The Innovator's Dilemma: When New Technologies Cause Great Firms to Fail

This chapter focuses on the theory of resource dependence, which states that customers exercise a certain level of control over a company’s allocation priorities, limiting their choices as a matter of survival. Companies are inclined to sustainably provide what their customers need so that they can continue to access the resource base that allows them to operate. This dynamic implies that managers have less influence on company decisions than customers do.

While many managers view this theory as controversial, Christensen’s research supports its validity, which raises the question of how managers can actually leverage disruptive technologies in spite of market demand. Rather than fight against the organization, Christensen suggests the alternative of harnessing the technology through an independent organization that meets the needs of emerging customers.

Innovation and resource allocation are directly tied together. When resources are invested in new product development, it is because the company has identified the product as being relevant to customer needs. Middle managers weed out irrelevant proposals, and senior managers select the most profitable innovation from a shortlist of relevant proposals. As middle managers prioritize projects in the interest of customer need and profitability, they cede more control to the customers than to executives.

Christensen posits that disk drive companies have used resource dependence and customer control to their advantage. For example, Control Data Corporation isolated its 5.25-inch production team from company headquarters. Though the company fell from its original position, this separation allowed it to survive when 5.25-inch disks became the mainstream product. Rather than affirming that managers are powerless against customer influence, Christensen argues that managers can sidestep the issue entirely by entrenching independent organizations in the value networks where the disruptive products will find value.

Christensen offers a few more examples to demonstrate this principle in context of the computing, retail, and printer industries. First, he focuses on Digital Equipment Corporation (DEC), who tried to breach the personal computer value network directly. Forced to plan around two separate cost structures, DEC managers couldn’t justify investing resources in a product that reaped low margins. IBM, on the other hand, devoted an independent subsidiary to entering the same market and was able to effectively build a new customer base and reap relatively high margins.

Next, Christensen shows how established retail firms Kresge and Dayton Hudson successfully transitioned from traditional department and variety store retail to discount retail through the creation of their autonomous subsidiaries. Finally, Christensen discusses how printer manufacturer Hewlett-Packard developed an autonomous division to leverage the disruptive ink-jet technology, which competes against the company’s own laser-jet unit. (At the time of the book’s writing, ink-jet printers were still just emerging.) Christensen predicts that although the laser jet is a superior product, it is only a matter of time until the ink-jet printer effectively obliterates the laser-jet division.

The book turns to the issue of organizational size, pointing out that large companies are compelled to grow their revenue in order to grow their organization. The small markets where disruptive technologies thrive therefore seem less viable to pursue, unless, as Christensen proposes, the organization were to embed their projects in smaller organizations that match the size of the market.

Christensen explains that many organizations are unsure whether to lead innovation development or give up first-mover advantages in order to avoid risks. Neither approach is definitively more correct than the other, but certain conditions can tell managers which option is best to pursue, as shown in the example of the disk drive industry. When established disk drive firms introduced the thin-film read/write head to their products, the pioneers failed to gain any substantial advantage over entrant competitors, whose later products would far exceed the pioneers’ products in performance. In this respect, pioneers do not gain any advantage when a sustaining technology is involved.

On the other hand, when pioneers are developing disruptive technologies, they gain a wide lead on competitors who follow them into the value network. Comparing the 83 American companies that entered the disk drive business from 1976 to 1993, Christensen finds that nearly 40% of firms who achieved substantial annual revenues with disruptive innovations generated a combined total revenue of $62 billion. This amount dwarfs the six percent of successful firms who achieved the same annual revenues in established markets, where they produced only $3.3 billion in total. Christensen concludes that the latter groups chose the risk of competing with entrenched firms over the risk of exploring an emerging market.

As Christensen has already discussed, however, it isn’t always easy for established firms to make the jump to new markets. The further they are from the size of the companies that are characteristic to new markets, the harder it is for established firms to justify the need to enter the competition. A company’s growth rate influences its stock price, and a strong stock price enables access to more capital. On the other hand, a company that has already achieved considerable growth will find it increasingly difficult to sustain further growth. Thus, it makes little sense for a large company to transition into an emerging market whose total value does not meet their growth requirements. To tackle the dilemma of harnessing disruptive change as a large company, Christensen tests three possible approaches against case studies in the computer, disk drive, and motor controls industries.

The first approach involves accelerating the emerging market’s growth rate so that it fits the growth requirements of the large company. For example, Apple Computer, which was already a leading personal computer firm by the early 1990s, saw the opportunity to enter the personal digital assistant business and boost their growth. The aggressive commercialization of their product, the Newton, was seen as a failure, despite the fact that it sold more units than the company’s flagship products of a decade earlier. Apple had created the market for personal digital assistants, but it rushed the discovery process for customers, leaving many of them unsure of the Newton’s value. At that time, the market was too small to contribute to Apple’s growth, and Apple lost out on their significant investment.

