(FORTUNE Magazine) – A year ago, getting Jeff Bezos to talk about making money was a bit like getting Bill Clinton to define sex. Last fall, when asked when he thought Amazon.com would turn a profit, he hemmed, hawed, and mumbled something about not "missing out on the big opportunities of the Internet." Pressed further, he gave this murky response: "Look at USA Today; it took 11 years to become profitable."

Today you can't get Bezos to stop talking about making money. Literally. During one recent hourlong interview, the word "profit" or some variation thereof came up 25 times, as in, "We are very much driving toward profitability." Indeed, Bezos insists Amazon is thisclose to making money. He's even set a deadline. "We have, for the first time, set an internal goal with the date for when the company as a whole is going to be profitable," he proudly declares. In an internal e-mail sent out to 7,000 Amazon.com employees, which he showed FORTUNE, Bezos laid it out. "We're putting a stake in the ground: We're going to become profitable," he wrote. "That's right: We're aiming to have sales of $5 billion, produce over $1 billion in gross profits, and achieve solid operating profitability by ..." Bezos blotted out the date, citing the company's policy against making forward-looking financial statements. But one of his 7,000 employees told me the deadline was next Christmas.

Of course, the reason for the new rhetoric is no secret. A year ago, when e-tailing was hot, when Bezos was Time's Man of the Year and could blow off questions about profits, Amazon's stock price was $113--up 6,500% from its IPO. But, as Bezos wryly puts it, "you go from Internet poster child to Internet whipping boy--it takes like 30 seconds." Thirteen Wall Street analysts downgraded their ratings on the stock this year. This past summer a bond analyst declared that the company was at risk of running out of cash by early 2001. An article in Barron's suggested that Bezos was a fraud. With his stock now trading at $27--76% off its high--Bezos simply must talk about profitability, whether he likes it or not.

What's surprising is that Bezos, after five years in which his company has lost a total of $1.74 billion and borrowed $2 billion, actually seems to mean what he says about profits. Over the past year he has walked the walk: He has put Amazon through a massive overhaul by slashing spending, revamping the culture, laying people off, and hiring old-economy whizzes to teach him Six Sigma and inventory management. The result is a leaner, more efficient retailer, one that most analysts believe will turn a profit by the end of 2002. That seems auspicious, until you consider that even now, at 76% off the stock's high, investors value Amazon at $10 billion--more than Barnes & Noble, Kmart, and J.C. Penney combined. So, as the company gets closer to profitability, the question lingers: What kind of company will this really be? Can Amazon live up to the dream it once cast before our eyes, that it would become the first truly great company of the new economy? Or is Amazon in fact something far more mundane, something as mundane as--gulp--a retailer?

We'll return to that question later. First, the good news: Forget the hyperbole about whether Amazon will survive the dot-com shakeout. It will. This is a company that should sell nearly $3 billion of books, Razor Scooters, TVs, lawn mowers, and more to some 25 million customers in 150 countries this year. Its brand name is more recognizable than Burger King, Wrigley's, or Barbie. In addition, the company has $900 million in cash--more than enough to last it through 2001. Even Amazon bears concede that the company isn't about to disappear. Fay Landes, an analyst with Sanford C. Bernstein, who rates the stock an "underperform," states firmly: "Amazon is not going to go bankrupt." While Landes does expect Bezos to miss his internal deadline, she thinks the company will break even by mid-2002.

That should be encouraging to Bezos, a former Wall Street exec who (despite protestations to the contrary) pays close attention to stock analysts. Indeed, it was thanks to analysts that he knew long ago that Amazon would have to change its act. In particular, it was the questions he received during an analysts' call in the summer of 1999--a full nine months before last April's tech-market crash. "Everyone thinks it all happened in April of this year," says Kelyn Brannon, former CFO of Amazon.com International, who left last January. "We saw the turn--how the market was going--after the June [1999] conference call. The tone of the calls, the questions, the whisper conversations that went on afterward--it was a different tone. Rather than saying great quarter, great revenue growth--we stopped getting so many 'greats.' You get right into hard questions: Can you talk to me about direct margin? Can you talk to me about the operational efficiencies in your distribution center?"

