(FORTUNE Magazine) – SIT DOWN TO visit with some of the old-timers at Microsoft Corp., and it doesn't take long to observe the imprint Bill Gates has made on the company he dropped out of Harvard to create 15 years ago. His adolescent-sounding techie slang creeps into nearly every conversation on the wooded Microsoft campus near Seattle. A few younger managers carry the mimicry almost to the point of caricature: When they talk, they give their vowels the same singing emphasis their 34-year-old chairman does. As discussions get animated, they hunch forward, prop their elbows on their knees, and start rocking back and forth in their seats, just like Chairman Bill. Yet Microsoft is no personality cult. Nor is it a corporate dictatorship. With some 5,200 employees, over $1 billion in revenues this year, and the broadest array of products in the personal computer software business, it is too big and too diverse to be either. From early on, Gates has relied on a handful of strong professional managers. Despite his reputation as the consummate computer nerd, Microsoft's growth testifies to his skills as manager and leader. Microsoft -- more than any other company to emerge from the personal computer revolution -- is the technological, organizational, and commercial ; embodiment of one man's vision. Its clarity of purpose, competitiveness, tenacity, and technological self-confidence emanate from Gates. And despite his company's swelling size, he has found an effective way to keep communicating all that directly. In a business where people eat innovation for breakfast, his personal qualities have helped him keep the magic going after Microsoft's initial, spectacular success. The company's other unique strength is its central role in a volatile, hotly competitive industry. That results from the dominance of DOS (for disk operating system), the software that Gates sold to IBM to control the innards of its first PC. IBM PCs and clones dominate the business, with 74% of their market. Microsoft has sold more than 40 million copies of DOS. That preeminence lets Gates plug into a lot of other people's secrets, because throughout the PC industry anyone developing new programs or machines is virtually compelled to work with his products. For years the owlish entrepreneur has been the industry's leading technological visionary and power broker. As early as 1975, when he and Paul Allen, a high school chum, started the company, he recognized that an obscure newfangled computer chip called the microprocessor would make possible full-fledged computers small and cheap enough for people to use in their homes and offices. Six years later he persuaded IBM to let him sell DOS to all comers, which allowed other manufacturers to make PC clones. The ensuing whirlwind of competition drove down prices, stimulated innovation, and created an $83-billion-a-year industry worldwide. Though much of the computer chip and PC hardware industries moved to Japan and elsewhere overseas in the 1980s, the role of DOS as the unifying component of most PCs has helped entrench the U.S. as the epicenter of world software. Gates's empire competes in nearly every nook and cranny of that industry -- including the computer languages professional programmers use to build their products, the operating systems and networking software that control and link groups of PCs, the spreadsheets and word-processing programs (''word prossers,'' in Gatespeak) that have become standard equipment in office desktop machines. With Microsoft's ballyhooed introduction of Windows 3.0 in late May, Gates is trying to put a friendlier face on PCs -- and carve out yet another lucrative market for his company (see box). Along the way he has amassed a personal fortune of over $2.8 billion, most + of it tied up in his 36% stake in Microsoft stock. Says Stewart Alsop, a prominent industry analyst who publishes P.C. Letter: ''Gates sometimes reminds me of the 19th-century industrial barons who by force of will and business genius built the oil, steel, and banking monopolies. He's been that important to the computer industry.'' Gates is in a class of his own. Other young, charismatic PC pioneers -- notably Steven Jobs of Apple Computer, now CEO of Next Inc., and Mitchell Kapor, who started Lotus Development -- built successful companies. But Jobs left Apple in 1985 after a power struggle; Kapor grew disillusioned with running a big corporation and stepped aside. Most high-tech companies make at least one major misstep on the way to becoming billion-dollar behemoths. That still hasn't happened to Microsoft. HOW HAVE GATES and his lieutenants managed to manage Microsoft so well? One key is his ability to spread himself around. While the technological strategies bear the Gates signature, he and his managers break them down into concise business goals that can be handled by small, independent ''business units'' of programmers and marketers. The groups are small enough so that Gates can sit around a table to chat with key members and inject his ideas personally. While he would prefer not to be bothered with operational details, he is quick to identify both organizational weaknesses and the management talent required to repair them. Although Gates freely delegates, his capacity for retaining business and technical details enables him to understand intimately both the forest and the trees. At meetings with his managers, he is as likely to check the math in handouts and