Editor’s note: Almost from the moment I met Mike Ruettgers, I liked him. This was in the early 1990s, when Ruettgers had just become chief executive of EMC Corp. (www.emc.com), which made data-storage equipment, and I was a reporter covering the computer beat for the Wall Street Journal. I had met plenty of executives who hoped to have the Journal write that their company had made the biggest breakthrough since Prometheus, so I braced for the usual sales pitch. Instead, Ruettgers was matter-of-fact. He liked his chances, but he understood if others didn’t.

It is true that EMC’s stock has taken a beating this year, as technology spending has fallen and as some of the froth has come off technology stocks. But what Wiersema describes as the core of EMC’s success—an obsession with customer service—still keeps EMC well ahead of its powerful competitors. —Paul Carroll

Imagine that your business is utterly dependent on an uninterrupted flow of data from your computer systems. Although everything appears to be running smoothly, you get a call from someone at the company whose data-storage devices you use, telling you that something is about to break down. She arrives the next day, replaces a couple of logic boards, gives the unit a pat, and leaves. Disaster has been averted. You never receive a bill for the call. It is on the house.

At EMC, that has been the standard way of doing business for more than a decade now, based on state-of-the-art monitoring technology and, even more, on a common-sense mantra: Listen to customers; discuss their problems; hold their hands; and help them find solutions, big or small. EMC manages customer service as an investment center. It doesn’t try to earn a profit on service, as so many companies do in the information-technology sector.

The reward for EMC, founded in 1979 as a maker of add-on memory boards, is a lock on a business whose growth is exploding. Demand for memory storage is rising 86% a year, and, if EMC is correct, the market for storage will leap to $100 billion over the next five years—twice the size of the market for servers. EMC, which generated $8.87 billion in revenue last year, predicts that world demand for storage will rocket to 10,000 petabytes, or 50 times today’s total capacity. The fiber-optic infrastructure will have grown a thousandfold to 20 billion miles. Data-compression technology will multiply transmission and storage capacity another thousandfold. Any person could generate a trillion bytes of stored information—medical records, school data, photos, video, whatever—and organize and index it for rapid retrieval.

Until recently, few people paid much attention to data storage. Since the early days of computers, the details of where and how all those billions of 1s and 0s were stored and retrieved has been the boring side of the business. The glamorous side was the computers themselves—mainframes, personal computers, laptops, and then network servers—and the software that kept them running. These days, EMC is a bit more familiar. Customers tend to sit up and take notice when they learn that two-thirds of the world’s critical data is stored on EMC products. When it comes to storage systems, nearly all customers are streamliners. That is, they don’t really want storage units; they want what the machines can do for them, and they want it with as little hassle as possible. Streamliner customers want their suppliers to be around five to 10 years from now. They require reliability and responsiveness: Nearly all the systems must work 24 hours a day, seven days a week, 365 days a year. A crash is intolerable.

“Only 10% of our job is fixing problems,” says Joe Walton, EMC’s senior vice president in charge of global customer service. “The other 90% is predicting, pre-empting, [and] preventing problems, so that customers can get on with their business.”

EMC’s engineers follow a guilty-until-proven-innocent policy in dealing with problems. Whether a glitch originated in an EMC product or another part of the customer’s system, EMC will keep on trying to fix it until it is clear somebody else was to blame. That approach surely plays to the streamliner’s urge to make a purchase and have no hassle about it, but it also helps EMC design better boxes. That saves a lot of money. As Executive Chairman Mike Ruettgers told me, “We believe the adage that if you fix a problem in design, it takes one unit of effort. If you fix it in manufacturing, it takes 10 units. If you have to fix it in the field, it’s 100 units.”

EMC collaborates with the major hardware and software companies to build and install complete systems. When a customer has to change its system, EMC engineers will preview the change and then will test any questionable features in EMC’s laboratories, using equipment that matches the customer’s system. If anything goes wrong after that, EMC will fix the problem at no charge.

What is more, for the first two years after a customer buys a storage box, service is free. Sacrificing this flowing profit stream is only tactical. EMC’s obsession with performance and reliability makes its streamliner customers willing to pay considerably more than rock-bottom prices. The service philosophy has paid off in loyalty, too: 80% of sales are to previous customers, and EMC’s customer retention rate is 99%.

One vital element in EMC’s success is the company’s “call home” support centers, which can monitor EMC units anywhere in the world and automatically report anything that goes wrong or is about to go wrong. The problem almost always is straightened out before the customer experiences any trouble.

Another benefit: The centers are closely linked to the engineering and research and development operations, providing EMC engineers with an intimate connection to their customers that helps them keep improving their products and maintain their leadership position in the industry.

When Ruettgers arrived as an executive vice president in 1988, he gave his senior managers airsickness bags, saying, “The quality of our products makes me want to puke.” Today, Ruettgers claims EMC’s customer satisfaction rate of 99.4% is by far the highest in the industry, reflecting its continuing investment in quality, its engineers’ close involvement with the customers, the reliability of its products, and the company’s expertise in product research and development.

