Chapter One: Introduction Competing in the Age of the Internet
Occasionally, the world experiences a technological revolution that changes the way people live and interact. Ancient peoples experienced the emergence of agriculture, irrigation, and civil engineering. These developments led to the creation of cities and urban culture. Medieval peoples experienced the invention of the printing press. This technology gradually made books, magazines, newspapers, and the printed word -- information -- ubiquitous. Early modern Europeans championed the Industrial Revolution and new fields of science and engineering. New inventions, such as engines and factories, substituted mechanical devices and inanimate power for animal and human labor. Technology then progressed dramatically after the mid-19th century. The world has recently seen, in relatively rapid succession, the emergence of the telegraph, the telephone, radio, automobiles, airplanes, television, and the computer -- to name the better-known inventions in communications and transportation.
And now we have the Internet. The Internet is a network of computers, tens of millions of them, large and small, around the world. More accurately, it is a network of networks, based on a set of software technologies that drive computer hardware to send, receive, and locate "packets" of information traveling a worldwide electronic highway at lightning speed.1 The Internet has launched a technological revolution that is changing the way individuals, as well as organizations, live and interact. Imagine combining the power of the printing press (and most of the newspapers and magazines on earth) with the power and speed of the telegraph, telephone, radio, television, and computer. Then make this package easy to use and cheap enough for the mass market. You would then have the potential of the Internet in its most usable form, the World Wide Web (known as "the Web" for short).
We are not exaggerating when we say that the Internet and the World Wide Web, with the browser as its user interface, are revolutionizing mass communications, as well as mass networking technology. It is unlike anything we have seen before. The Internet has the potential to link easily and almost instantaneously every computing device with every database with every person who has access to a communications device (telephone, cable, satellite, etc.). As a consequence, the Internet is recasting the most traditional organizations, ranging from the U.S. Internal Revenue Service to your local grocery store. Tens of thousands of companies, both large and small, have created Web sites through which you can purchase goods and services or receive valuable (and not so valuable) information. This means that consumers can do common tasks on the Internet, such as ordering groceries or books and searching for stock prices. They can also do far more complex tasks, such as creating ideal travel itineraries, getting investment or medical advice, or holding a videoconference while sharing documents with people around the world. For anyone in the industrialized world, and for many people in developing countries, access to this great wealth of information and services is already available. The cost is usually the price of a personal computer (PC) or a cheaper device like a handheld computer, a TV set-top box, or the new network computer, as well as a local phone call and a charge of a few dollars per month.
The Age of the Internet
To understand the managerial and competitive implications of the Internet, we draw lessons from the experiences of Netscape Communications Corporation, the fastest-growing software company in history. After Netscape's explosion on the scene in 1994, it became synonymous with the Web. One year later, Netscape gained even greater -- if unwanted -- notoriety when Microsoft Corporation, the world's largest company dedicated to software production, challenged Netscape to a life-or-death battle. Together, these two companies are struggling to control key components of the Internet, including browsers, which provide a graphical user interface to the Web; servers, which are special software programs that run on powerful PCs or mainframes and deliver, or "serve," information (including pictures or sound) to the browsers; and portals, Web sites like Yahoo! and AOL.com (America Online) that aggregate information and become the jumping-off point for users surfing the Web.
It seems hard to believe that Netscape and the World Wide Web were not even on the horizon a decade ago, and the Internet was a little-known curiosity. The Internet began in the late 1960s as an arcane network connecting university and government computers. Scientists wanted to exchange data and electronic mail. Government officials wanted to be able to communicate if a nuclear war caused conventional communications technologies to collapse. The Internet remained the province of these small groups for 20 years. Then, in 1989, Tim Berners-Lee, a British researcher at the European Laboratory for Particle Physics (CERN) outside Geneva, created a system that would make it easier for scientists to use the Internet to share information. Berners-Lee defined the core elements of the Web -- a text formatting system (Hypertext Markup Language or HTML), a communications standard (Hypertext Transfer Protocol or HTTP), and an addressing scheme to locate Web sites (Uniform Resource Locators or URLs). Then he built a rudimentary browser.2 In 1993, a handful of students working for the National Center for Supercomputing Applications (NCSA) at the University of Illinois took Berners-Lee's invention, integrated graphics and multimedia features into the browser, and made it run on mass-market computing platforms, such as Windows and the Macintosh. The result was Mosaic, a wildly popular toy and information access tool. Most of the browsers available today, including Netscape's Navigator and Microsoft's Internet Explorer, have descended in some way from NCSA's Mosaic.
