06 Sep 2004  Research & Ideas

The Innovator’s Battle Plan

Great firms can be undone by disruptors who analyze and exploit an incumbent’s strengths and motivations. From Clayton Christensen’s new book Seeing What’s Next.

 

When companies have the same capabilities and motivation, they care about the battle and have the necessary skills to fight it. Skills in execution make the difference here—and because other scholars have addressed these challenges quite capably, we do not focus on them in this book.8

The more interesting scenarios occur when there are asymmetries—important differences of motivation or skills. Asymmetries of motivation occur when one firm wants to do something that another firm specifically does not want to do. Asymmetries of skills occur when one firm's strength is another firm's weakness.

The section discusses three topics:

  1. How asymmetries power the process of disruption
  2. How to identify the company with the shield of asymmetric motivation and the sword of asymmetric skills on its side
  3. How to identify circumstances in which a high-potential disruptive development will prove disappointing, ending in either a brutal fight or incumbent co-option

How asymmetries propel disruptive entrants

Asymmetries allow disruptive attackers to enter a market, grow without incumbent interference, and mitigate the incumbent's response when it is finally motivated to counterattack. The result of asymmetric battles often is the seemingly sudden end of a great firm. From the incumbent's perspective, every action it takes is rational. But the outcome is devastating. Disruption is the strategy that creates and capitalizes on asymmetries of motivation and skills. It naturally follows a three-step process:

Step 1: Entrants enter behind a shield of asymmetric motivation; early incumbent response leads to cramming. Incumbents pass over what in retrospect turn out to be multibillion-dollar opportunities because attackers take advantage of asymmetries of motivation. When people say "flying beneath the radar," they really mean "taking advantage of asymmetries of motivation." Disruptive markets start among customers that appear to the incumbent to be either undesirable or nonexistent. The initial absolute size of a disruptive opportunity is generally too small to justify any substantial amount of investment or even management attention.

Asymmetric motivation shields companies from competitive response, because their potential challengers are just not interested in fighting. Even if they fight, their hearts aren't in it. The battle matters more to their opponents. Remember the old saying: "It is not the size of the dog in the fight. It is the size of the fight in the dog."

Gauging asymmetries is so essential to predicting outcomes because the availability of information about a threat or opportunity has little influence on who wins and who loses. What makes the difference is what a company does with that information. Incumbents usually see the same technologies that entrants do. Because of their processes and values, however, incumbents predictably "cram" the technology into the largest and most obvious market applications.

What causes cramming? Managers tend to adopt innovations in ways that make sense to their businesses. They know the markets their companies serve. Those markets appear large and measurable. The companies have established processes to help them penetrate that market. So the established competitors in a market naturally choose to bring all innovations—sustaining and disruptive—to their core markets where their best customers reside.9 For example, had Western Union purchased Bell's patents, we would predict that it would not have commercialized the technology until it had toiled away in its labs modifying the technology so it could be useful to Western Union's most profitable real-time long-distance data customers.

When an existing firm tries to insert a product or service with disruptive potential into its processes, what comes out the other end tends not to be disruptive. Instead of embracing the innovative product's inherently disruptive nature, the incumbent inevitably tries to morph the product to fit into its existing processes and values. It alters the innovation to enhance its appeal to core customers and fit within its operating model. The problem with cramming is that it changes the innovation in ways that obviate its inherent disruptive energy. It takes an innovation from a circumstance in which its unique features are valuable to a circumstance in which its unique features are a liability.

Cramming is like trying to stuff a square peg into a round hole. What signs indicate that cramming is occurring? When companies spend a lot of money fixing product deficiencies, they may be cramming. Large charges or expenses to integrate an acquisition are a good tip-off. Another sign is when companies must convince customers to change their behavior or put up with something they don't seem to want.

For example, Kodak first began to sense that digital imaging might pose a threat to its core business in the mid-1990s. It invested more than $2 billion in research and development. However, it framed the challenge as, "How do we make digital imaging good enough to serve as a viable replacement to silver halide film in our core market?" By seeking to create high-priced, performance-competitive digital products, Kodak missed much of the disruptive growth driven by inexpensive digital imaging. Kodak eventually established a strong market share after introducing a very low-cost camera, but only after spending $2 billion trying to maximize its cameras' performance.10

The result of asymmetric battles often is the seemingly sudden end of a great firm.

