Sign In

Communications of the ACM

Viewpoints

Technology Strategy and Management: The Legacy of Bill Gates


View as: Print Mobile App ACM Digital Library Full Text (PDF) In the Digital Edition Share: Send by email Share on reddit Share on StumbleUpon Share on Hacker News Share on Tweeter Share on Facebook

Half a year has now passed since Bill Gates retired from his day-to-day position at Microsoft last June. Gates remains chairman of the board and still spends about one day a week at the company. But the media marked his departure as a major event, and rightly so. Gates built and led, for more than 30 years, one of the most successful technology companies in history, and Microsoft's products touch nearly all of our lives every day. At the same time, Gates lives with the reputation that he was simply "lucky." Gates and Paul Allen founded Microsoft in 1975 to make programming languages for a new PC that had yet to come to market. Gates later got the IBM contract to produce DOS in 19801981 after first turning it down because he did not make operating systems. At least part of his legacy stems from this good fortune, especially since IBM was not clever enough to control the rights to the operating system (nor the microprocessor design, which Intel controlled) that Gates then sold to clone producers. Another view is that Gates was primarily a talented "hacker," with little or no management skills, and that Microsoft succeeded because of its growing market power stemming from DOS and then the Windows and Office monopolies.a

There is much truth to the argument that Gates and Microsoft were in the right place at the right time, but that fact does not explain why they were able to exploit the opportunities that appeared. In 1995, Richard Selby and I wrote in Microsoft Secrets: "Bill Gates of Microsoft may be the shrewdest entrepreneur and the most underrated manager in American industry today."b I still stand by this judgment, but now is a good time to reflect on what might be the ultimate legacy of Bill Gates.

On the positive side, Gates correctly foresaw the future when, in 1975, he dropped out of Harvard to co-found Microsoft based on the belief that every desktop would one day have a PC on it. He wanted those machines to run Microsoft software and he played a huge role in making this vision come true. Second, Gates recognized that hardware prices were decreasing and that most of the future value from computing would be in software. At the same time, he argued that users should pay for software so that programming could become a livelihood and software companies would have the money to invest in product development. The free software movement and the shift to services-driven business models are complicating the software business today but this strategy served Microsoft remarkably well for three decades.

Third, Gates correctly recognized the potential of the software products business, in contrast to the custom software, services, and maintenance business of the mainframe world. He realized that commonly used software programs could be standardized, packaged, and sold cheaply, like any other commodity. He was not the first to see this but he exploited the potential of shrink-wrapped software better than anyone else. Fourth, Gates early on understood the power of creating products that could become "platforms" (see my September 2008 column comparing Microsoft and Apple). Windows and Office continue to generate so much revenue because they are at the center of enormous ecosystems of users and complementors. Microsoft designs products that are more open than Apple and has worked much harder to cultivate partners. Fifth, as Selby and I also wrote in 1995, Gates stands out for how he, and the people he hired, uniquely combined an understanding of "the technology and the business." Simply being a great programmer did not land you a job at Microsoft. You had to understand how to make money from software as well as how to work in a team. Sixth, Gates led his company through at least two major technological transitions, either of which might have done in a less able leader and organization: from character-based to graphical computing, and from the desktop to the Internet. He also directed Microsoft's diversification from programming languages to desktop operating systems, applications, enterprise software, video games, and the MSN online business, among others.c

Finally, Gates has distinguished himself as a truly charismatic leader of a unique kind of companya software companywhich defies conventional rules of productivity and operations given that the marginal cost of reproducing a software product is essentially zero. Gates never had formal management training and displayed many rough edges in his first decade or two on the job. But, make no mistake about it. Gates learned how to manage, delegate, and lead thousands of extremely smart, talented people. He continues to set the standard for other high-technology entrepreneurs-turned-CEOs.