Mosaic launched a wave of innovation that led, in turn, to an ever-expanding technological alphabet soup. People working with the Internet have had to learn new concepts and new vocabularies almost daily. In addition to HTTP and HTML, two other early standards that defined how the Internet could send and receive information were FTP (File Transfer Protocol) and TCP/IP (Transmission Control Protocol/Internet Protocol). Many other standards quickly emerged for sending data and even video pictures and telephone conversations across the Internet. The proliferation of these technologies is testimony to the dynamism of the Web. More important evidence, however, is the explosion of Internet-based software and services in just a few short years. Utilizing the technologies of the Web, some companies have quickly grown to hundreds and even thousands of employees, hundreds of millions of dollars in revenues, and billions of dollars in market value. The "Internet future" has been unfolding so fast that managers in the industry tell us they cannot confidently predict exactly what products and features to build, what technologies to use, or what customers will buy more than six months to a year in advance. Nevertheless, many products that companies want to create, such as new operating systems, browsers, servers, or groupware applications such as electronic mail and electronic bulletin boards, take 18 months or more to design, build, and test.
Life was not like this before! In past decades, many companies extolled the virtues of long-term planning -- looking forward five to 10 years into the future. Compared to today, companies also took their time in product development. For example, Microsoft launched an operating system for IBM-compatible PCs, called MS-DOS, in 1981 and only made evolutionary changes in this technology until 1990, when it introduced Windows 3.0. Apple unveiled the Macintosh computer in 1984, and 14 years later, it had yet to deliver any breakthrough changes to its software technology.
Not only has the technology been moving fast, but the number of users has been growing geometrically each year. In 1993, the primary users of the Internet were scientists, professors, and engineers at university and government labs and a handful of corporations. By 1998, there were 130 million users from all walks of life.3 Web commerce also exploded from nothing in 1993 to $22 billion in 1998, with predictions of hundreds of billions of dollars early in the next century.4 This rapid expansion of the network is a classic example of what economists describe as "positive feedback loops," "increasing returns," and "network externalities."5 Behind the jargon, the dynamics are easy to follow. As more people and organizations connect to the Internet, more people and organizations create more tools and applications that make the Internet even more useful. And the more users, as well as tools and applications, there are, the more valuable connecting to the Internet becomes. As a result, more people start connecting, more tools and applications appear, and even more people sign on, ad infinitum. The technology community likes to describe this phenomenon as Metcalfe's Law, which states that the usefulness of a network, like the Internet, grows exponentially as the number of users grows.6 If a company could control a significant piece of the network or the technology, the potential returns would grow exponentially as well.
These competitive dynamics naturally lead managers to race for market share. Companies hope to grab the biggest share of customers, set the standard for the market, and reap huge benefits down the road. This psychology puts a huge premium on speed. This is especially true for firms that are using the Internet or racing to add to its capabilities. In the past, for example, AT&T might have taken a decade to design and build a new telecommunications switch, write the software to make it work, and deploy the new product widely to customers. Similarly, Microsoft and Lotus would spend two to three years just to produce new versions of their desktop applications. Today, Intel upgrades its microprocessors several times per year. Cisco unveils a new communications switch or router almost every month. PC manufacturers, like Dell and Compaq, use new processor technology to launch new computers every few months. And Netscape and Microsoft have been competing to deliver new versions of their browsers and servers as often as the market will absorb them.
Competing on Internet Time
The conventional wisdom about competition in the age of the Internet is that the business world has become incredibly fast and unpredictable, and we need to throw out the old rules of the game. We decided to test this hypothesis by looking in the place where the world was most likely to have changed -- the small constellation of companies that are building the new information infrastructure and hope to accelerate the pace of life for everyone else. After more than a year of intensive investigation, we are inclined to agree with some (but not all) of the hype. Some things really have changed because of the Internet, and some traditional forms of business practice have become much less useful than in the past.
