The Innovator's Solution

by Clayton Christensen

  FIGURE 7 - 1

  A Framework for Finding the Right Organizational Structure and Home

  The lower horizontal axis asks managers to assess whether the organization’s values will allocate to the new initiative the resources it will need in order to become successful. If there is a poor fit, then the mainstream organization’s values will accord low priority to the project; that is, the project is potentially disruptive relative to its business model. The upper horizontal axis in figure 7-1 captures on a continuum the level of autonomy needed by an organizational unit attempting to exploit an innovation. For disruptive innovations, setting up an autonomous organization to develop and commercialize the venture will be absolutely essential to its success. At the other extreme, however, if there is a strong sustaining fit, then the manager can expect that the energy and resources of the mainstream organization will coalesce behind it because the project is sustaining. There is no reason for skunkworks or spin-offs in such cases.

  The right vertical axis maps three types of organizational structures that can be used to either exploit or overcome existing processes. The development team charged with shepherding an innovation to market can be either heavyweight, lightweight, or functional (all defined later in this chapter). The four regions in figure 7-1 integrate the challenges of dealing with different types of fit with the mainstream organization’s processes and values. Region A depicts a situation in which a manager is faced with a breakthrough but sustaining technological change. It fits the organization’s values, but it presents the organization with different types of problems to solve and therefore requires new types of interaction and coordination among groups and individuals. This circumstance mandates a heavyweight project team (described later). In region B, where the project fits the company’s processes as well as its values, the new venture can easily be developed by coordinating across functional boundaries within the existing organization. Region C denotes a disruptive technological change that fits neither the organization’s existing processes nor its values. To ensure success in such instances, the managers should create an autonomous organization. Region D typifies projects in which products or services similar to those in the mainstream need to be sold within a fundamentally lower-overhead business model. These ventures can leverage the main organization’s logistics management processes, but they need very different budgeting, management, and profit and loss profiles.19

  In using figure 7-1, it is important to remember that disruption is a relative term. What is disruptive to one company might have a sustaining impact on another. For example, Dell Computer began by selling computers over the telephone. For Dell, the initiative to begin selling over the Internet was a sustaining innovation. It helped Dell make more money in the way it was already structured to make money. Not surprisingly Dell adopted Internet retailing very successfully. For Compaq, Hewlett-Packard, and IBM, however, marketing directly to customers was decidedly disruptive because of its impact on their retail channel partners. They couldn’t make room for Internet distribution within their existing organizations, and so their attempts to incorporate this new channel were far less successful. The only way they could have succeeded at becoming leaders in marketing computers directly to customers was to have done it within an autonomous business unit, and possibly with a new brand.

  Similarly, the Internet is a sustaining technology relative to catalog retailers such as Lands’ End, and so we would expect them to incorporate it into their existing processes. But it is disruptive relative to bricks-and-mortar retailers such as Macy’s, which would require autonomous units in order to exploit on-line retailing in a way that could create truly disruptive growth.20 Similarly, the Internet is a sustaining technology to discount stockbrokers such as Ameritrade, and a disruptive technology to full-service brokers such as Merrill Lynch.

  Organizations cannot disrupt themselves. So when Merrill Lynch implemented Internet-based equities trading within its mainstream brokerage organization, the effect was essentially to bring better information to Merrill’s full-service brokers, to help them do an even better job servicing the needs of their high-net-worth clients. The Internet system was shaped as a sustaining technology relative to Merrill Lynch—and no other outcome would be possible. Furthermore, this was a wise thing for Merrill Lynch to do. Its brokerage business for wealthy clients is a beautiful, profitable business, and Merrill would be crazy not to make it even better and even more profitable.21 But Merrill’s executives should not conclude that they have addressed the threat and opportunity of discount online brokerage that is posed by Charles Schwab. They could only do that if they acquired or created an autonomous unit whose values or cost structure helps them earn attractive profits at discount prices.

  This is an important reason for our observation in chapter 4 that established companies are prone to cram disruptive ideas into the mainstream market, forcing them to compete against consumption on a sustaining-technology basis. As long as the strategies for developing and commercializing these disruptive innovations are developed within the mainstream organization, this is the only outcome that we can expect. An organization’s processes and values ensure that it can only implement sustaining innovations.

  Creating New Capabilities

  The RPV model can be a useful guide for executives who determine that they need to create new capabilities because those that their organization presently has aren’t well suited for building new-growth businesses. This can be framed as a make-or-buy decision. We typically frame make-or-buy decisions as relating to resources, such as training managers internally or hiring them from outside. But processes and values can also be made or bought, as the following discussion describes.

  Creating Management Bench Strength

  In many ways, building the management bench strength required to launch a sequence of new-growth businesses is a chicken-or-the-egg problem. Maximizing the probabilities of success means identifying managers who are able, here and now, to grapple successfully with the challenges of building new businesses. But to develop managers for the future, organizations need to put up-and-coming managers into situations and responsibilities for which they are not yet qualified. It is the only way they can learn the skills required to succeed. You need to be creating successful businesses in order to have the right curriculum within your internal schools of experience in which next-generation managers can learn. And yet having capable managers in place is a prerequisite to building these growth businesses. Successfully wrestling with these dimensions of the innovator’s dilemma is a critical responsibility of a director of human resources.

  By the time a company reaches substantial size, most executives have established processes to identify a set of early-career, high-potential managers who should be prepared, ready and waiting, with the skills to succeed in the situations that will confront the company in the future. In many companies, employees are chosen for this high-potential management track based on early evidence of right-stuff attributes. In these firms, recruiting and promoting up-and-coming leaders entails sifting through lots of people in order to find those few who possess the desired end-state attributes in some nascent form.