overhead slides for errors as he is to critique fuzzy marketing strategies. He loves to join Microsoft programmers in the brainstorming sessions that give birth to new products. ''It's very important to me and to the guys that work for us that Microsoft feel like a small company, even though it isn't one anymore,'' says Gates. ''I remember how much fun it is to be small, and the business units help preserve that feeling.'' Says Scott Oki, 41, senior vice president for U.S. sales and marketing, an eight-year Microsoft veteran: ''Bill has the cumulative intellect of an octogenarian and the hormones of a teenager. Each part of the company has a life of its own now, but Bill is the glue that holds it all together.'' Gates puts it this way: ''No doubt the company would do well without me at this ) point, but I like to think the Microsoft clock runs faster because of me.'' Consider just how fast that clock has been ticking: -- Since 1982, the first full year after the deal with IBM, Microsoft has grown nearly 50-fold without a blip. Analysts predict revenues of $1.2 billion for the fiscal year ending in June, a 48% gain over fiscal 1989, and profits of $274 million, up 61%. -- Because of the company's near monopoly on operating systems software for IBM-compatible PCs, analysts estimate that it earns 60% pretax on this business. -- Microsoft's after-tax profit margins of 24% are the highest in the software industry, mainly because the company gets by with fewer programmers and marketers than its rivals. -- Gates pointed Microsoft abroad long before its competitors. Microsoft products predominate in nearly every foreign market they are in; they generate more sales and profits overseas than at home. -- The stock is one of Wall Street's top performers. It traded recently at a multiple of 34, vs. the teens and low 20s for most of its peers. -- Late last year Microsoft surpassed competitor Lotus Development, maker of the popular 1-2-3 spreadsheet, as the biggest seller of applications software -- word processors, spreadsheets, and the like. Lotus, which edged ahead again in March, then tried to merge with Novell Inc. of Provo, Utah, maker of the best-selling networking program for linking groups of PCs. That's the one type of operating system software Microsoft doesn't yet dominate, and the combined companies could have formed a counterbalance to Microsoft. But the deal fell through after a month, leaving Gates as firmly in command as ever. Proud as he is of Microsoft, Gates considers it far from being a company ''where everything is golden. We're in a serious investment mode'' -- that's Gatespeak for losing money -- ''in a whole slew of businesses.'' Among them: helping IBM establish a new, higher-performance operating system called OS/2 as a replacement for DOS. Microsoft is also trying to dislodge Novell as the leader in networking software. That is sure to become an attractive market as more businesses link PCs into networks that share programs and data.

The key to Gates's management strategy is Microsoft's dozen business units, some of which employ as few as 30 people. Each group is in charge of a particular type of software, and each knows exactly how it stacks up against its competitors by every conceivable measure, from the technical sophistication of a program to the amount of labor that went into building it. Business unit managers -- known inevitably as BUMs -- study the products of competitors along with any financial and productivity information their rivals make public. If the Microsoft unit doesn't outperform its competitors, the unit hears about it -- often from Gates, who is known to do the same homework himself. Microsoft works its employees hard. Many routinely put in 75-hour weeks, especially near shipment deadlines for new products. But turnover is low -- 8% to 10% annually for marketing and administrative types, even less among programmers. In top management it's practically nil. (An exception: President Jon Shirley, 52, who retires at the end of June to explore Puget Sound in his cabin cruiser and perhaps restore an old car or two. His successor is Michael R. Hallman, 44, former president of Boeing Computer Services.) Jeremy Butler, 46, senior vice president for international operations, brags that since Microsoft moved overseas in a big way in 1983, only two of the company's 21 country managers have left. They too retired. Employees don't earn enormous salaries; Gates pays himself $190,000 a year. Nor are they handed many perks. Nobody has a company car, and Oki had to pay for the fancy wallpaper, artwork, and plants in his office. But the company is generous with stock options. For example, before it went public in 1986, all of the several hundred programmers on the staff got options as if they were top managers. The leap of Microsoft shares (see charts) has transformed dozens of them into paper millionaires. Three-fourths of Microsoft's employees also buy shares at a 15% discount through a stock purchase plan. ''I could retire today and live off the interest from all the money I made from my options,'' says Jeff Harbers, 38, one of seven Microsoft ''architects'' -- the panel of technical high priests who advise Gates, figure out how to fulfill his grand software strategies, and explore potential new technologies. He adds: ''I took off last year thinking I might be tired of working, but I came back because this is what I want to do. Microsoft's job isn't finished.''