The second approach involves entering when the market has already emerged and grown to fit the company’s growth requirements. Christensen argues that firms that adopt this approach will have to compete with the entrant firms who have developed advantageous capabilities while simultaneously creating the market. Priam Corporation experienced this when trying to enter the 5.25-inch drive market, which Seagate Technology had spent three years creating. In that time, Seagate formed the capability to develop products on a one-year cycle, which outpaced Priam’s capability of developing products on a two-year cycle. Similarly, Seagate found itself outmatched by Conner Peripherals, which pioneered the 3.5-inch drive market with portable computer manufacturers. By moving first into the market, Conner developed the capability to design its products according to the needs of its biggest customers. By contrast, Seagate had waited too long for the market to fit its growth needs.

The third approach involves tapping into small organizations to commercialize disruptive products. Since small market opportunities don’t fit the managerial priorities of large companies, it makes sense to create a spinoff organization or acquire a relevant and appropriately sized company to engage with the target market. From the perspective of a smaller organization, managers and employees alike will view the opportunities as being relevant to their growth, thus resolving the issue. Christensen recalls the Control Data Corporation from Chapter 5 as an example of one company that leveraged a spinoff unit to develop the 5.25-inch drive. For acquisitions, he points to the Allen Bradley Company, a leading motor controls firm that acquired several firms that specialized in disruptive electronic control components. In doing so, this company maintained a lead over the entrant firms who created the value network with factory automation companies, and it outperformed its mainstream competitors, who struggled to integrate the disruptive innovation into their motor controls business. Johnson & Johnson applies a similar strategy of aggressive acquisition and subsidization, which ensures its lead as new technologies emerge.

Because disruptive markets cannot be predicted or known until they exist, Christensen advises managers to devise learning strategies to better understand the market’s characteristics. At the moment, many of the common learning capabilities are geared toward sustaining innovations and known markets, which makes such capabilities irrelevant to disruptive technology and emerging markets. This dynamic raises the question of how companies can identify cost structures and performance trajectories when the market has yet to mature, let alone exist. Christensen uses the chapter to address this question, examining several case studies from the disk drive, motorcycle, and microprocessor industries.

He points to Disk/Trend Report, an annual publication whose exhaustive data on the disk drive industry enables it to make forecasts in each annual issue. Upon testing the accuracy of these forecasts, Christensen notes that Disk/Trend Report’s forecasts in established markets and sustaining technologies were largely accurate, whereas its forecasts in emerging markets and disruptive innovations were typically off the mark.

Meanwhile, in the motorcycle industry, Honda established itself in the North American market through a process of trial and error. After failing to sell its low-cost motorbikes in the same markets where Harley-Davidson and BMW’s high-powered motorcycles held the lead, Honda’s executives stumbled upon an unintended application in off-road dirt-biking. Despite forecasts showing that Honda’s product was unique in the market it wanted to enter, the company ultimately found success in a new network it was creating by accident. Christensen notes that this incident revealed the true nature of Honda’s low-cost bikes as a disruptive technology, which implied the need to value the product according to different performance attributes than the ones used to grade the established firms’ products. When the company’s North American efforts proved profitable, the market leaders attempted to follow Honda into the market, but they found themselves at odds with their native customer base. Hence, the competitors retreated upmarket, and Honda continued to win more of the market.

Intel found greater success in the microprocessor industry when it inadvertently transitioned away from its flagship products to invest in microprocessors. Intel’s internal resource allocation processes automatically funneled capital and manufacturing capabilities into the disruptive technology, even as management continued to prioritize their mainstream business. In other companies, managers might have prevented the diversion because the microprocessor industry was still hard to analyze. Research forecasts could not anticipate the microprocessor’s popularity with the personal computer market.

Christensen concludes that it is ultimately moot to predict disruptive technology applications and the markets where they will find value. Instead, companies are better off assuming that their early entry strategies will be wrong. Importantly, Christensen differentiates between failed ideas and failed businesses, noting that multiple companies have gone through failed ideas before succeeding as businesses. He concludes that companies are therefore wise to conserve resources for new plans when the first one fails. However, this approach conflicts with conventional management wisdom, since managers are unwilling to brace failure without the guarantee of immediate results. Naturally, they are uncertain of the results because the market is still developing, which is not the case with sustaining technologies. In an established market, managers can trust in the forecasts and aggressively execute their initial plans.

Christensen thus reiterates his suggestion to adopt plans for learning, rather than plans for implementation, whenever disruptive technology is involved. He defines discovery-driven planning as a process that “requires managers to identify the assumptions upon which their business plans or aspirations are based” (157). More often than not, these assumptions include forecast accuracy and target market characteristics. Christensen therefore moves away from the traditional management philosophies that target unexpected failures by stressing the need for more agnostic marketing, which he defines as “marketing under an explicit assumption that no one […] can know whether, how, or in what quantities a disruptive product can or will be used before they have experience using it” (157-58). Managers can thus venture into unexplored territory and seize first-mover advantages as long as they properly anticipate the risks.