Cramming happens all the time. It is almost never successful. It costs a lot of money and usually ends with disappointing results. Cramming explains why so many disruptive innovations originate from within incumbents but are ultimately commercialized by separate organizations. When a firm develops the germ of a disruptive idea, it often falls prey to the temptation to cram that opportunity to its mainstream market. The momentum behind the widely practiced mantra "serve our customers" is often nearly impossible to overcome. The mainstream market tends to reject the innovation. Frustrated managers and engineers leave, form a new company, and discover a new market where the innovation has value.

Step 2: Entrants grow and improve; incumbents choose flight. As disruptive attackers follow their own sustaining trajectories, they make inroads into the low end of the market or begin pulling less demanding customers into a new context of use. What happens when the disruptive entrant begins to make inroads? A good way to visualize what incumbents can do when faced with a disruptive attack is to consider how humans respond to a perceived threat. Our body immediately reacts. We produce adrenaline. Our heart rate goes up. Our respiration rate goes up. Blood flow redirects from nonessential areas to critical areas. Our body is prepared for one of two actions: fight or flight.

Incumbents naturally choose flight. What looks highly attractive to the entrant continues to look relatively unattractive to the incumbent. The asymmetric motivation leads to incumbents naturally fleeing the low end. They cede that market to the entrant. AT&T, for example, initially ceded the lowest end of the existing long-distance market to MCI. Western Union clearly ceded the new local communications market to the Bell companies.

Remember, incumbents focus on delivering up-market sustaining innovations that allow them to earn premium prices by reaching undershot customers. They view flight as a positive development. When there are large groups of undershot customers in the higher tiers of a market, incumbents can beat a long and profitable up-market retreat. The customers incumbents leave behind tend to be the least attractive. Sales can go up (high price points replace low price points). Margins typically go way up. The incumbent stops worrying about disloyal, dissatisfied, low-paying (overshot) customers whom outsiders may term "distractions" and instead focuses on very loyal, highly satisfied, high-paying customers. In essence, incumbents can disassociate themselves from a business in exchange for a better one.

How can we identify when flight is in progress? Changes in customer or product mix can be a clear sign of flight, as are plans to discontinue low-end product lines or to stop servicing old versions of products. Companies tend to announce that they are "focusing on the core" or "seeking higher margin opportunities" when flight is in progress. Sometimes companies will launch diversification efforts to flee from an attacker.

Step 3: Entrants utilize the sword of asymmetric skills. What starts small gets big—too big for an incumbent to ignore. The number of undershot customers dwindles. But when incumbents become cornered, they face two problems. First, asymmetric motivation still stymies effective response. Even though the new opportunity may appear big, it typically requires a different business model. Even worse, incumbents are now at the mercy of asymmetric skills. Remember, disruptive innovations typically introduce new benefits to a market, usually centered on convenience, simplicity, customization, or affordability. As the entrant steadily solves unique problems, it builds the ability to do whatever is required to succeed in its context. When the incumbent has retreated into the highest tiers of its market and has to fight because there is no room for further retreat, it is at a competitive disadvantage. As the game changes to the one the disruptor plays best, it is very hard for the incumbents to develop new skills quickly.

By the time the Bell companies firmly established themselves, they developed unique competencies related to transmitting the human voice over relatively short distances. They established skills in acoustics, network management, customer service, and so on. Western Union had none of these skills. Its business did not need to solve these problems. It was on the wrong side of asymmetries. Western Union couldn't suddenly become a viable competitor after the telephone had been improving for twenty-five years. Similarly, Digital Equipment Corporation couldn't match the flexibility of the personal computer assemblers' processes, Sears couldn't match the inventory turns and low prices of the discount retailers, and so on.

It is important to note that almost thirty years elapsed from the introduction of the telephone before telephony operators began making serious inroads against Western Union. Similarly, wireless telephony existed for twenty-five years before it seriously began to erode the wireline business. Both innovations grew for a very long time in markets that were different from the incumbents' cores. Incumbent firms that take action when data shows a downturn in their core businesses take action too late. The only signal to take timely action is sound theory.