Until then Bezos and Wall Street had formed a mutual admiration society. The Street side was embodied by Merrill Lynch analyst Henry Blodget, who defined the Amazon dream in his 1999 report entitled "What Amazon Could Do." Free of the costs and problems of physical stores, Amazon could be the most profitable retailer anyone had ever seen. According to Blodget, its operating margins could grow to 12%; Wal-Mart's are 6%. Losses? Amazon's losses weren't real losses, he explained during a quarter in which the company blew through $138 million. "Amazon is investing money," he wrote, "not losing it."

In that atmosphere the company had just one mission: make sales grow. "Our initial strategy was very focused and very unidimensional," says Bezos. "It was GBF: Get big fast. We put that on our shirts at the company picnic: They said GET BIG FAST, and on the back, EAT ANOTHER HOT DOG." The theory was simple and beguiling: Once Amazon got big enough (Blodget, Bezos, and the other believers never made clear how big), the efficiencies of a virtual store would kick in, and Amazon would start making money.

GBF also meant, more or less, spend at will. Any expense was justifiable, as long as it helped the company grow. There was no formal budget. Employees simply spent what they needed to keep sales up. If it cost $100 to get out a $10 order, so be it. One former Amazon employee recalls sending items out regularly by messenger service to make sure customers didn't cancel orders.

Launches of its virtual stores were lavish. Over the past two years Amazon has opened 31 new businesses--selling toys, electronics, cars, kitchen appliances, and so on. Last year those new stores lost $300 million. "Say we knew that we wanted to launch auctions by a certain time," says program manager Jonathan LeBlang. "We would have gone and bought a larger, more expensive computer that we could get this week rather than a smaller, cheaper one that had a three-week delivery on it." To ensure that those new stores had enough Pokemons, George Foreman grills, and PlayStations, the company stocked up on inventory--a strategy that resulted in a $39 million write-off in the fourth quarter of 1999.

GBF also wreaked havoc in the warehouses. Last year the company spent an estimated $200 million building seven distribution centers around the country--from Nevada to Kentucky. The idea is that if these three million square feet of warehouse space are run efficiently, Amazon will be able sell to as many customers as, say, Wal-Mart at a fraction of the cost. But when Jeff Wilke, Amazon's new operations chief, got a look at them for the first time last fall he was stunned: They were a mess. There were defective products on the shelves and mystery shipments arriving that no one remembered ordering. Once a truckload of kitchen knives showed up--no one knew where it came from or where it was supposed to go. It just sat there. "We kept it all--we just kept it," says Wilke, shaking his head. "We put it on the shelf and said, 'I don't know. What matters is the customer.' " By the end of 1999 it seemed that GBF was still in the saddle: While sales were up 169%, to $1.6 billion, net losses had risen from $125 million in 1998 to $720 million.

But the truth was that behind the scenes Bezos was working to ensure that Amazon got its spending under control. His key move, in fact, had occurred in June when he reached into the old economy to hire Joe Galli, a 19-year Black & Decker vet.

Right off the bat, chief operating officer Galli played hardball. He hired a slew of new managers, all with old-economy backgrounds: Delta, NBC, AlliedSignal, MCI. He pushed Amazon to start structuring itself like, well, an old-economy company. He forced it to set up a formal budget. He established an approval process for expenses. For example, all major purchases--computers, distribution machines--required signoff by top management (i.e., Galli). He hired Wilke to apply the Six Sigma methods to the distribution centers. He also presided over the company's first major layoffs, when he let 2% of the work force go in January.

Galli was not a popular guy. Accustomed to Bezos' freewheeling leadership, employees found Galli heavy-handed. He was stifling innovation, they complained, hurting Amazon's close-knit culture. "It was cruel the way he laid those people off," sniffs one former exec, who left in part because of Galli. The COO became notorious last fall when he took away the employees' free Tylenol and aspirin. It was a small perk, but for many employees Galli's move was a symbol of everything they hated about him. The outcry was so fierce that the company restored the perk within a week. "That was, frankly, just a mistake that Joe made. Like, oops, you know, people do really spend a lot of time at their computers," recalls David Risher, general manager of Amazon's U.S. virtual stores. Galli, who left in July to become CEO of VerticalNet, insists his moves were for the good of the company. "Some steps are less popular than others, because anytime you become more disciplined, you have to change your behavior," he says.