Gates, whose father is a prominent Seattle lawyer, says the promotion and compensation plan for his programmers is similar to that of a law firm. Each programmer is rated at one of six levels between 10 and 15, with architects at 15. ''When you hit 13, it's like making partner,'' says Gates. ''We have a big / ceremony and everything,'' including awarding the programmer more stock options. While some programmers also perform some management chores, they don't have to be managers to climb the scale. A nifty piece of programming will do the trick. All this inspires tremendous loyalty. The company's central role in the PC industry also helps, as does Gates's personal magnetism and accessibility. Chairman Bill encourages employees to communicate with him directly through the company's electronic mail system, and dozens do each day. He tries to respond to each message the same day he gets it. ''That's a big deal,'' says William Sahlman, a Harvard business school management professor. ''As a company gets larger and the charismatic figurehead gets further and further away from the bowels of the company, it's particularly useful for people to feel they have some direct access.'' Employees say Gates's messages are blunt and sometimes sarcastic but always entertaining. (He also communicates with his parents via E-mail, although his mother, Mary, a member of the University of Washington board of regents, admits that ''he doesn't always answer me right away.'') Good programmers are in short supply, but Microsoft can afford to be choosy. It's not unusual for the company to hire less than 1% of all those who apply for jobs, says Mike Maples, 47, a former IBM executive who is Microsoft's vice president for applications. The company is so selective in hiring partly because the boss is just plain stingy. ''There are two levers in our business -- head count and advertising,'' Gates says. ''Advertising expense you can change very easily, and if you make a mistake, it's easy to correct. With head count you have to be more conservative. Once you allow managers to think that it takes 100 people to do something when it should be 20, that's extremely hard to reverse.'' One chronic problem at Microsoft is partly of Gates's own making. Time after time the company has targeted a new market only to introduce a mediocre product the first time out. That happened with its original word-processing program for the PC, and more recently with Windows. The first versions of OS/2 and LAN Manager, Microsoft's networking software for linking dozens of PCs, also got off to slow starts. Says Vern Raburn, a former Microsoft vice president who orchestrated the company's move into applications software in the early 1980s: ''With few exceptions -- say, DOS and Microsoft's products for the Macintosh -- they've never shipped a good product in its first version. But they never give up, and eventually they get it right.'' Raburn says the problem can be traced straight back to Gates. ''Bill is too willing to compromise just to get going in a business. That's what happened to OS/2. Also, he's very action oriented and believes that it's better to make a decision and correct it later than not to make a decision.'' Gates finds it frustrating that a product can run through three versions before it finally clicks. ''We can't get used to this, because customers will start saying, 'Okay, let's skip the first version.' It isn't written in heaven that only the third version is a gold mine.'' He adds: ''The challenge of either competing with a market leader or establishing a new software category is very, very hard. If everybody at Lotus went on vacation for a year, we'd still have a hard time competing with 1-2-3 because its momentum would still be there. But once you get to the dominant position, the rewards are incredible, so you can afford to keep going after it. Besides, I hate abandoning things.'' DOES MICROSOFT have weaknesses besides a tendency to rush a product to market too quickly? Even Gates's biggest fans see a few cracks. ''Microsoft is a very seat-of-the-pants operation in certain ways,'' says Michael Swavely, president of Compaq Computer's North American operations. ''In particular, their marketing tends not to be well thought out, and their understanding of market research is only rudimentary. But that is typical of a fast-growing company, and it's fixable.'' A case in point, Swavely says, is the disappointing result of Microsoft's joint effort with IBM to promote OS/2, the next-generation operating system for IBM-compatible PCs. In