In disruptive circumstances, entrants win because they arc—often unknowingly—taking advantage of the asymmetries on their side. An incumbent's strengths are its weaknesses. Its values, which make sure it delivers excellent products to demanding customers, stop it from going after markets where ultimately its strongest competitors will forge their processes and values. The incumbent's processes, those that allow it to serve its customers well, are weaknesses when the game changes and new capabilities are necessary. Fleeing from the disruptive attacker feels good in the short term but further deprives the incumbent of the necessary skills to compete. The end can come swiftly and can appear stunning to the untrained eye. Typically, the best an incumbent can do is to belatedly acquire the winning firm and stave off ultimate destruction.

Seeing What's Next

In sustaining circumstances, incumbents win because they are the ones attacking with asymmetries on their side. Consider an entrant that tries to bring a radical sustaining innovation to an incumbent's best customers. There are no asymmetric motivations here. Incumbents are very motivated to go after this market. The incumbent's processes are relative strengths, not relative weaknesses. A new firm that attempts to introduce a radical sustaining innovation almost guarantees itself a long, bruising fight. Unless the new firms sell their advanced product quickly to the established leaders or have extremely patient, deep-pocketed investors, incumbents will spend more, produce better products and services, and ultimately force the entrants out of the market.11

Identifying the firm with the sword and the shield of asymmetries

The power of asymmetries is why the signals of change question is so important. Companies that develop different ways to target nonconsumers and overshot customers can create a new market or attack the lower tiers of a market almost free from interference from an incumbent that views the opportunities as unattractive. They have the potential to develop legitimately different skills and business models….

There are three factors that contribute to asymmetric motivation, all of which relate to firm values. The first factor relates to an opportunity's absolute size. An opportunity that looks interesting and large to a small firm might look uninteresting and small to a big firm. To the emerging telephony companies, the emerging local communications market looked boundless. Through the lenses of Western Union's values, the market looked inconsequential. Similarly, a meaningful growth opportunity for a Fortune 50 company will be very different than that of a start-up. The second factor relates to an opportunity's initial customer. Disruptive markets start among customers that appear to the incumbent to be either undesirable or nonexistent. Entrants are motivated to serve the very customers incumbents are motivated not to serve. The final factor relates to an opportunity's business model. Disruptive entrants use business models that do not fit the ways established firms make money. Gross margin per unit sold tends to be lower but turnover or asset utilization tends to be higher. Disruptive innovations tend to be off-the-shelf products, in which the customer turns either to a group of specialist firms or to themselves to provide postsales service. A company that has a business model based on long-term relationships and multi-year support agreements will have little interest in selling a product that obliterates those revenue streams.

How can you observe asymmetric motivation in action? When companies take completely different actions that make sense to both of them, it is a sign that there are asymmetries. When one firm calls an industry "unprofitable" while another firm calls that market "important," asymmetries are at work. For example, in the 1990s, Cisco raced to capture more of the networking equipment market while IBM sought to leave networking equipment to focus on higher-margin services. On the other hand, when an incumbent announces an entrant's emergent growth market is a strategic priority, it could indicate a lack of asymmetries.

Asymmetric skills act as a weapon a company can brandish to attack its opponents. Remember, a company's skills come largely from its processes. A process comes from repeatedly solving a particular class of problem. Processes are designed to get the same thing done, over and over—and as such they tend to be inflexible. Asymmetric skills arise when one firm, through repeatedly completing the same task, has developed a unique ability to do something that its competitor is uniquely unable to do.

How can you tell if combatants have asymmetric skills? Make a list of the tasks the company has repeatedly addressed, for which formal and informal processes have likely coalesced. Compare the list to the nature of tasks required to succeed in the disruptive market. If a company's processes facilitate its doing what it needs to get done in a market context, it has the requisite skills to take that market. Companies all have strengths and weaknesses. When one firm has strengths in markets in which another firm's capabilities are weaknesses, the firms have asymmetric skills.

Circumstances in which high-potential disruptive development will prove disappointing

There are certain situations in which—even though disruptive entrants seem to be on an unencumbered trajectory toward success—incumbents still emerge triumphant.

