques, How to Start your Own Country, and Life Without Friends. Those are not the kind of t.i.tles most bookstores carry regularly. Former programmer Peri Hartman appreciated the silly but upbeat att.i.tude that Bezos embodied like a kid winning a goldfish at a street carnival. Although stressful at times, "Amazon was a fun place to be," says Hartman. "It was stressful in a good way. Jeff has a very positive att.i.tude toward employees."

Hartman also praises Bezos for fostering a sense of camaraderie, noting that there was no infighting between teams vying for attention at Amazon, something that can destroy a young company. The goals, says Hartman, were "innovation, working hard, doing the best job you can. Everybody pulled together to make this happen. You were competing with outside companies, not internally."

Bezos also has the technical chops that allow him to recognize what features could make or break a company. He understands the technology, the problems, and the solutions, and can offer an understanding ear and knowledgeable guidance. But, according to Kaphan, that kind of help didn't always come frequently enough. In the early days, when the company was understaffed and overworked, that meant the programmers had to wing it and hope for the best. "Generally there was enough to worry about that we had to just solve our design problems however we could and move on to the next challenge," says Kaphan. "If anything, there was not enough in the way of design review [from Bezos], which caused some serious problems in the engineering culture of the company, and led to a lot of problems later."

Hartman agrees with both these views. "Jeff didn't do code," he says. "He was the idea guy. He came up with ideas all the time. He probably thought things through more than others." The fact that he focused on the big picture rather than the actual code was still a huge benefit to Amazon, something that Hartman believes "let the company grow into other [new] ideas."

An idea guy can also seem a bit like a king overseeing his kingdom, or perhaps a CEO of a much larger company who can leave the heaviest lifting to his subordinates. In the early days, Bezos did chip in with a lot of mundane tasks, such as packing up books and getting them into the mail for customers.

But he didn't always put in the long hours required of everyone else. Working overtime is not an option for employees at a start-up, it's a mandate. The programmers had to pull many all-nighters in order to meet deadlines. But, frustrating for some, Bezos did not join the late-night work sessions. One former employee says he would "make a show of always getting his eight hours of sleep."

Still, he provided the right kind of guidance from the top, the kind that a CEO absolutely must give. He demanded a robust computer infrastructure, a database management system that could handle orders as the company grew, an ordering system that was easy to use, and the back-end tools that would ensure products could get to customers quickly. Kaphan notes that Bezos paid a lot of attention to the flow of the checkout process and the warehouse order processing software. And everything had to be stable enough, able to handle enough traffic that it would not crash and leave customers stranded, a common problem, especially in the early days of commercializing the Internet. "He was scared to death that we would get all these customers, and then they would go away because the system didn't work well, wasn't easy," says Hartman.

His reputation of being notoriously cheap was very important to him. Not only in hiring and renting office s.p.a.ce, but even for the process of furnis.h.i.+ng his headquarters. For many years he made a big deal out of the fact that the desks at Amazon were made from doors with legs attached. Other furniture was bought at garage sales and auctions.

In fact, the reputation was sometimes more important to him than the actuality. It's part of Bezos's carefully crafted public image. He made it a point to publicize his early frugality for years, sure to talk about his door-desks in interviews. His personal desk was even featured in a photo shoot in Vanity Fair.

The way Jeff put it, the doors represented the company's wisdom in serving the customers rather than the employees. "We have a strong focus on trying to spend money on things that matter to the company and not spend money on us," he said in an interview with a Silicon Valley magazine. "What our furniture looks like does not matter one whit to our customers."

But appearance was even more important than reality. "Sometimes, that meant spending a little bit more to reinforce the idea that we weren't wasting money," said Gina Myers, Amazon.com's first comptroller. "Jeff would say if it looks cheap-even though it's a little bit more expensive-we should buy it because it reinforces our culture of being cheap and not wasting money." Since few customers ever saw the company's furnis.h.i.+ngs, that perception must have been intended for potential business partners, investors, and the press.

Yes, Jeff Bezos is a showman. Some of his techniques may represent a shoot-from-the-holster approach and, as is always the case with successful start-ups, there was a certain amount of luck involved. As Kaphan noted in his email to me, "Amazon's rapid growth covered over a large number of sins. Given the magnitude of some of the mistakes, I think the company's survival is at least in some significant part due to luck."

