How many gas stations are there in the United States?

Jeff Bezos likes to put this question to executives who want to work at Amazon.com. It has nothing to do with Amazon's business--petroleum products are among the few things the online retailer does not yet sell--but everything to do with Bezos's assessment of raw intelligence. Bezos does not care about the exact number of gas stations, only the back-of-the-envelope calculation that the candidate devises to get the answer: The process yields a snapshot of how the prospect's brain is wired.

While all corporate chiefs want to hire bright people, Amazon's founder prizes "genius" as a gold standard, and hiring it has been a long-held imperative. Amazon's screening process sometimes includes asking younger job candidates for their SAT scores.

Bezos's appreciation for powerful minds was nurtured in a series of privileged environments. At 9, he was a subject of a book about education for child prodigies. As a Miami high school senior, he celebrated his acceptance to Princeton University by taking his girlfriend for a weekend in New York. At 26, he was a wizard among whiz kids at one of Wall Street's most exclusive hedge funds. He was 30 when he founded what would become the dominant force in online retailing.

Today, as the era of dot-com indulgence fades, Amazon is suffering through a season of bad projections, front office defections and stock deflations. The company has come to embody all that is possible and precarious in Internet business. But in a series of conversations this summer, Bezos (pronounced BAY-zos) remained resolutely confident of his company's health.

People who know Bezos say the philosophies he has applied at Amazon follow a principle he has long embraced: Intrinsic ability rules, trumping acquired skills and accumulated experiences. By any measure, Bezos and his super-achievers have started a revolution in electronic commerce. But the question now is, where is it leading?

Among the leaders of the New Economy, Bezos offers perhaps the starkest contrast with the up-from-nothing titans of the Industrial Age. Where many of his corporate forebears had firsthand experience with poverty, Bezos is the child of affluent, suburban comfort and a close-knit family. From within this world view, his competitive drive stems more from his joy in the test than from an appreciation of want or failure.

One of Bezos's former Wall Street co-workers dubs him "a Master of the Universe for the Internet Age," meaning he has the same blue-chip credentials and sense of invincibility that marked Wall Street in the 1980s. Indeed, Bezos's designs and intentions seem to argue against the idea that Internet commerce is a newly leveled field where anyone can play because the costs of participation are so low.

In the shorthand mythology of cyberspace, Bezos invented Amazon.com while he and his wife were driving cross-country in a used Chevy Blazer. As Amazon grew and Bezos's paper wealth shot into the billions, he cultivated the persona of an Everyman for Giddy Times. He speaks about his company's fortunes with the awestruck enthusiasm of a lottery winner.

"We were hoping to build a small, profitable company," Bezos says. "And of course, what we've done is build a large, unprofitable company." Then he lets loose with his trademark laugh, a resounding hiccup in which he strings six articulated "hahs" together--hah-hah-hah-hah-hah-hah!--"like a jackass gargling bumblebees," says his friend Paul Saffo.

The laugh comes out of nowhere and packs uncertain meaning, much like Amazon itself. Bezos didn't set out to build a retail empire in 1994, any more than he set out to be valedictorian of the Palmetto High School Class of 1982; but as soon as he saw what outsize goals the Internet would allow for and what its investors would fund, his achiever's addiction kicked in.

Bezos spends much of his time recruiting and screening executives. One of his standard interviewing techniques is to pop oddball questions, a common practice in high-tech companies. Bezos's favorite is the one about how many gas stations there are in the United States.

He asks it sheepishly. "I realize it's a little weird to ask things like this to people who have 800 people reporting to them somewhere," he says. If it were put to him, Bezos would draw on his own experience: He'd figure there are 3,000 people in Cotulla, Tex., where he spent his boyhood summers with his grandfather. Cotulla had two gas stations. "I'm going to say there's a gas station for every 1,500 people," he says. "And now I know there are 280 million people in the U.S. Divide that by 1,500, and there's my guess."

That figure is 186,666. The American Petroleum Institute says there are 175,000. Off the back of his envelope, Bezos comes within 6.7 percent.

'A' Team Credentials

"There's nothing wrong with asking for SAT scores," Bezos says. Several of his employees blew away his 1,450, in fact. SATs are one minor metric in an elaborate assessment of job candidates. "If you start out with A's, you get to keep A's," he says. "If you start hiring B's, B's hire B's."