Understanding asymmetries allows us to identify two specific circumstances in which the disruptive process is likely to come off the rails. The first circumstance occurs when the industry context makes incumbent flight unpalatable. The second occurs when an entrant fails to develop a distinctive business model or to create unique skills in its early stages, making co-option the natural choice for the incumbent. In these circumstances, expect to see tough competitive battles after the aggressive growth phase.

Industry contexts in which flight is unpalatable

Fleeing up-market is only a natural choice for incumbents when there is an adequately sized, attractive market at the high end. This isn't always the case. Firms might not have the ability to reach the next tier of undershot customers. Or undershot customers might not exist. Or an incumbent's cost structure and business model sometimes prevent it from leaving low-end markets.

Think about the battle between AT&T and MCI. MCI grew without real interference from AT&T for a few years. But AT&T eventually had to respond to MCI. AT&T's business model depended on amortizing large fixed costs over a wide number of users. Flight to the high end—which typically entails focusing on smaller numbers of customers who are willing to pay high prices for the most advanced products and services—really wasn't a competitive option. Because AT&T cared about the customers MCI took from it, it ultimately decided to fight back. The resulting "long-distance wars" of the 1980s and 1990s cost both companies millions in advertising, with the "winner" capturing a bigger slice of an increasingly less lucrative market.

When firms can't choose to flee, we can expect them to fight. Entrants can no longer hide behind asymmetric motivation. And if this happens early enough, entrants don't have the time to refine their asymmetric skills. The result is often a bruising battle for market share. This concept helps explain the no-win situation facing many players in the aviation industry.

Lack of fully developed asymmetries make co-option a natural choice

Incumbents can't flee when they have no up-market to penetrate. When else are incumbents strongly motivated to respond to disruptive attackers? When an entrant makes money in similar ways or uses similar processes to incumbents. An entrant runs into trouble if it does not develop a business model that is unattractive to the incumbent, or if it does not hone unique skills that are matched to the disruptive business. It can create initial growth by taking advantage of the incumbent's disinterest in small, immeasurable markets. But once the entrant has proved to the world that the market exists, the incumbent can muster its internal resources to co-opt the innovation, unless the entrant has made that path unattractive. Natural incumbent motivation shifts from flight to fight. And in direct fights with comparable motivation to win, the incumbents have some real advantages to bring to the battle.

Asymmetric skills act as a weapon a company can brandish to attack its opponents.

What indicates an attempt at co-option? One sign is an effort by incumbents to develop products or services that mimic the entrant's offering. Acquiring and attempting to integrate one of the class of disruptive entrants is a clear signal that the incumbent is attempting co-option. When business model or process asymmetries don't exist, co-option becomes a viable incumbent response strategy.

Incumbents might still miss much of the growth created by the disruptive innovation, of course, depending on when they initiate following a co-option strategy and where they implement it. Growth-driven co-option needs to occur early and entails going after an entrant's core customers. In essence, an incumbent co-opts the disruption by augmenting its existing product and service offerings to appeal to new customers.12 When confronted with wireless technologies, for example, the incumbent phone companies chose growth-driven co-option. They developed wireless technology using internal resources and tried to market it to new customers.

Defensive co-option usually occurs later in a technology's development. Incumbents recognize that they have lost the game in the volume end of the market and do what they can to block incursions from below. For example, Oracle introduced its disruptive relational database in the minicomputer market in the 1980s.13 When IBM realized that Oracle had decisively won the game in that market, it introduced a relational database at the low end of its mainframe market, attempting to block Oracle's advance. Look to target customers and company announcements to see if companies are following growth-driven or defensive strategies. For instance, a company announcing that the disruptor's home market is a strategic priority is a clear signal of a growth-driven response.

Our theories suggest co-option efforts will be fruitless when success requires different skills or motivation. But when co-option is viable, incumbents can still ultimately master the technology and detain or swallow potential attackers. Existing telephone companies knew how to build and maintain networks. They had existing customers. They had a ton of cash. These were tough capabilities for wireless entrants to overcome. Selecting a business model that looked attractive to incumbents and providing a means for incumbents to learn put wireless entrants at a long-term disadvantage.