While Galli got the blame, Bezos was the man behind the orders. In fact, when asked about Galli, Bezos makes the chain of command clear: "The senior management team was very much in step, and we knew exactly what we wanted to do." And in the days since Galli's departure, Bezos has not let up. Gone are the days of "Get big fast." Bezos' new motto is "Make some great cash, baby." Instead of looking for ways to grow sales, employees look for ways to save money. "So, if it's a tradeoff--two weeks to do a project for $200 or three weeks for $100--18 months ago we would take the two-week, $200 approach," says LeBlang. "Today I do it in three weeks for $100." Now every division must carefully account for what it spends. Each meets weekly to go over its numbers. Executives must also write operating plans that outline specific financial deadlines and target precise sales goals and margins. "Now we have discipline," says Brian Birtwistle, product manager of the online software store. "You map out everything--what marketing you'll do for the year; what initiatives you'll launch; what you have to do to make those numbers realistic." Sounds like Business 101, and indeed it is. To learn the basics of P&Ls, balance sheets, and cash flow analysis, Amazon employees now take Finance 101 courses offered at the Seattle headquarters. When they've completed that, they take Finance 102.

In October the company showed Wall Street that it had learned its lesson. Analysts had expected a third-quarter loss of 33 cents a share, but the company's loss was just 25 cents. Operating losses as a percentage of sales dropped from 22% to 11%. Gross margins were a company record 26%, up 20% from a year ago. What's more, the books/music/video division actually turned a profit, earning $25 million on $400 million in sales. Wall Street was impressed and gave Amazon's stock its first boost in a while, sending it up 28%. "They've made tremendous progress," says David Ponneman, an analyst at Lincoln Capital. "I believe Amazon has the potential to become one of the most successful retailers in the world."

Lincoln Capital has a 5% stake in Amazon. So it's fair to ask, What kind of an investment is that, really? If Amazon turns out to be no better than an ordinary retailer--a Kmart, a J.C. Penney, a Sears--it's a bad one, to be blunt. Kmart trades at just 0.2 times sales; if Amazon were valued on that scale, its market cap would tumble from $10 billion to $540 million. And if Amazon turns out to be "one of the most successful retailers in the world," like, say, Wal-Mart...it's still a bad one. Wal-Mart trades at 1.2 times 2000 sales; value Amazon on a similar scale, and its market cap would drop to $3 billion.

That's why Bezos knows he has to make the case that Amazon is fundamentally different from a regular retailer, fundamentally another, better beast altogether. It's a case he loves to make, and here's how it goes:

"The foundation for e-commerce is technology, as opposed to the foundation for retail commerce, which is real estate," he begins. "As real estate gets more and more expensive, technology gets cheaper and cheaper. That's the fundamental analysis for e-commerce." In the real world, Bezos argues, every time Wal-Mart opens a new store, it has a brand-new set of costs: new rent, sales staff, electricity bill, cash registers, and so on. When Amazon, on the other hand, opens up a virtual store, it has none of those costs. Over the past few years the company has paid upfront to build its basic foundation--the Website, distribution centers, servers, and customer-service centers. That foundation serves as an e-tailing mold, so Amazon can sell just about anything--electronics, kitchen appliances, shovels--without incurring a bundle of new costs. If Target, on the other hand, wanted to start selling lawn mowers, it would have to pay for more space and sales staff; Amazon just has to tack a new tab onto amazon.com.

That's the theory, anyway. The trouble is that it turns out to be as full of holes as "Get big fast." E-tailing is not a one-size-fits-all business, as Amazon's own experience so painfully shows. In fact, every time Amazon has opened a new virtual store, it has taken on a whole new set of costs.