Is there a place for the B student at Amazon.com? "Certainly the answer is yes," says David Risher, an Amazon senior vice president. "But at the leadership level the answer is no, in part because that person would stick out like a sore thumb, and frankly would not interact very well with Jeff."

In 1996 Bezos interviewed Joy Covey, a candidate to be Amazon's chief financial officer. He noted that she had finished second in a field of 27,000 the year she took her CPA exam. "Why not first?" Bezos asked, half kidding. Covey, who had law and business degrees from Harvard University, said that she didn't study. She got the job, and remained at Amazon until recently.

Candidates for top positions can be subjected to five to 10 reference checks, which Bezos often conducts himself. They can take 60 minutes each and include a battery of 23 questions, including: Can you think of a problem that everyone thought was unsolvable that this person solved?

"If this person were really brilliant, you can remember these things," Bezos said. "If they can't think of anything, it doesn't mean they're not brilliant. But it's certainly a negative indicator."

Bezos, no brooding nerd, is a rare extrovert among high-tech pioneers. He is 5 feet 8 and 170 pounds, but seems endearingly smaller than life, speaking around a mouthful of peanut butter crackers, fumbling with his cell phone when his wife, MacKenzie, calls to remind him to bring something home ("Love you, too"). At 36, he has thinning brown hair and a puppy's brown eyes that invite a mothering impulse. Several of the 60 people interviewed for this profile said they felt very "protective of Jeff."

He wears blue dress shirts and khakis, and often roams around Amazon headquarters. He tells jokes and eagerly flashes his badge at security checkpoints, as if showing his normal-guy credentials. You might never know he was Time magazine's 1999 Person of the Year, except that the framed cover of the issue is one of the first things a visitor sees in the front lobby.

Amazon's headquarters conveys other mixed messages. It is a stately, 12-story brick complex southeast of downtown Seattle, a converted hospital. There's a foosball table inside the front entrance, near a security guard who escorts visitors to the restroom. The company has all the egalitarian trappings of a typical dot-com--casual dress, flattened hierarchies, stock options for everyone.

Since super-achievers often cluster, Amazon, like many high-tech firms, tends to recruit managers from a New Economy version of the old boys' network. Many of them graduated from the same schools and worked together at the same companies. Recent hire Jonathan Leblang graduated from Palmetto High with Bezos. Risher attended Princeton with Bezos and was a member of the same eating club, where he recalls Bezos was a ferocious player of beer pong. Risher, like several Amazon managers, came from Microsoft Corp., a company Bezos studies closely and admires for its rigorous hiring practices.

As many technology leaders do, Amazon executives trumpet their organization as the ultimate meritocracy, an institution where skill, commitment and drive matter most. Bezos says that brilliance alone is not all he cares about, and that Amazon's hiring process is not one-dimensional.

But his idea of diversity tends to focus on "the diversity of genius." He wants to assemble a collection of minds that are variously wired. "There is no Amazon.com 'type,' " it says on the "our culture" page on the Web site. "There are Amazon.com employees who have three master's degrees and speak five languages . . . people who have worked at Procter and Gamble and Microsoft . . . a professional figure skater . . . a Rhodes scholar." One of his employees was a National Spelling Bee champion, Bezos says. "You can shout out 'onomatopoeia' in the hallway, and you'll get an answer." He is looking for "a superstar dimension" in every job candidate, he says.

In interviews about hiring, Bezos and top Amazon managers do not bring up cultural or racial diversity; sources within the organization say it is simply not a priority in recruiting. Amazon does not provide a demographic breakdown of its employees. Its top executives and board of directors are predominantly male and white, but this profile is common among technology firms. Eighty-seven percent of managers at high-tech companies are white, according to the most recent Equal Employment Opportunity Commission figures available; 75.3 percent are men. Amazon will be represented at job fairs for black and Hispanic MBAs later this year, a company spokesman says.

Bezos argues against the notion of diversity as defined purely by race and ethnicity--and the notion that Amazon's staff is homogeneous. "I believe a company would be cutting its throat if didn't consider a wide variety of talent," he says. Amazon is committed to becoming more ethnically and racially diverse, he says.