For companies competing in the new information economy, the Internet is forcing managers and employees to experiment, invent, plan, and change their ideas constantly while they are trying to build complex new products and technologies. The Internet also requires companies to face the reality that competitive advantage can appear and disappear overnight. This is because the Internet makes it possible to organize your business in new ways, to offer new products and services, and to distribute those products and services to tens of millions of people almost instantaneously via telephone lines, cable TV networks, and wireless communications. It was the electronic distribution capability of the Internet that allowed Netscape to burst onto the scene in 1994 and, in only a few months, emerge as one of the most serious threats Microsoft had ever faced. This sudden rise to prominence of new companies can and will happen again.
We also found, however, that some of the strategic precepts of the pre-Internet world continue to ring true. Several core elements of competitive advantage -- vision, leadership, innovation, quality, barriers to entry, customer lock-in, switching costs, and partner relationships -- remain critical to the overall equation for creating a successful company, even in the most turbulent environments. The bewildering pace of the Internet may even put a premium on these old-fashioned virtues. In addition, while the Internet compels managers to speed up several activities, such as product development and product launches, at the same time, other activities, such as strategic planning processes, can operate on more "normal" time scales. Microsoft, for example, found that its customary three-year planning cycles worked just fine, as long as you can "pulse," in the words of Microsoft's president Steve Ballmer, and make quick adjustments.
The Internet may be stimulating a revolution in competitive dynamics and some business practices, but it has not revolutionized everything. Some of the new technologies associated with the Internet, such as the new Java programming language, are too immature to be the foundation for many companies, or even a full suite of products. The tyranny of the installed base is also very real. Despite all the hype about the forthcoming supremacy of new devices such as network computers and set-top boxes, PCs running Windows remain the primary access devices for the Internet. Assuming the imminent death of the installed base is a recipe for distraction.
We argue, therefore, that competing on Internet time is about more than just being fast. The apparent compression of time is only one dimension of life in and around the Internet. For us, competing on Internet time is about moving rapidly to new products and markets; becoming flexible in strategy, structure, and operations; and exploiting all points of leverage for competitive advantage. The Internet demands that firms identify emerging opportunities quickly and move with great speed to take advantage of them. Equally important, managers must be flexible enough to change direction, change their organization, and change their day-to-day operations. Finally, in an information world where too many competitive advantages can be fleeting and new entrants can easily challenge incumbents, companies must find sources of leverage that can endure, either by locking in customers or exploiting opponents' weaknesses in such a way that they cannot respond. In short, competing on Internet time requires quick movement, flexibility, and leverage vis-à-vis your competitors, an approach to competition that we define later in this chapter as "judo strategy."
Netscape and Its Battle with Microsoft
No two companies better capture the essence of competing on Internet time in a virtual judo match than Netscape and arch-rival Microsoft. Both have moved rapidly to new products and markets, built flexibility into their strategies and operations, and exploited leverage for competitive advantage. Netscape, which is the primary focus of this book, is a particularly powerful model because it has been a catalyst and a driver and could even become a casualty of the new Internet age. Netscape's decision to ship a new browser electronically over the Internet every few months, with "beta," or pilot, versions even more frequently in the early years, made the company an immediate symbol of competing on Internet time. While most established firms have been struggling to change the way they do business, Netscape started with a clean slate. Managers and engineers at Netscape deeply embedded the Internet into the fabric of the company: Netscape designed its human resource systems, strategic planning procedures, product development processes, and product distribution mechanisms with the Internet fully in mind. Netscape also had the advantage of being a recent start-up, with none of the historical baggage that has plagued other companies trying to be fast and flexible in how they operate and compete. As a result, the company was uniquely positioned to capitalize on the market shifts and opportunities created by Internet technology.
Netscape's early history reads like a fairy tale. The company was started in April 1994 by Jim Clark, the founder of Silicon Graphics, and Marc Andreessen, a recent college graduate who had headed up the Mosaic team.7 The seasoned entrepreneur and the untested computer geek hired half a dozen of Andreessen's former colleagues at the University of Illinois and set about creating software for the World Wide Web. Initially, Netscape's business model called for developing two sets of products -- the browser, which would catapult Netscape to fame, and Web servers, which would pay the company's bills.