In this chapter, Christensen distinguishes individual employee capabilities from the organizational capabilities that exist independent of human resources. Building on the previous chapters’ suggestion to create autonomous organizations for emerging markets, he singles out the popular idea of core competencies and sets out to provide a clear framework for identifying the competencies that are most relevant to disruptive challenges.

The framework is built around three categorical factors: resources, process, and values. Resources are acquirable assets that enable a company to execute its operations. Most managers assume that resources are the most significant factor for assessing change-driving capabilities, but this is only because resources are the most visible capability factors. The next factors drive capabilities in subtler ways.

Processes are “patterns of interaction, coordination, communication, and decision-making” that convert resource inputs into valuable goods and services (163). These include formal processes that have been standardized by the company, as well as informal processes that occur naturally over time. As organizations develop their capabilities, they concretize their activities for efficiency, making it easy to complete certain recurring tasks. However, this approach also makes it difficult to execute tasks that fall outside these processes. Inasmuch as organizations define their capabilities through sustained performance, they also imply their organizational disabilities. Critical to Christensen’s thesis, certain processes influence investment decisions. If they are inflexible to change, those processes will affect the company’s performance in the long run.

Finally, values are the “criteria by which decisions about priorities are made” (164). More than just the moral guidelines by which companies build their integrity, values are conceptualized in Christensen’s framework around decision-making processes and priority ranking. Between two projects with exclusive sets of tasks, values enable employees at every level to determine where they will invest their time, attention, and energy, among other resources. On the organizational level, corporate values usually align with the business model or cost structure, so that employees prioritize the tasks that generate revenue. Similarly, corporate values empower managers to deprioritize projects that fail to realize the company’s operating requirements. That said, engaging in low-margin projects becomes a unique capability that established firms typically do not possess. When firms move upmarket or grow at a sustainable rate, they change their cost structures and market priorities—and thus their values—losing the capability to engage with previous market tiers. In short, the bigger the company, the less capable it is of managing innovation.

Recalling the five disruptive innovations that shifted leadership in the disk drive industry, Christensen points out that the established firms’ capabilities were defined by processes they perfected while developing more than 100 sustaining innovations, as well as their value for higher margins. The intermittent quality of disruptive innovation and low profit margins did not align with their capabilities, even if they had the resources required to carve out new markets. On the other hand, entrant firms who have fewer resources are more capable of developing disruptive innovations for margins that are high from their perspective. Their processes are formed as they search for profitable market applications.

While this insight seems to favor entrant firms, Christensen notes that entrant firms sometimes fail to develop consistent processes or effectively evolve their product line. This is where company founders play a critical role in shaping corporate culture. Founders who properly direct the company’s processes and values imprint their approaches on managers and employees. Christensen describes this as a migration of capabilities. Resources may enable the company to operate at its inception, but culture ensures that the company can sustainably solve the same types of problems as the company grows. This is where change becomes difficult to implement.

Because processes and values are usually inflexible, they will be difficult to shift through training. Although it isn’t impossible to create new capabilities within an organization, Christensen explains that companies will need to redraw their organizational boundaries and willingly sacrifice some of their existing capabilities in the process. This is easier said than done, however, since the process-values mechanism is predicated upon consistent performance and is therefore inflexible. Companies cannot execute two different sets of processes simultaneously.

Christensen explores two alternate approaches: acquiring organizations that match the market tasks, or creating spin-out organizations and creating the capabilities from within. Acquisitions can be beneficial to large companies depending on which capability factor is valued for its potential. When a company is acquired on account of its resource capabilities, it makes sense to integrate that subsidiary with the parent company as a way of augmenting the parent company’s capabilities. On the other hand, when the subsidiary’s processes and values are the source of its success, Christensen cautions against integrating that company with the parent firm, as doing so will merely overwrite the subsidiary’s capabilities. To illustrate the point, he explains that Johnson & Johnson used its acquisitions to gain a foothold in several disruptive markets. By refusing to merge its new companies, it enabled them to retain their processes and operate at greater capability with their parent company’s resources.

Creating spin-out organizations can benefit companies whose values are deeply entrenched, making it impossible to shift priorities toward innovation development. Spin-out organizations will have a perspective of cost structures and growth opportunities that differs from that of mainstream organizations. This approach requires the close supervision of the mainstream firm’s chief executive, since founders define the processes and values that allow the company to realize value.

Ultimately, Christensen introduces a framework for assessing innovation projects against organizational capabilities. The presence of process-driven capabilities will only require lightweight teams who can facilitate the implementation of those capabilities in the new firm. The absence of those same capabilities calls for heavyweight teams to work together and generate new processes. On the other hand, when a new project requires values that the company does not possess, only then should the company create a spin-out organization.