To sum up, incumbents tend to respond to potential disruptive incursions in one of three ways: they either cede a market, attempt growth-driven co-option, or attempt defensive co-option.

Footnotes:

8. See, for example: Larry Bossidy and Ram Charan, Execution: The Discipline of Getting Things Done (New York: Crown Business, 2002); James C. Collins and Jerry I. Porras, Built to Last: Successful Habits of Visionary Companies (New York: HarperBusiness, 1994); James C. Collins, Good to Great: Why Some Companies Make the Leap . . . and Others Don't (New York: HarperBusiness, 2001); William Joyce, Nitin Nohria, and Bruce Roberson, What (Really) Works (New York: HarperBusiness, 2003); Steven C. Wheelwright and Kim B. Clark, Revolutionizing Product Development: Quantum Leaps in Speed, Efficiency and Quality (New York: Free Press, 1992).

9. Of course, it is worse than this. Not only does an incumbent try to bring the innovation to its existing customers, it typically tries to bring it to its best existing customers. Ironically, these customers value the new attributes of the disruptive innovation the least.

10. For more on this, see Clark Gilbert and Joseph L. Bower, "Disruptive Change: When Trying Harder Is Part of the Problem," Harvard Business Review, May 2002, 94-101; and Clark Gilbert, "Can Competing Frames Co-exist? The Paradox of Threatened Response," working paper 02-056, Harvard Business School, Boston, 2002.

11. The best strategy for entrants in this situation is to try to briefly leap in front of incumbent companies and then sell themselves to incumbents. Many venture capital firms, whose partners have less and less tolerance for taking market risk, are adopting this strategy. They fund companies that are developing a sustaining innovation and seek to sell the start-up to an incumbent for a quick profit. Airbus is a notable exception that proves the rule here (discussed in more detail in Chapter 6). Airbus has pursued Boeing with a sustaining innovation strategy—but survival and success have required European governments to cover tens of billions in losses.

Entrants can undoubtedly move more quickly than established firms can. But once established firms marshal the resources to respond, entrants are in trouble. Other researchers have noted that some industries teeter on a razor's edge. These industries are so precarious that companies have no margin for error. In these industries, entrants can triumph by introducing better products, rapidly locking in customers, and quickly moving up the experience curve. Rebecca Henderson noted this phenomenon in her doctoral thesis, in which she studied the photolithographic alignment equipment industry. In four successive generations, new firms ended up toppling established market leaders with what we would classify as sustaining innovations. Our belief is that these industry circumstances do exist but are quite rare. See Rebecca M. Henderson and Kim B. Clark, "Architectural Innovation: The Reconfiguration of Existing Systems and the Failure of Established Firms," Administrative Science Quarterly 35 (1990): 9-30.

12. Again, remember that we mean incumbents to be firms already entrenched in a market context. We would classify an existing firm launching a disruptive attack on a new market as an entrant.

13. Hierarchical database software was designed for use on mainframe computers in the late 1960s and 1970s. Data in hierarchical databases is organized in a "tree" format, like a classic organization chart. These databases required expert programmers but could perform specific searches very quickly and used the mainframe's resources efficiently. The databases were ideal for demanding customers who needed heavy-duty transaction processing, such as financial institutions. The leaders in the industry were companies such as IBM and Cullinet, which in 1978 became the first U.S. software company to go public. In the early 1970s, a researcher at IBM came up with the theory behind a new form of database, called a relational database. All pieces of data in a relational database are interrelated. The initial relational databases were slow and used a lot of computer resources, but allowed users to relatively quickly and easily run customized queries without having expert knowledge about the underlying database structure. This technology initially had little value to mainstream customers because its functionality was not good enough. With their relative simplicity and flexibility, relational database were clearly more convenient than hierarchical databases.

About the authors

Clayton M. Christensen is a professor at Harvard Business School.

Scott D. Anthony is a partner at Innosight LLC.

Erik A. Roth is a consultant in McKinsey & Company's Boston office

Reproduced by permission of Harvard Business School Press. Excerpt from Seeing What's Next: Using the Theories of Innovation to Predict Industry Change. Copyright 2004 Harvard Business School Publishing Corporation; All rights reserved.

[ Buy this book ]