Netscape Navigator, the company's browser, was a spectacular success. Less than two months after being released in December 1994, it captured more than 60 percent of the market. Navigator owed much of its popularity to the fact that users could get it for free over the Internet. In the public's mind, Netscape became synonymous with software that was innovative, fun, and free. But Netscape was doing more than just giving its products away. Corporate customers who wanted support were happy to pay, which allowed Netscape to generate $80 million in sales in its first full year (see Figure 1.1). The browser accounted for 60 percent of Netscape revenues, with server-based products making up most of the rest. By the end of 1995, Netscape was a fairy tale come true: Less than two years after its founding, the company was valued at $7 billion.
Netscape initially aimed its products at the World Wide Web, but soon became a pioneer in using Internet technology and protocols as the basis for business applications. In 1996, Netscape was an early mover in intranet software. These "internal Internet" systems use "open" (i.e., not controlled by any one company) Internet standards and browser technology to support applications such as electronic mail and information sharing behind a corporate "firewall." (The firewall is a software barrer that prevents outsiders or unauthorized users from gaining access to the information on the intranet.) In 1997, Netscape extended its push into extranets, which use the same underlying Internet communications technology to connect multiple businesses together over a secure channel. Both moves required that Netscape develop a broad portfolio of increasingly sophisticated products. They also transformed Netscape from a browser company into a growing enterprise software firm, whose sales primarily targeted corporate information systems managers. Unlike Netscape's first customers, the early adopters who loved getting their hands on the latest technology, Netscape's new corporate customers were generally more conservative, with much higher standards for product quality and reliability. The share of revenues represented by corporate sales climbed from 7 percent in 1995 to 41 percent in 1996 and 62 percent in 1997 (see Figure 1.2).
At the same time, Netscape's share of the browser market began a steady decline after peaking at close to 90 percent in early 1996. The primary cause of this decline was, in a word, Microsoft. For much of 1995, Microsoft remained preoccupied with the challenges of bringing out Windows 95. As far as the Internet was concerned, the company seemed to have buried its head in the sand. Most Microsoft employees paid little attention to the Internet, and some had not even heard of Netscape, browsers, or the Web. Not surprisingly, a Web browser was not part of the original Windows 95 product specification.8 Bill Gates, however, was not ready to be counted out of the game. By the end of the year, he had turned Microsoft around. On Pearl Harbor Day 1995, he announced that Microsoft was "hard core" about the Internet and planned to "embrace and extend" the Internet across all Microsoft's products. As an opening salvo, Gates promised that Microsoft's Web browser, Internet Explorer, and its Web server would be free to all -- not just to noncommercial users.
Netscape's stock promptly fell 28 percent. A classic battle was looming: Netscape and its 700 or so employees were cast as David against Gates's Goliath, which had more than 17,000 employees and almost $6 billion in sales in 1995. Netscape continued to grow at a spectacular rate from the end of 1995 through most of 1997, with 10 quarters of revenue growth averaging 41 percent and 10 quarters of growth in pretax earnings. But with Microsoft attacking aggressively on all fronts, the competition eventually took its toll. At the end of 1997, Netscape was forced to declare a fourth-quarter loss of $88 million, and its share of the browser market had dropped close to 50 percent by the middle of 1998.
Netscape has been trying to regroup since early 1998. The company's management mapped out a new strategic direction, while Microsoft was forced onto the defensive in response to a barrage of state and federal antitrust suits. Questions regarding both companies' futures remained in the autumn of 1998. Yet no matter what happens to Netscape in the future, it will always have the distinction of being one of the fastest-growing start-ups ever. In a little more than three years, the company reached an annual sales rate of more than $500 million. It took Microsoft almost 14 years to reach comparable revenues! In addition, Netscape's browser, Netscape Navigator, has become one of the most successful desktop computer applications in history (see Figure 1.3). After 18 months, Navigator had an installed base of more than 38 million users, making it the world's most popular PC application. By early 1998, users had downloaded more than 90 million copies of Navigator from Netscape's Web site (see Figure 1.4). Netscape's servers attracted less attention, as public interest focused on the drama of the "browser wars." However, Netscape established a solid position in this market as well. By early 1998, it was the leading supplier of Web servers to large U.S. corporations (see Figure 1.5).