If there is a recurring criticism of Bezos among people who know him, it is that he can be impatient with frailty. This has become increasingly difficult as Amazon has grown into an organization with thousands of front-line employees. They often work 55 or 60 hours a week for $11 to $14 an hour, with small grants of stock options that have lost value as Amazon's stock price has declined. The employees' output and performance is monitored closely. The company has struggled with low morale and high turnover among some customer service and distribution center workers. A union organizing campaign, uncommon at Internet companies, has sprouted within the ranks.

"I found the culture at Amazon to be terribly unforgiving," says Julie Drebin, who worked as an Amazon customer-service representative for two years until she left in May. "If you dared to raise issues about the company, it was as if you were defying these people who were so elite, they were beyond being questioned." Read this quote, Bezos chose not to comment.

"I love Jeff Bezos," adds a former colleague who worked with Bezos in the early 1990s at the Manhattan hedge fund D.E. Shaw. "But he has blind spots to certain realities, especially when it involves the limitations of other people."

Bezos himself has always inspired enormous faith--from parents, bosses and financial backers--and the experience has nurtured in him an unusual sense of possibility. His ascent has come on a series of intellectual playgrounds where he has had to contend with few limitations of his own.

The Gifted Student

One day when Jeff was 6, he was playing with a friend and mentioned that he had attended his parents' wedding. That's impossible, the friend said. You weren't born yet. When he got home, Jeff had a question for his mother.

"I explained it all to him in terms that made me comfortable," says Jackie Bezos, "which meant I was maybe dancing around the subject." What she explained was that she had been married before, it didn't work out, and she and Jeff went back to live with Nanny and Pop. Jeff looked up with his wide eyes and quietly asked, "You mean you got a divorce?" His mother nodded.

In 1964, two weeks after she turned 17, Jackie Gise Jorgensen gave birth to a baby boy. She was a high school student in New Mexico. Eighteen months later, her marriage ended. She and her son moved back home with her parents.

"From the minute I knew Jeff was going to be born, all I wanted to do was be a good mother," says Jackie Bezos. She hauled Jeffrey to her community college classes, changing his diapers at lunchtime.

When Jeff was 2, she took a job at a bank, and there she met Miguel Bezos, a recent immigrant from Cuba. Miguel says he fell in love with both mother and son simultaneously. He and Jackie married when Jeff was 4, and Miguel adopted Jeff. Until that day when he was 6, Jeff assumed that Miguel was his biological father.

When Jeff was 5 and 6, his siblings, Christina and Mark, were born. Miguel took a job as an engineer at Exxon Corp. in Houston and put in long hours.

Jeff had another male role model in his grandfather, Lawrence Preston Gise, who worked on space and defense technologies for the government. In 1968 he retired to his ranch in Cotulla, and Jeff spent almost every summer there. Pop doted, telling stories about missile defense systems and teaching him to lay pipe and castrate bulls.

That Jeff was unusually bright was confirmed by a standardized test when he was 8. His parents enrolled him in a pilot program for gifted students at Houston's River Oaks Elementary School, 20 miles from their home. One of his early feats involved a teletype machine that could be connected to a mainframe computer by a modem. None of the teachers knew how to use the computer, but Jeff and a few other children figured out how to program it.

When Jeff was 9, he was described in the book "Turning on Bright Minds: A Parent Looks at Gifted Education in Texas" as "friendly but serious," a boy of "general intellectual excellence."

"It also said I had no leadership skills whatsoever," he recalls. "Hah-hah-hah-hah-hah-hah!"

His mother pushed him to go beyond his geeky comfort zone, signing him up for a youth football league--a terrifying notion to Jeff, who was small for his age and had never played before. He compensated by memorizing his team's offensive plays and defensive formations. Coach Meisner charged him with calling plays on the field.

In 1978, Exxon transferred Miguel to Miami, where the family lived in a four-bedroom house with a pool in the affluent Palmetto district of Dade County. Jeff enrolled at Palmetto High, an incubator of high achievers. He gravitated to a group of about 10 kids from his honors classes. They engaged in gentle high jinks, such as hanging yarn from classroom lights before school. "He was not a problem kid," says Jackie Bezos. "If Jeff was ever guilty of anything, it was crimes of the mouth."