To maintain close relations with its streamliner clients, EMC invites customers’ engineers and designers to meet their EMC counterparts in separate technology councils. Every day, an average of 25 client companies send groups of two to 40 people for customized briefings on industry trends and possible solutions to their problems at special briefing centers at EMC’s three regional headquarters in Massachusetts, Ireland, and Japan. EMC also tailors its sales and support staff to its customers, with one style for the relatively buttoned-up Global 2,000 customers, and a younger, hipper approach for start-up businesses. EMC staff members on the Global 2,000 beat tend to work from 9 a.m. to 7 p.m. and may even wear suits. Those working with start-up businesses dress and talk accordingly, and work their shifts within a window that runs from 10 a.m. to 10 p.m.

EMC also has proved itself a master at keeping up with a market that changes at breakneck speed. The company has gone through five major business reinventions and four product generations in just 10 years. Now, it is accelerating its evolution to what its managers call TNT: The Next Thing. These people are nothing less than serial disrupters of their own business model and technology—yet, somehow, they manage to do this so seamlessly that their streamliner customers barely notice.

The first revolution came in 1989, when EMC took on an arrogant IBM in going after the data-storage market for large computers. Big Blue was terrible at listening to its customers. What is more, it was focused on a complex, expensive form of storage, for which it had long set the technical standards and controlled the market. EMC already was reforming its culture to center on customers, and one of its bright ideas was a revolutionary approach to storage. Instead of trying to compete in the ever-more-complex form of memory that IBM dominated, EMC’s engineers assembled a lot of small, low-priced disks and linked them into what was called a redundant array of inexpensive disks (RAID). At first, this approach wasn’t as reliable as existing devices, but it offered a big advantage in price and speed and took up less space. EMC overcame initial skepticism by offering the devices on a trial basis, and, soon, many trials turned into purchases. Within three years, it had become the leader in market share for mainframe storage.

By that time, customers were shifting away from mainframes to decentralized computer networks, with servers linking dozens of outposts. The server makers, like the mainframe producers before them, left an opening for EMC. Server makers treated storage as an afterthought and, not focusing on what customers really wanted, offered devices that could be hooked up only to servers made by the same company. EMC captured the storage market by offering devices in 1992 that could be hooked up to servers from any manufacturer.

In 1995, it reinvented its business again, moving to enterprise storage—creating one box to handle all the storage needs of a company’s computers, whether mainframes, servers, or some other type. The change helped customers share and manage data better. By cutting the link between data and a particular computer, the change also made it easier for companies to upgrade or replace computers.

A few years later, EMC went through yet another transformation by getting seriously into software and services to make its boxes more independent and able to deal with a still greater variety of computers. With the new software, customers can set up duplicate storage systems, recording everything simultaneously in two locations to ensure the system will never be down for more than a second or two. This was a major advance because, at the time, computer-reliant companies that needed to guard against failure had to make backup tapes of data at frequent intervals and store them in distant warehouses, to protect against fire or flood. When something did go wrong, the companies resorted to what was facetiously called CTAM—the Chevy truck access method. Workers at the warehouse stacked tapes on a truck, drove to the site of the problem, and made copies. The process took at least two days.

EMC moved rapidly into the emerging market for storage networks, in which a large number of servers are connected to a large number of storage units. The latest reinvention is still happening: The market is moving from enterprise storage to a whole new “info-structure” to deal with what EMC calls the “content big bang”—an Internet-based information explosion over the next five years that will dwarf all increases in information storage.

Each of EMC’s serial disruptions required foresight and the commitment to enormous investments in research and development: $5 billion over the past 10 years, with another $2.5 billion planned for the next two years. What stands out most is that all of these innovations and changes were somehow implemented without major interruptions for the customer. To achieve seamlessness, EMC has had to develop several technologies concurrently; start working on the next breakthrough before the current one even has reached general availability; make its own products obsolete in their prime; and preserve its customers’ investment by offering software that works across generations.

All this would seem incredible if the incredible hadn’t become commonplace. Some 40 years ago, a terabyte of data would have required a storage facility the size of Argentina. Today, EMC stores 19 terabytes in a box the size of a desk. As the cost of storage has been driven down, it becomes easy and affordable to keep nearly everything anyone might want. For every price drop of 1%, EMC says, demand for storage rises 4%.

Ruettgers, who takes nothing for granted, knows that there is no guarantee of continued success, and he promises “continuous reinvention and investment, seamless product transitions, and superior execution on a global scale.” In the old days, he recalls, EMC could afford to make only one bet at a time. Now, it is working simultaneously on two or three promising technologies that could revolutionize the storage world again.

From The New Market Leaders by Fred Wiersema. Copyright ©2001 by Fred Wiersema. Reprinted by arrangement with the Free Press, a division of Simon & Schuster Inc.