We cannot predict, with any certainty, the eventual outcome of Netscape's battle with Microsoft, and we would not bet against Microsoft in any market in which the company takes a serious interest. Nonetheless, we believe there is much to learn from observing what Netscape has done and how it compares and competes with Microsoft. The lessons from Netscape go far beyond the challenges facing a start-up. As Netscape became big, so did its challenges. Many of Netscape's current problems resemble those of larger companies trying to cope with ultrafast-paced competition and very nimble competitors. Netscape executives also have been worrying about how to respond to intense competition from multibillion-dollar companies, how to manage thousands of employees who must adapt constantly to the Internet, how to maintain a huge installed base of products and customers, and how to solve the puzzle of turning the company's Web site into a money-making machine. In other words, Netscape managers have been worrying about the same types of problems plaguing many enterprises today. And, of course, Netscape has a special dilemma: the Microsoft problem. Microsoft is not only a multibillion-dollar giant with millions of existing customers. Microsoft is also as fast and as nimble as any company of any size. As Bill Gates told Time magazine, "I don't think you'd be interviewing me if we were any less nimble. You'd be writing our epitaph."9
Plan of the Book
This book extracts lessons from Netscape and its battle with Microsoft to help managers learn to compete more effectively on Internet time. We seek to learn from the many things both companies did right, as well as from their mistakes. We have organized the remainder of this book into four main chapters and a conclusion. Each of the following chapters focuses on principles that represent our observations and interpretations of key practices or concepts used at Netscape, and sometimes at Microsoft, to operate and compete (see Table 1.1). These principles do not necessarily represent "best practice" in entrepreneurship, strategic planning, project management, or software engineering. Nor did Netscape or Microsoft implement each of these principles perfectly in every instance. At the very least, however, they represent effective practices that should help many companies become faster and more flexible and exploit strategic leverage.
Chapter 2 explores the lessons of Netscape's start-up phase and the scaling up of a company on Internet time. We suggest that Netscape's early success came from a combination of great vision, experienced people, and a flexible, yet loosely controlled and decentralized organization. In addition, Netscape was the first company to take full advantage of the Internet, enlisting millions of users as a virtual testing organization, transforming its Web site into a virtual distribution channel, and exploiting the press as a virtual marketing organization. Netscape's spectacular success, however, also had its dark side. Netscape did a great job in tapping free external resources, but the company was much less successful in building corporate relationships. At times, Netscape managers seemed so taken by their own accomplishments that they concluded the company could do too many things itself. That arrogance hurt Netscape with potential allies, especially companies that made personal computers and software applications or provided news, financial information, and other kinds of Internet "content."
The next three chapters examine the strategic, technological, and operational dimensions of competing on Internet time. We suggest that firms can prosper in these types of fast-paced and unpredictable environments if they adopt judo strategy, which we are using to describe a philosophy of competition that emphasizes rapid movement, flexibility, and leverage. In chapter 3, we develop principles for competing with judo techniques, and in chapters 4 and 5, we focus on the technical and operational aspects of implementing these techniques.
On the strategic level, we suggest in chapter 3 that firms competing on Internet time must begin by developing a capability to move rapidly to new markets and uncontested ground. Particularly for new players, like Netscape, it is crucial to avoid head-to-head combat with industry giants, such as Microsoft and IBM. The second strategic principle is to be flexible, giving way to superior force when directly attacked. Netscape had to be willing to retreat when confronted with a direct assault that it could not win. The third principle, which is the heart of judo strategy, is to exploit leverage that uses the weight and strategy of your opponents against them. Microsoft has a deep commitment to a "proprietary" (the opposite of "open") software architecture, Windows, and to convincing customers to upgrade to the "latest and greatest" version of its operating system and applications. Netscape's judo strategy has been to turn Microsoft's commitments to Windows into a disadvantage by supporting open standards and all the platforms that Microsoft wants to eliminate. These include older Microsoft products (e.g., Windows 3.1) as well as UNIX and Apple operating systems. A problem with judo strategy, however, is that a company's vulnerability increases with its success. As Netscape got bigger, an agile Microsoft used a number of judo moves against Netscape. At the end of this book, we will answer whether Bill Gates or Jim Barksdale is the better judo master.