But in the end, Bezos's aim was spectacular. Kaphan adds that Bezos fostered a "bias for action.... We tried a lot of things and made a lot of mistakes, but managed to avoid any fatal ones." Luck aside, "It also took a lot of intelligence and skill," says Kaphan.

What really made the difference for the company was Bezos's dedication to iteration. He worked at things over and over again until he got it right, at least most of the time. When certain things looked like a lost cause (such as trying to create a search engine to compete with Google), he was willing to drop them. The Kindle, which he considers to be a crucial product for the future, took three years to develop and a couple more to really catch on. Large companies, with businesses to protect and stockholders to please, often have a hard time innovating. "One of the biggest problems with big companies doing clean sheet innovation is that even if you see it, you have to be a really long term thinker, because for a long time it will be a tiny slice of the company," he has said. "The key thing is to be willing to wait 5, 7, 10 years. And most companies aren't willing to wait ten years."

In the end, success makes up for a lot of the pain and frustration that employees might have had to put up with. With that success came an enormously successful IPO, and extraordinary rewards. Bezos showed his appreciation for the work that his early employees did in more ways than one. After the IPO, in order to celebrate Shel Kaphan's fourth anniversary at Amazon.com, Bezos organized a "Shelebration." He chartered a jet to take Kaphan, the engineering staff, and their spouses to Maui for a four-day surprise vacation. When they arrived, Kaphan was greeted by a group of old friends from the Bay Area waiting for him, flown there on a second jet Bezos had chartered just for them.

And that's not even counting the monetary rewards Kaphan received for finally realizing his dream of helping to create a spectacularly successful start-up. Kaphan's hard work netted him over one million shares of stock. When Amazon went public on May 15, 1997, Kaphan suddenly found himself worth over $25 million. At today's prices, depending on how much stock Kaphan has sold over the years, he could be worth over $170 million. Today, Bezos is one of the richest executives in the world. He made number eighteen on Forbes magazine's list of the world's wealthiest people in 2010, with a net worth of $12.6 billion.

A hundred million dollars, old mismatched shoes, praise from a prostrate CEO . . . the rewards of working for Jeff Bezos are memorable, often spectacular. It's all part of the unusual world of Bezos. And those were just the early days. Bezos was to prove to be full of surprises in the years to come. The one thing he should never be, however, is underestimated.

Chapter 16.

Head in the Clouds.

When Netflix streams movies on demand to your home, the programs are sent to you from computers on Amazon. Netflix can't afford (at least not yet) to buy all the computing power needed to load up films instantly and stream them to thousands of customers at any moment. So it rents computers from Amazon's vast store at pennies per minute to handle the tasks, tapping into just as much computer power as it needs at any given moment. It's all part of a surprising business from the online retailing company, called Amazon Web Services, which is part of a larger trend known as cloud computing. Services like this bring in half a billion dollars annually in revenues to Amazon.

Buying companies is a relatively easy way for a stock-rich company to expand its business. But sometimes a great executive will stumble upon an unexpected new idea, or one of his employees may come up with something. The key is the ability to look beyond the current conventional wisdom and embrace a radical new idea. Jeff Bezos has that ability. He doesn't create any structured "skunk-works" organization specifically tasked with the job of creating new businesses, but engineers within the company are given the opportunity to experiment, and good ideas are embraced quickly.

That happened around the turn of the century, even as Amazon seemed to be sinking into the dot-com abyss. Internal tinkering by Amazon engineers led to a new business opportunity that put Amazon at the forefront of cloud computing. Around 2000, some Amazon engineers got the idea of enhancing the a.s.sociates and Marketplace programs, which created retail relations.h.i.+ps with outside companies and Web sites. What if partners who referred buyers to Amazon also had more access to the rich data and retail software that Amazon had developed? Would that make it easier for other sites to help Amazon sell its wares?

Robert Frederick, then a thirty-one-year-old senior technical manager, was one of the prime developers of the project. He had already created software that could pull out data from Amazon and reformat it to be displayed on the tiny screens of mobile phones and personal digital a.s.sistants so that people could browse the site from their mobile devices. Some of his bosses suggested the idea of making similar data available to partners. So Frederick separated the databases holding product information from the software that ran the company and stored pa.s.swords and credit card information. That prevented sensitive data from leaking out. Despite some concerns from higher-ups about security, Frederick recalled, "The funny thing is that it did not take a great deal of convincing."