On the negative side, first, part of the legacy must be that Gates himself was never very good at anticipating trends in the software business. His great strength was the ability to identify movements relatively early and then mobilize resources to exploit those trends. But he did not anticipate. This is true of graphical computing (led by Apple), Internet computing (led by Netscape), Internet search and other services (led by Yahoo, Google, and others), multi-platform computing (led by Sun Microsystems), open source computing (led by the Linux community and best exploited by Red Hat and IBM), the transition from products to services in the computer industry (foreseen best by IBM), software as a service (pioneered by Salesforce.com), or cloud computing (led by Amazon and Google). Second, Gates took too long to adjust to the position of power Microsoft attained and never quite understood the rules of competition differ for a firm with a monopoly market share (in this case, two monopolies). He testified reluctantly and weakly in the anti-trust trial. He never realized it was unnecessary as well as illegal for Microsoft to cross the antitrust line to battle competitors such as Netscape.d

Third, Gates let the company "get away" from him when it should not have, and he never demonstrated a "second wind." His early principles of how to manage a software company focused on hiring only top-notch people who understood both the technology and the business, and keeping development teams and products small in size. These principles started slipping in the mid-1990s. Microsoft reacted in a panic to Netscape and threw hundreds of programmers at building Internet Explorer and then dispatched thousands more to rebuild Windows and Office. Moreover, Gates was still chief software architect when Windows grew too large and complex in the early 2000s, tying too much functionality together to make every new feature an "integral part of Windows." Gates also kept the company tied to Windows when he could have decided that it was a platform company, whether the platform was Java or Linux or the Internet, and not just a Windows company. Microsoft has been recovering from the Windows quagmire and it has a new partnership with Novell to support Linux. But other managers have pushed these changes.

Shareholders should be disappointed that Gates grew weary of being targeted for his success during the antitrust trial and started withdrawing from his responsibilities then, first relinquishing the CEO title to Steve Ballmer in 2000. While he took on the title of chief software architect, he spent increasing amounts of time and money on philanthropic ventures, as well as raising his new family. This is a natural and highly laudable transition for a person at his pinnacle to make. But we know from the case of Steve Jobs and Apple that it is not impossible for CEOs, even with families and outside interests, to reinvent themselves and reenergize the companies they lead. Steve Jobs has taken Apple to its greatest heights only in the last couple of years, with the iPod and the iPhone. Though Apple may never match Microsoft in revenues or profitability, Jobs and Apple clearly have achieved a thought leadership position in computing, digital services, and multimedia. Apple shareholders have benefitted greatly in the past few years while Microsoft's stock has languished or fallen. Moreover, Gates is departing just when Microsoft's business model of packaged software products is undergoing its most serious challenge since the Internet from free software and remotely delivered software as a service and "cloud computing," and even a new browser (or Web operating system?) from Google.

Perhaps it is unfair to end a discussion of Gates' legacy by describing what he was not or was not trying to do. What he did is more than enough for any career. No individual did more to grow the PC software business by bringing inexpensive, powerful software products to the masses. This achievement should be Gates' most enduring legacy. He will always be one of the greatest (and richest) entrepreneurs and managers ever produced by any country, and he is no longer underrated.

Back to Top

Author

Michael Cusumano (cusumano@mit.edu) is Sloan Management Review Distinguished Professor of Management and Engineering Systems at the MIT Sloan School of Management and School of Engineering in Cambridge, MA.

Back to Top

Footnotes

a. The best account of Gates and Microsoft through the early 1990s is Stephen Manes and Paul Andrews, Gates: How Microsoft's Mogul Reinvented an Industryand Made Himself the Richest Man in America. Doubleday, New York, 1993.

b. M.A. Cusumano and R.W. Selby, Microsoft Secrets. Free Press/Simon and Schuster, New York, 1995, 23.

c. My thanks to David Yoffie for reminding me of this last point.

d. See M.A. Cusumano and D.B. Yoffie. Competing on Internet Time: Lessons from Netscape and Its Battle with Microsoft. Free Press/Simon and Schuster, New York, 1998. Also, M.A. Cusumano, That's some fine mess you've made, Mr. Gates. Wall Street Journal (Apr. 5, 2000), A26.

DOI: http://doi.acm.org/10.1145/1435417.1435429


©2009 ACM  0001-0782/09/0100  $5.00

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.

The Digital Library is published by the Association for Computing Machinery. Copyright © 2009 ACM, Inc.


 

No entries found