Chapter 4 analyzes Netscape's attempts to use product design to achieve strategic leverage as well as increase the potential flexibility and quickness of its engineering organization. Netscape's key differentiating skill vis-à-vis Microsoft was the ability to build products that worked on multiple operating system platforms, especially for the UNIX server market. Although cross-platform design had some penalties for productivity and product performance, it allowed Netscape to seek customers that Microsoft shunned. We discuss the specific techniques that Netscape evolved for cross-platform design as well as its experience -- more negative than positive -- with Java, a cross-platform programming language. We also examine other areas of design strategy, such as how Netscape (and Microsoft) tried to increase the technical and organizational flexibility of their development teams by raising the level of modularity in their products. Both Netscape and Microsoft also used the idea of "parallel development" to overlap design and engineering work in different projects, including the next and the next-plus-one versions of the same product. Parallel development facilitated rapid movement in the marketplace by shortening development cycles and minimizing the intervals between new product releases.
Chapter 5 focuses on Netscape's efforts to increase flexibility as well as innovation and speed in the software development process. We also make many comparisons to Microsoft in order to understand how Internet time may have affected software development techniques. In Netscape's first few years, for example, managers and engineers emphasized the rapid creation of new products and features. As customers evolved from leading-edge Internet users to more conservative corporations, however, the company had to adapt its priorities and engineering culture to place more emphasis on values such as product stability and customer support. Microsoft had to make a similar transition in earlier years but then had to speed up some new projects to compete in Internet markets. Netscape also enhanced its ability to accommodate design changes quickly during a project by following the "synchronize-and-stabilize" process, an approach that Microsoft pioneered in the late 1980s for PC software development. This process allows developers lots of freedom to innovate and experiment during a project but keeps their design changes synchronized through daily "builds" (working prototypes). Product teams then periodically stop in order to stabilize their changes and reevaluate the evolving product. To promote speed and efficiency in managing design changes, Netscape pursued another strategy, test automation, with some, but not complete, success. In addition, Netscape revolutionized the use of beta testing. Like Microsoft, it also adopted a variety of other strategies, such as extensive internal product usage, customer data analysis, and project postmortems to improve products and processes as rapidly as possible.
Chapter 6, the conclusion, explores the implications of judo techniques for competing on Internet time. To some extent, the fast pace of the Internet is slowing; we can see this in the lengthening intervals between new product releases for browsers and servers. Nonetheless, we believe that technologies like the Internet have permanently altered the nature of competition in certain markets and permanently accelerated the flow of information and new products and services around the world. As a result, managers in the age of the Internet must build the capabilities to change quickly all the time.
We begin this final chapter by summarizing the key strategic, organizational, and operational principles that worked together to facilitate movement, flexibility, and leverage. In the process, we distill lessons learned from both Netscape and Microsoft about how to compete effectively on Internet time. The importance of these principles goes well beyond understanding these two companies, however. We believe our principles suggest a more general approach for how companies can be faster, more flexible, and find new sources of leverage in the information economy. In addition, we suggest a number of lessons about what not to do when competing on Internet time. Both Netscape and Microsoft made mistakes as they battled for dominance of the Internet. We identify the most serious errors that managers should avoid. Finally, we end with some thoughts about the future of Netscape. In the wake of huge losses at the end of 1997, Netscape has indeed transformed itself. It is pioneering new markets and technologies, such as electronic commerce and application servers, as well as leveraging the browser to build traffic on the Netscape Web site. But the company's success is not completely under its control. Part of its future will depend on the U.S. Department of Justice and its success at limiting Microsoft's dominance of the software industry. Despite this uncertainty, we speculate about how Netscape is likely to fare in three central struggles -- the browser wars, the server wars, and the portal wars.
Copyright © 1998 by Michael A. Cusumano and David B. Yoffie