After the mentorship of Pop Gise and a diet of "Star Trek," Jeff fixated on outer space. He had always been taken with far-off exploration and magical kingdoms--he had visited Disney World seven times. Bezos believed that space travel would soon be a booming sector, and he wanted to start a shuttle service that would spare companies the cost of launching a booster rocket.

He studied and competed exhaustively. At the beginning of his senior year, he realized he had a chance to be valedictorian of the Class of 1982. He had aced all of his honors classes and liked his chances. On the day class ranks were given out, seniors were called into the principal's office, their fates inscribed on index cards. "Mine had a 1 on it," Bezos recalls. "And it felt very gooood . . . hah-hah-hah-hah-hah-hah!"

He left Miami for Princeton University, the only school he wanted to attend. "Einstein was there, for goodness' sake." He dreamed of being a theoretical physicist until he discovered he was no longer the brightest kid in the room.

'A State of Awe'

One night during his freshman year, Bezos was struggling over a partial differential equation. After a few hours, he and his study partner visited the dorm room of a classmate, who glanced at the equation and said, "Cosine."

"After we expressed some incredulousness," Bezos says, "he proceeded to draw three pages of equations that flowed through and showed that it was cosine."

It led to a practical realization: There were people whose brains were wired to process abstract concepts in a very graceful way, and he was not one of those people. "This is not something that leads me to be depressed," he says. "This is something that leads me to a state of awe."

Bezos switched his concentration to electrical engineering and computer science and compiled a final grade-point average of 3.9, 4.2 in his major. (A-pluses counted as 4.3.) Like many college students, he grappled with questions of identity; this led him back to the silent issue of his biological father. For the first time since age 6, Bezos raised the topic with his mother. One summer day he asked a series of questions, which she does not share. When he was through, he gave her a hug and said, "You did a great job, Mom." Today, he says his biological father is "a total non-issue." Miguel Bezos is his father, end of story.

After graduating summa cum laude from Princeton in 1986, Bezos weighed job offers from Intel Corp., Andersen Consulting and Bell Labs. But he took a job at Fitel, a New York start-up that was building a global telecommunications network to settle cross-border equity sales. He liked the challenge and deemed the founders brilliant. He was employee No. 11, entering the flamboyant domain of New York finance in the 1980s.

Running Into Resistance

He slept on People Express planes. He shuttled regularly across the Atlantic. At 24, he was head of customer service and software development, and those divisions were based in New York and London, respectively.

Fitel offered an early look at how connected computers could automate a business process. Bezos--adept at both debugging code and schmoozing clients--was part of a team that built what was essentially a mini-Internet for investors, sellers, brokers and banks.

But those clients tended to think conservatively about new technology, and Fitel struggled to attract new customers. In April 1988, Bezos left for a job at a financial services powerhouse, Bankers Trust Corp., where he developed software applications for the company's pension fund clients. After 10 months, at 26, he was made a vice president. As at Fitel, however, Bezos faced resistance to the technology he was developing.

"There was an old guard inside Bankers Trust that saw no reason to change anything," says Harvey Hirsch, a former BT executive who ran Bezos's division. Bezos was not shy about advocating personal computers, but several colleagues scoffed, saying that PCs did not have sufficient memory to handle elaborate financial data.

"He always assumed his brilliance would get him ahead," says Kelsey Biggers, Bezos's closest friend at BT and a fellow proponent of speeding the pace of change. Bezos was more amused by than hostile to the old guard, Biggers says. "We'd say, 'We're the mammals eating the dinosaurs' eggs.' "

By 1990, Bezos was tired of being a technologist at a non-technology company. But a headhunter urged him to consider one more: D. E. Shaw.

It had a mystique as one of the most technically advanced financial shops in the world. The founder, David Shaw, had a PhD in computer science from Stanford; he was interested in both technology and finance, and was an expert in devising new trading strategies based on complex mathematical formulas.

