The Innovator’s Dilemma


June 2001

The Innovator’s Dilemma

CIDMIconNewsletter Reviewed by Ginny Redish, President, Redish & Associates, Inc.

The Innovator’s Dilemma is the most interesting and useful book on organizations that I have read since Crossing the Chasm by Geoffrey Moore. Where Moore explains why so many new companies have difficulty moving their products to a mass market, Christensen explains why so many old companies have difficulty dealing with new ideas that would mean they have to cross the chasm back in the other direction. (Christensen refers to Moore only once, near the end of the book, and does not explain most of his thesis with reference to Moore or the “chasm,” but that is in essence what he is talking about.)

Understanding the Dilemma

Christensen’s book is about organizational failure but not about all the reasons for failure that you would typically think of. As he says, “Companies stumble for many reasons, of course, among them bureaucracy, arrogance, tired executive blood, poor planning, short-term investment horizons, inadequate skills and resources, and just plain bad luck” (page ix).

But those are not the companies Christensen is talking about. Instead, this book is about “well-managed companies that have their competitive antennae up, listen astutely to their customers, invest aggressively in new technologies, and yet still lose market dominance” (page ix).

Who is he talking about?
Remember Wang and 8″ disks? Remember DEC (Digital Equipment Corporation) and the VAX minicomputer? Remember when Sears was the dominant clothing store in the US (rather than Target or Wal-Mart)? Christensen is talking about companies that were so dominant in their field that they were unable to successfully change when the world was changing around them.

He is not only talking about these companies losing their dominant position-and, in many cases, disappearing entirely. He is talking even more about how difficult it is for people inside a successful company who do see the signs of the future or have innovative ideas to get heard within the company.

What type of change is so difficult?
Companies do change. They do grow. They do bring out new products. But Christensen points out that what successful companies do well is to keep on a straight path of whatever made them successful.

He draws a distinction between what he calls “sustaining technologies” and “disruptive technologies.” A “sustaining technology” is a technological advance that helps a company do better at what it has been doing, that fulfills what major customers perceive as their needs. A “disruptive technology” is an innovative idea that is related to the company’s business but that means changing course, changing customers, taking risks again like the company probably did in its early days.

An example
Christensen has examples from many industries, but his main case study is disk drives-from 14″ disks to 8″ disks to 5¼” disks to 3½” disks. In each case, the companies with the then-dominant size kept making their disks better (faster, more capacity), moving in the direction that their big customers were pushing them. But meanwhile, the technology that they were continuously improving was becoming less and less relevant.

Here is how Christensen explains just one instance of the dilemma:

Until the mid-1970’s, 14-inch drives with removable packs of disks accounted for nearly all disk drive sales…Hard disk capacity…increased at 15% annually…and capacity increased at 22%…reaching beyond the mainframe market to the…supercomputer market.

Between 1978 and 1980, several entrant firms-Shugart Associates, Micropolis, Priam, and Quantum-developed smaller 8-inch drives with 10, 20, 30, and 40 MB capacity. These drives were of no interest to mainframe computer manufacturers, which at that time were demanding drives with 300 to 400 MB capacity. These 8-inch entrants therefore sold their disruptive drives into a new application-minicomputers…. Although initially the cost per megabyte of capacity of 8-inch drives was higher than that of 14-inch drives, these new customers

[minicomputer manufacturers like Wang, DEC, etc.] were willing to pay a premium for other attributes that were important to them-especially smaller size. (pages 16-17)

Christensen’s book explains the ironic reality that so many companies that were started as risk ventures find it almost impossible to risk again once they have become successful. Where are Shugart and Priam? Where are Wang and DEC? These upstarts in the 8″ and minicomputer markets repeated the history of their 14″ drive predecessors, spending their resources on researching and developing ever better versions of their proven technology as their market got smaller and smaller. (Micropolis and Quantum actually did make the transition-more on that later.)

Good management causes this dilemma
As Christensen points out, poor engineering or poor management does not cause the failures he is talking about. In fact, the success of engineering and management in these companies (in support of what made the company successful) is the very thing that (ironically) leads to eventual failure. Good management can be a root cause of the problem.

  • Good managers listen to their major current customers. Christensen says: “Customers effectively control the patterns of resource allocation in well-run companies” (page 99). (Ironically, of course, eventually those same customers leave the company for the technology that they earlier had no use for and effectively kept from being developed within the company.)
  • Good managers insist on a business case that shows markets and profits. The “disruptive technology” when it is new is by definition unproven and often in search of a market. Christensen says: “The ultimate use or applications for disruptive technologies are unknowable in advance. Failure is an intrinsic step toward success” (page 99). Initially, the new disruptive technology is likely to have worse performance than what the company now produces. Because it is cheaper, simpler, smaller, and more convenient, it usually has a smaller profit margin than what the company now produces. What middle manager is going to go to bat for something like that? It would be managerial suicide.

Is it hopeless to think that a successful company can spawn its own disruptive, successor technology? Certainly, in many cases, success in the new technology has come from a brash, young start-up. Overnight delivery did not come from within the postal service or the trucking companies. Intuit’s QuickBooks grabbed 70 percent of the small business accounting market because it was not developed under the close guidance of CPAs as earlier attempts at software for small businesses had been (page 175). We all know stories of entrepreneurs who left an established company because they could not get attention or resources for their innovative idea-even when they would have preferred to remain as loyal employees with the company’s backing-even when they would have preferred to remain as technical developers and not to have become entrepreneurs at all. (The innovator’s dilemma!)