Consider, for example, Amazon's online electronics store, which it launched in July 1999; after a heavy promotional blitz, the business is Amazon's second biggest, after books. But trying to sell electronics is completely different from selling books. Books are small, cheap to ship, have a long shelf life, and come from a very limited group of suppliers. Electronics come in oddly shaped packages (think TVs--there's a 13-inch version, a 65-inch, and a flat-screen); their shelf life is short (who wants last year's TV?); they are very expensive to ship (big-screen TVs, for example, require special delivery). What's more, there are dozens of manufacturers--Sony, Uniden, Pioneer, JVC, Toshiba, Konka. And several--including Pioneer, JVC, and Panasonic--don't want to alienate their existing vendors by doing business with Amazon. So they don't sell their products to Amazon.

Thanks to all of that, Amazon had to run up big costs to launch its electronics store. It had to buy Pioneer, Panasonic, and JVC products from resellers at a markup of around 5%. It had to redesign its warehouses to handle such cumbersome packages as, say, a home theater system or a 70-pound TV stand. To compete with local electronics stores it had to hire service people to go to customers' homes to help set up their DVD players or TiVos. What's more, it had to advertise--a lot. Turns out that just having an electronics tab on the site isn't enough to turn an Amazon book buyer into an Amazon TV-set buyer. In fact, one group of Amazon book customers surveyed by Sanford C. Bernstein said that they weren't even aware that the site sold other products. "I probably wouldn't buy a TV on Amazon," says Steve Martin, a 37-year-old software engineer who's been buying books and software there for the past four years. "You pay all that money, it comes, it doesn't work--you want a local company to come deal with it."

What's true for electronics is true for toys, kitchen appliances, hardware, pool and patio--you name it. Each new business Amazon goes into has its own set of problems--problems that incur a new set of costs. Amazon sells Eureka vacuum cleaners in its kitchen section, but it has no relationship with Eureka--so it must buy from resellers. Or how about the lawn business, where the company sells garden tractors among thousands of other items: What's the synergy between the reader of Wuthering Heights and the rider of a lawn tractor? In 1999 Amazon's big new business was supposed to be toys, but it messed up so badly that excess toy inventory accounted for most of last year's $39 million write-off (the company launched a partnership with Toys "R" Us in August). And when Amazon ventures overseas, these problems are exacerbated. This year Amazon launched sites in France and Japan. So it has to pay to build two new distribution centers, hire two new staffs on two continents, puzzle through different sets of tax codes, and so on. Finally, none of this addresses the cost of promoting and advertising the new stores--a cost that's hard to forget, given that the company spent $224 million on marketing last year.

Indeed, just like real-world retailers, Amazon's costs seem to grow with sales. For example, in the first nine months of this year Amazon's sales grew 86% over the same period last year; its fulfillment costs--the company's biggest expense--grew 118%. Not surprisingly, Amazon's margins end up getting squeezed about as tightly as those of real-world retailers. As analyst Jeff Fieler of Bear Stearns puts it, "Let's face it: This company is a retailer, and the margins just aren't that great." According to Bezos' own calculations, when Amazon turns a profit its gross margins will be 20%. Wal-Mart's gross margins are 22%; Best Buy's are 20%; Barnes & Noble's are 28%. Most bullish analysts now estimate that when Amazon becomes profitable, its operating margins will be around 6%. That's not bad at all; in fact, it's a number all kinds of retailers are happy to deliver, from Wal-Mart to Target. The only problem is that it's about half of what Blodget once conjured; it's a long way from the dream. Says Dan Nordstrom, CEO of nordstrom.com: "Expectations are so high--higher than Amazon can deliver."

Bezos, of course, rejects such pessimism. "This is still day one. It's still the very beginning," he says. "We're not even in our awkward teenage years yet." Asked to describe the Amazon of the future, he says, "Our mission is to be earth's most customer-centric company. We will raise the worldwide standard of what it means to be a customer-obsessed company."

Bezos' beautiful dreams have swayed investors before and may sway them yet again. But it's getting harder to keep reality from creeping in and spoiling the dream.

FEEDBACK: kbrooker@fortunemail.com