Shaw was a "true genius," Bezos says, a person unusually adept at using both sides of the brain, seeing big pictures and small patterns. Many of the business practices Bezos later adopted at Amazon he learned from Shaw, beginning with hiring. "We don't always recruit for specific positions," Shaw says. "We're happy to warehouse a truly gifted individual on the assumption that they may someday make us money."

After two years, Bezos was named a senior vice president, at 28 the youngest of four at Shaw. He spent much of his time exploring new business opportunities for the firm. Bezos spent many nights in a sleeping bag he kept rolled up in the back of his 39th-floor office.

He was also seeking a wife. He was in his late twenties, making a salary in the high six figures and ready to settle down. He devised a principle called "woman flow," a variation on the Wall Street term "deal flow," which ensures a stream of potential deals or, in this case, miserable blind dates.

In the end, Bezos fell in love with MacKenzie Tuttle, a Princeton graduate and D. E. Shaw research associate who worked in his group. There was no mutual interest when she was hired. When it developed, Bezos was terrified of acting on it for fear of a sexual harassment claim. Finally, drawn by his laugh, she approached him. They were married at the Breakers resort in Palm Beach, Fla., in 1993.

In early 1994, Bezos began working directly with Shaw to investigate business opportunities online. After decades on the esoteric fringes, the Internet was coming mainstream with the advent of the World Wide Web. Bezos had first used the Internet in 1985, in a Princeton astrophysics class, but he never thought about its commercial possibilities until the spring of 1994 while working for Shaw.

Bezos and Shaw met for a few hours each week to brainstorm ideas; Bezos would then go off on his own and research them. He enumerated 20 products that could be sold online, among them software, trading services, compact discs and, somewhere down on his list, books.

But books kept rising. They could be easily sampled online. There were about 3 million books in print at the time, 10 times the number of CDs. The Internet's search capabilities made it easy to "browse" by author, title, publisher and keyword--an arduous process in a physical bookstore.

Bezos also found a study that said Internet use was growing by 2,300 percent annually. "I didn't even believe the figure when I saw it," he recalls. "I read through the methodology, and it became clear to me that this was a reasonable methodology."

Bezos's research coincided with another realization: He was ready to start his own business. He had contemplated possibilities for years, increasingly as he approached 30. He told Shaw he was thinking about starting an online bookstore himself.

They talked for two hours. Shaw told Bezos he could have a bright future at the firm but said he understood his desire to start his own--something Shaw himself had done when he left Morgan Stanley in 1988. Bezos took 48 hours to make a final decision.

He contemplated in what he calls a regret-minimization framework. "Most regrets in life are of omissions, not commissions," he says. He placed himself in a rocking chair at age 80, asking what he would regret more: leaving a job that came with perks like a six-figure Christmas bonus, or missing out on the chance to shape the Internet.

Bezos gave Shaw his final answer and planned a road trip.

From a Seattle Garage

In Amazon lore, Bezos headed West with no specific destination and veered toward Seattle. In fact, Bezos scrutinized many factors--supplier proximity, labor pools, state sales taxes--before he left Manhattan. Seattle won because it had a large reservoir of tech workers and it was a few hours from the largest book distribution center in the country.

Company name? Bezos liked Cadabra, as in "abracadabra," conjuring his fondness for the magical.

"Cadaver?" asked the Seattle lawyer who helped him incorporate the company.

Bezos buried that idea and moved to his next: Amazon.com. He didn't have a single customer yet, but he loved the name. Something huge. With no specific reference to any product, it would allow him to move the business in any direction.

This is no small point in understanding the rise of Bezos and Amazon. Selling books forever seemed too limiting. He needed to overhaul commerce. Computer screens would be the new store windows; physical stores would survive only if they provided something that could not be replicated online--physical ambience, entertainment value. He was not hostile to traditional retailers, only to their mediocrity. Strip malls, he would often say, were doomed.

Even though Amazon would be introduced to the world as an online bookstore, Bezos preferred "book service." "Thinking of yourself as a store puts you in a box," he says. "Services can be anything."

The company was launched in 1994 with a $300,000 investment from his parents and loans from his own bank account. Beyond that, Bezos scrambled to raise $1 million from 20 local investors--a major accomplishment, since Bezos knew few people in Seattle and the Internet was still unknown to most. Kelsey Biggers visited Bezos at the time, finding him sleep-deprived and frustrated. "He said maybe we'll make it, or maybe we'll be a tasty morsel for some other company," Biggers says. "That's the only time I ever heard Jeff show a flicker of doubt."