Resolving the Dilemma

The first half of The Innovator’s Dilemma (Part One, Chapters 1 through 4) builds the framework to help us understand “why sound decisions by great managers can lead firms to failure” (page xiii). The second half (Part Two, Chapters 5 through 8) works to resolve the dilemma. (Chapters 9 and 10 wrap up the book with a first person case study of how Christensen might manage the disruptive technology of electric cars if he worked for a major automaker [Chapter 9] and a summary of the major points of the book [Chapter 10].)

Four principles for success
Christensen says that successful companies can foster their disruptive innovators and reinvent themselves into the new world that the innovative technology will eventually bring. Those that have succeeded have followed these four principles (paraphrased from Christensen, page 99 and Chapters 5 through 8):

  • Put the project in an independent, autonomous unit that is allowed to have its own value system, its own customers, its own budget, its own profit margin-and that is often geographically away from the main organization.
  • Put the project in an organization that is small enough to get excited about small opportunities and small wins.
  • Recognize that the first attempts are most likely to fail. Assume that it will take an iterative process of trial and error to find the right product and the right market. Do not believe forecasts. Do not gear up for huge production until you have been through a few rounds of experimentation.
  • Assume that the new technology will start out with a new market that values what makes this different from what your mainstream clients want now.

Put the project in its own organization
Another aspect of the irony of the innovator’s dilemma is that quite often the initial idea for the new disruptive technology comes from developers within the successful old technology company. But the idea gets stymied there. In fact, Christensen shows why it cannot succeed within the main organization of a well-managed company.

Christensen supports what is known as the “resource dependence theory” of management-that great companies are driven by customers and investors, not by their managers and executives. He says:

Organizations will survive and prosper only if their staffs and systems serve the needs of customers and investors by providing them with the products, services, and profit they require. Organizations that do not will ultimately die off, starved of the revenue they need to survive. (page 101)

But by definition, current customers do not want or need a disruptive new technology. The people who were buying minicomputers from DEC saw no reason to have PCs, when PCs were an untried idea-a gleam in an innovator’s mind.

Furthermore, Christensen points out, executives usually don’t even get to see or hear about all the great ideas their developers have. The ideas get filtered on the way up the chain. And even if the ideas get up there and get approved, many crucial decisions about resources are made day to day by mid-level managers who have to deal with priorities among competing approved ideas. Christensen points out that DEC actually tried four times to get into the PC market and failed. Why?

DEC launched all four forays from within the mainstream company… Even though executive-level decisions lay behind the move into the PC business, those who made the day-to-day resource allocation decisions in the company never saw the sense in investing the necessary money, time, and energy in low-margin products that their customers didn’t want. (page 110)

IBM succeeded because they started a whole new, separate, autonomous unit in a new geographic location to do PCs. They didn’t try to get their mainframe people to do it. HP succeeded in the inkjet market in the same way, letting their laserjet business keep doing its thing for its customers and starting a separate, autonomous unit in a different state to do inkjet technology.

Put the project in a small organization that values small wins
The value system of a company with a new technology has to be different from that of an established company. (Basically, Christensen is saying that a company must act and expect differently when it is in Moore’s early adopter stage than in his mainstream, majority stages.) Large, growth-oriented companies need continual large growth.

Christensen points out, for example, the difference between the experience of early Apple computers (when Apple was founded as a truly disruptive technology) and the experience of the Apple Newton (trying to develop a disruptive technology within a company that now had mainstream growth expectations). In the first two years of the Apple computer, the company sold 43,000 units and considered that a huge success. They went through several generations of Apples and the failure of the Lisa before they really found their market with the Macintosh. The Newton sold 140,000 units in its first two years!-almost four times as many as the original Apple. But that was considered a failure, and the project was shut down without the generations of trial and error that had been allowed for the earlier computer technology (pages 134-135).

Recognize that the first attempts are most likely to fail
“Markets that do not exist cannot be analyzed: Suppliers and customers must discover them together,” Christensen says (page 147). To be starting up rather than to be an ongoing business takes a very different managerial style and criteria for managerial success.

Assume the new technology needs a new market
The mistake that established firms make with a new idea, Christensen says, is to assume that they can perfect it for their traditional market. They keep it bottled up in the lab trying to make it good enough for their current customers, when what succeeds is being first to market for those who value the very differences that make the new idea “disruptive.” Managers for new technologies have to be in discover-and-learn mode, looking for users. And we could have a major role in helping find new markets. See box on page 72.

Two more examples
Remember Quantum and Micropolis, two companies that made it big with 8″ disk drives. Why didn’t they die off as the others did?

Quantum didn’t die off because it followed Christensen’s four principles. Quantum missed the 5¼” market entirely, but when a group of Quantum developers came up with the 3½” drive, Quantum financed them, sent them off as a separate company, and kept an 80 percent share. When the new company was successful, Quantum bought the other 20 percent and basically gave the old name to the now-successful new company (pages 104-105).

Micropolis didn’t die off because it didn’t follow the four principles (exception that proves the rule). After trying unsuccessfully to both continue with the sustaining technology (make better 8″ drives) and also build up the new technology (get into 3½” drives), the CEO of Micropolis realized he could not do both. So he scuttled the old business, gave up all his old customers, and put all his resources into the new business. In essence, he crossed the chasm back again and started over (page 106).

In Summary

The Innovator’s Dilemma is a fascinating look at how organizations grow, survive, and die. It is an insightful explanation of how new products and new ideas succeed or fail. It explains why the strategies that make managers extremely successful in one situation are the very strategies that make them fail in other situations.

The Innovator’s Dilemma won the 1997 Global Business Book Award as the best business book of the year. It’s worth reading, and it’s worth a place in your thinking right next to Crossing the Chasm.

About the Author