the mid-1980s both companies recognized that the original DOS, pulled together hastily so IBM could ship its first PCs, wasn't advanced enough to keep up with the power of the more sophisticated hardware of the future. They decided to start from scratch and build what they thought would be the ideal new operating system. They proclaimed it the successor to DOS when they began shipping it in 1988. BOTH COMPANIES assumed that by 1990 OS/2 would be on its way to dominating the PC operating system market. After all, its many new features include the ability to run several programs simultaneously; it also has graphic menus, much like those on Apple's popular Macintosh, that make a computer easier to operate. But so far only about 300,000 copies have been sold, vs. some 25 million for DOS over the same period. Microsoft and IBM failed to anticipate that software companies would take a long time to write programs to run on the new operating system. (Microsoft itself has yet to ship a version of Word for OS/2.) It didn't help that the price of the extra memory required to run OS/2 efficiently quadrupled around the time the new operating system was announced, forcing users to pay a stiff $2,000 or so to upgrade their PCs. Other critics say that Microsoft products, while technically sound, lack the attractiveness and ease of operation that most computer users want. Some call it a philosophical problem. Says Mitch Kapor, founder of Lotus: ''I give Bill Gates an A for vision, because as a business person and a strategist he's brilliant. His flaw is that his view is not informed by a humanistic or compassionate vision of how to make computers work for people.'' Gates admits as much. ''One of my guys once said, 'We make the world's best plumbing, but we never think about the toilet seat.' But that's changing. I think we're getting better. Look at our new Windows product. I contend the icons in Windows blow everybody else's away.'' Rival Apple popularized icons, the tiny pictures on a computer screen that launch programs and open files when activated by a pointing device called a mouse. Microsoft faces some threats not of its own making. The distinction is beginning to blur between PCs and technical workstations, the high-powered desktop machines that scientists and engineers favor. So far workstations haven't made many inroads into business offices, the province of PCs, but they're trying. The longer it takes Microsoft and IBM to establish OS/2, the better the chance workstation makers have to grab a piece of that market. That's even more important to Microsoft than to IBM: While IBM has belatedly started making workstations, Microsoft writes programs almost exclusively for PCs. Then there are the recurring rumors that Microsoft's crucial relationship with IBM is unraveling. The two companies have been partners for nine years, most recently in developing OS/2. But they haven't always marched in step in marketing it. Also, in the past two years Big Blue has begun turning to others for help in developing new technologies. It has licensed some workstation operating system software from Steve Jobs's Next as well as technology ( developed by Go Corp., a tiny Silicon Valley startup, that lets computers read information handwritten by electronic pen, without the need for either a keyboard or a mouse. Neither IBM nor Microsoft will say that the relationship is strained. Says Gates: ''It's like we're married or something, and if IBM starts going out with somebody else, the gossip columns get all excited about it and come up with all these Machiavellian theories. I can understand that. But we're sincere people, and we mean it when we say this relationship is still fundamentally strong.'' It's clear, however, that there has been tension. Pausing a moment, Gates adds, ''In a way, because OS/2 hasn't taken off overnight, we've rededicated ourselves to working better with each other.'' Finally, what about Japan as a competitive threat to Microsoft and to longstanding U.S. dominance in PC software? ''I don't think the Japanese will ever do well here in that kind of software,'' Gates says. ''Japanese engineers are perfectly good at writing code. But a lot of what makes good software is marketing and other things, and there they suffer a cultural and linguistic disadvantage. They don't understand very well what word-processing features are important to English speakers and writers. Plus, once you have the infrastructure in place in one country, especially in operating systems, it's kind of hard to break in. They do well