The Bezos garage was snaked with extension cords and cramped with a potbellied stove, three large computers and three engineers--the founder and his first two hires. Their first goal was to show that this new industry could be self-sustaining. Growth would be steady and methodical--and funded by whatever profits the company could muster; other outside investments would come later.

That was the traditional retail model. But the plan changed almost as soon as Amazon opened for business, in July 1995. In Amazon's first week, it processed $12,438 worth of orders. Within the first 30 days, it sold books to people in all 50 states--without advertising. Internet commerce was becoming a land grab overnight.

On Aug. 9, Netscape held an initial public offering--and its share price more than doubled on its first trading day, even though the company had no profits. The much-hyped IPO confirmed what was becoming clear to the Internet's first movers: The rules of starting and expanding a business--and raising money for it--were changing.

This meant, to Bezos's mind, pivoting from turning a profit any time soon. It was not a bold choice, but an obvious one, he says. "To have thought otherwise would have been management malpractice," he says. For a brief period late in 1995, Amazon was slightly profitable, a milestone Bezos would renounce as "unfortunate."

Profits meant stinting on marketing, advertising and infrastructure. Now, those three elements formed a tripod supporting Amazon's new corporate mantra: "Get Big Fast." Amazon finished 1996 with sales of $15.7 million--a jump of 3,000 percent over 1995. Its losses were $6.2 million, compared with $303,000 in 1995. But by this time, investor trepidation about the Internet had succumbed to early-stage hysteria. "I would give them pieces of paper with our weekly sales growth," Bezos says, "and they would say, 'Where do we send the check?' "

Amazon wanted to use technology to change not only the delivery of products, but also the paradigm of shopping. Its engineers developed the "collaborative filtering" software that would customize the shopping experience to each user's tastes.

Bezos sees this as a crucial achievement of his uprising. He challenges any comparison between Amazon and established retailers such as Wal-Mart Stores Inc. No matter how huge they might seem, bricks-and-mortar superstores are intrinsically limited. "You can subdivide the demographics of these stores into maybe 50 segments," Bezos says. But Amazon caters to segments of one--each customer, several million of them.

"Jeff is launching this giant cruise missile in a general direction, but he doesn't know where it is going," says Paul Saffo, director of Silicon Valley's Institute for the Future. "He's inaugurated a business model of 'Ready, Fire, Steer,' not 'Ready, Aim, Fire.' "

The company sold stock to the public on May 15, 1997, and by the end of the year the shares had risen 233 percent. The media loved Bezos as much as Wall Street did. His laugh became the background Muzak to a dot-com exuberance that may or may not have been rational.

But the laugh is more complicated than joy, and he deploys it to convey a range of messages. "He laughs when he's really happy, and he laughs when he's really frustrated," says Rick Dalzell, Amazon's chief information officer. "Sometimes I'll ask him why he's laughing and he'll say, 'Because I'd be crying otherwise.' "

As the century closed, Amazon had gone far toward fulfilling Bezos's anything, any time, vision: Its virtual bookshelves swelled with compact discs and videocassettes, shaving cream and hardware, Legos and dinosaur fossils. The company bought into numerous online retailers, such as Drugstore.com and Pets.com. In 1999, the company lost $719 million on $1.6 billion in revenue.

When Bezos was named Time's Person of the Year for 1999, one analyst dubbed him the author of "one of the smartest business strategies in business history." The "person of the year" distinction ensured Bezos's credentials as an Internet pioneer and brought tears to his mother's eyes.

But that was nine months ago.

A Question of Profitability

Amazon reflects the "financial characteristics that have driven innumerable retailers to disaster throughout history," Ravi Suria wrote in a financial analysis for Lehman Brothers Inc. in June. "The party is over."

Suria's report was widely read. Agreement was hardly general, but the report signaled that the rules were changing again: Old indicators were being applied to Internet businesses.

It marked the beginning of a long summer for Amazon. On July 25, President Joe Galli quit. The next day, Amazon reported lower-than-expected revenue, suffered a 45-minute outage, and was subject to another harsh report from a Lehman analyst, longtime supporter Holly Becker.

Bezos dismisses Suria's report as "pure, unadulterated hogwash" and says Amazon has never been in a stronger position than it is today. Long-term profitability remains his goal. Questions involving stock price, short-term profits and traditional retail models are basically irrelevant--"noise," in his term. Amazon has become inured to doomsday reports, he says. Wall Street has not; Amazon's shares have dipped to $41.50 following a December high of $106.

If these seem like tough times, Bezos says, it's only because everyone is paying attention. This closer scrutiny, however, was brought on partly by Bezos himself, when he chose to take on more than $2 billion in debt to fund Amazon's growth. While stock analysts make a parlor game of predicting when Amazon will make money, bond investors are more sober-minded: They want their principal and interest paid on time and in cash. As the company's debt load has increased, so has the urgency surrounding the question of when Amazon will be profitable.

Bezos has grown weary of this question. He says only that Amazon will become profitable "when [its] ratio of mature businesses to new businesses is higher," and that its only mature business today is its domestic book, music and video division. New businesses include all other Amazon product lines, such as tools and cars. He provides no time frame, and his confidence is unwavering.

"Amazon has created the best retail customer experience on the Internet, hands down," says Bill Whyman, president of Precursor, an independent technology research group in Washington. "The billion-dollar question now is whether that's good enough."

Like many industry analysts, Whyman is conducting extensive research on the buying patterns of Amazon's 23 million customers to gauge the direction of the online retail market. Amazon's average customer spent $26 in the last quarter, Whyman says, a figure that has gone down progressively from $47 since early 1997. Meanwhile, Amazon's data tell a different story: Each "ordering account" is spending $125 on the site each year, up from $106 in December 1998. The company does not report quarterly figures. "Ordering accounts" do not include the 22 percent of customers who have not bought anything on Amazon in the past 12 months.

On July 28, Bezos wrote a memo to staff members, a copy of which was obtained by The Washington Post. "We're putting a stake in the ground," Bezos wrote. "We're going to become profitable. . . . The senior team met for three days last week and created the outline of a plan that takes us to profitability in Q4 [the fourth quarter] of 2001." By that time, the memo says, Bezos expects to have sales of $5 billion, produce more than $1 billion in gross profits and achieve "solid operating profitability."

The memo would seem to declare an end to the age of Get Big Fast. Or does it? The company is making a big push to "Go Global," the memo also says, to seize on the explosion of electronic commerce in Europe and Asia. Bezos declines to comment on the memo but acknowledges that the efforts to "go global" and "put a stake in the ground" seem to contradict each other.

Bezos often fixates on empirical data, but building a lasting online empire is hardly an exact science. He mostly agrees with Paul Saffo's assessment that Amazon is a study in "Ready, Fire, Steer." He says the online world demands risk-taking, and getting bogged down in "aiming" can stunt innovation.

This is a prime reason for trying to create the highest admission standards in corporate America. E-commerce is still in "the Kitty Hawk age," he says; his faith in brilliance represents the ultimate long-term investment. Bezos wants a company that can follow the same pattern he has followed his entire life: take what's available and push it to its extreme.

Employees are more than workers, he says, they are inventors. He cites an everyday example of what the assemblage of brilliance can achieve. A customer-service representative suggested that Amazon's site include a help page in Spanish. No one acted on her suggestion, so the rep used her fluency in Spanish to do it herself.

For her efforts, the service rep received an Amazon "Just Do It" award. "We gave her a used Nike tennis shoe," Bezos says, laughing.

Staff researcher Richard Drezen contributed to this report.

About This Series

The Industrial Age produced an American business archetype--the self-made entrepreneur who was driven to amass wealth by memories of growing up poor. But the New Economy, driven by computing power and the opening of international markets, has given rise to a new sort of corporate titan--the child of middle-class values who nonetheless came to dream of global dominance. In this series, The Post is examining the experiences that shaped the New Economy's boldest leaders.

Part 1 of the series traced Steve Case's competitive drive from his roots in the suburbs of Honolulu to his leadership of America Online Inc.