in their own country, though.'' Japan has its own boy wonder of software -- Kazuhiko Nishi, 34, who in the late 1970s and early 1980s was a Microsoft vice president and one of Gates's closest confidants. Nishi, now president of Ascii Corp., a Japanese software maker, agrees that the Japanese trail the U.S., especially in packaged software for business use. The reason: Japanese companies have built mainly upon technology from the U.S. and have been insufficiently creative in their own right. Says Nishi: ''Gradually, software companies in Japan are no longer following the U.S. lead but are coming out with unique products.'' He cites Nintendo's videogames as an example of how Japan can make software that sells in the U.S. Still, he says, ''I don't think we can surpass the U.S. soon.'' For all of Microsoft's accomplishments so far, just about everybody in the company believes its best days are still ahead. Gates has his brain trust of architects developing lots of new technologies. One group is working on handwriting recognition to compete with Go. Another is trying to build the ^ software foundation to make IBM-compatible PCs into multimedia machines that can manipulate photos, video, stereo sound, and animation as well as numbers and text. With them, computer users could make their own flashy productions and video presentations. Perhaps the most ambitious effort: an attempt to make software more modular, so professional programmers can create new software faster and nontechnical users can easily combine the parts of their favorite programs that they use most often. That way, for example, you could draw a picture in the middle of a typed letter without having to leave your word-processing program. Says Jeff Harbers, who heads the project: ''The way I see it, Microsoft's most important job in the world is to take software development from the age of blacksmithing to the age of machine tools.'' HOW WELL Microsoft can keep the magic coming will depend in part on its new president. Says Hallman: ''The big challenge for me will be how to deal with constant growth, because realistically there is no end in sight. We can't have a flat organization with all the business units on equal footing forever, and even with our electronic mail system, communications could become a big problem.'' Because the nature of Microsoft's business is changing, ''developing good technology is no longer the only critical path,'' Hallman adds. ''It is how computers affect the business organization that is critical. We have to learn to make our products be the tools that organizations use to change their business processes'' -- much as Microsoft's own financial database and electronic mail systems have enabled it to break up operations into tiny, product-specific business units. Gates insists that all the challenges only make his job more fascinating. ''I'd get bored if things just stayed the same,'' he says. Besides, deep down he's still a technology nut. With others tackling business and organizational issues full time, he still has leisure to do what he likes best -- talk techie with his architects and engineers. He tries to spend at least half his time doing that. Many evenings, long after the rest of the executive staff has left, Gates is dropping in on the business units to see what's really getting done. Says he: ''When I feel good at the end of the day, it's because I find a product group that is doing better than I expected, or because I contributed a good idea that ends up in a product.'' Will Gates ever step aside? Probably not any time soon. ''This company is my life,'' he says. Besides, Microsoft hasn't yet fulfilled the mission he and co-founder Allen set for the company 15 years ago: ''A computer on every desk and in every home, all running Microsoft software.'' Allen, who left the company in 1983 partly for health reasons, has just rejoined the Microsoft board. Together, he and the tenacious Gates are still chasing their ambitious goal. They may never get there, but it won't be for want of trying.

CHART: NOT AVAILABLE CREDIT: NO CREDIT CAPTION: ON THE RISE WITH CHAIRMAN BILL Since Microsoft went public in early 1986, revenues have jumped nearly 500%, profits are up almost 600%, and the company's stock has rocketed from the equivalent of just under $7 a share to over $70. Gates's holdings sat for a time around the $1 billion mark but now amount to more than $2.8 billion. REVENUES PROFITS STOCK PRICE BILL GATES'S MICROSOFT HOLDINGS

CHART: NOT AVAILABLE CREDIT: NO CREDIT CAPTION: INVESTOR'S SNAPSHOT MICROSOFT