  The school-of-experience theory, however, says that potential should not be measured by attributes, but rather by the ability to acquire the attributes and skills needed for future situations. The talent to be sought, in other words, is the ability to learn what needs to be learned from the experiences in which the high-potential employee will be schooled in the future. By focusing on ability to learn, it is possible to avoid the trap of assuming that the finite list of competencies important for today are those that will be required in the future. A performance appraisal form targeted at identifying high-potential people would certainly cover basic technical and cognitive requirements, but would not ask for a ranking on right-stuff attributes. It would focus on learning-oriented measures such as “seeks opportunities to learn,” “seeks and uses feedback,” “asks the right questions,” “looks at things fro
m new perspectives,” and “learns from mistakes.” Some attributes of a good learner will show up as achievements, of course, but the quest is to determine whether an employee is willing to learn new skills.

  Putting people in positions where they will learn, however, creates its own dilemma. Those who are “ready now,” who are deemed to be fully qualified to handle a given job, by definition have the least to learn by doing it. And those who have the most to learn bring the least experience to the task. McCall notes that, as a result, many managers who are intensely focused on delivering ever-improving results often are the worst at developing next-generation management bench strength. It takes extraordinary discipline and vision on the part of senior executives to balance the tension between deploying fully qualified employees to deliver results now versus giving learning opportunities to high-potential employees who need further development. But strike this balance they must.

  Some firms deal with this tension by turning repeatedly to the labor markets, raiding other companies for people with the requisite skills already in full flower. Harkening back to chapters 5 and 6, we believe that one reason why internal management training is becoming more pervasive is because managers don’t yet perform well enough. In-house management development processes in many ways can create an optimized, interdependent interface between the skills of the manager and the processes and values of the company. In situations where management performance is not yet good enough, outsourcing “modular” managers and attempting to plug them into a company’s complex, interdependent system of resources, processes, and values often does not work well.22

  A company that works to develop a sequence of new-growth businesses can build a virtuous cycle in management development. Launching growth business after growth business creates a set of rigorous, demanding schools in which next-generation executives can learn how to lead disruption. Companies that only sporadically attempt to create new-growth businesses, in contrast, offer to their next-generation executives precious few of the courses they need to successfully sustain growth.

  Making New Processes

  The right vertical axis in figure 7-1 shows the kind of development team that is required to create appropriate processes for a new-growth business. When different processes need to be created, it requires what Harvard Business School Professors Kim Clark and Steven Wheelwright call a heavyweight team.23 The term refers to a group of people who are pulled out of their functional organizations and placed in a team structure that allows them to interact over different issues at a different pace and with different organizational groups than they habitually could across the boundaries of functional organizations. Heavyweight teams are tools to create new processes, or new ways of working together. In contrast, lightweight or functional teams are tools to exploit existing processes.

  We can use the concepts of interdependence and modularity from chapter 5 to visualize a heavyweight team and understand when it is important to create one. When there is a well-defined interface between the activities of two different people or organizational groups—meaning that you can clearly specify what each is supposed to deliver, you can measure and verify what they deliver, and there are no unanticipated interdependencies between what one does and what the other must do in response—then those people and groups can interface at arm’s length and need not be on the same team. When these conditions are not met, then all unpredictable interdependencies should be incorporated within the boundaries of a heavyweight team. The team’s external boundary can be drawn where there are modular interfaces. New methods of working together can coalesce within this team as it addresses its task. These can then become codified as processes if the team is kept intact and addresses a similar task repeatedly.24

  To be successful, heavyweight teams should be co-located. Team members bring their functional expertise to the group, but they do not represent their functional group’s “interests” on the team. Their responsibilities are simply to do what needs to be done in order for the project to be successful—even if that course of action is not optimal for their functional group. Many companies have used heavyweight teams successfully as a method for creating new processes. Chrysler, for example, historically structured its product development groups around specific components, such as electrical systems. When the changing basis of competition in its industry forced Chrysler to accelerate the development of new automobiles in the early 1990s, Chrysler organized its development teams around platforms like the minivan, instead of the technical subsystems. The heavyweight teams that Chrysler created were consequently not as good at focusing on component design, but the teams forged new processes that were much faster and more efficient in creating entirely new car designs. This was a critical achievement as the basis of competition changed. Companies as diverse as Medtronic in cardiac pacemakers, IBM in disk dives, and Eli Lilly with its schizophrenia drug Zyprexa have used heavyweight teams as vehicles for creating different, faster processes.25

  Drawing flow diagrams does not create radically different processes. Rather, executives build processes by giving a group of people in a heavyweight team a new problem that the organization has not confronted before. After the team has successfully addressed the challenge, the team needs to confront a similar problem again, and then again. Ultimately, this new way of working will become ensconced within the team and then can diffuse throughout the organization.

  Creating New Values

  Companies can create new prioritization criteria, or values, only by setting up new business units with new cost structures. Charles Schwab, for example, set up its disruptive online brokerage venture as a completely autonomous organization. It priced online trades at $29.95, compared with the average price of nearly $70 that Schwab had been charging for trades executed through its telephone and office-based brokers. The separate unit was indeed disruptive to the mainstream. It grew so fast that within eighteen months the company decided to fold what had been the mainstream business into the new disruptive organization. The corporation’s values, which in our model are synonymous with its cost structure, were thereby transformed by launching a successful disruptive enterprise. Schwab’s corporate values changed when the disruptive business displaced the old organization, whose values were incapable of prioritizing the disruptive growth business.

  The reason an organization cannot disrupt itself is that successful organizations can only naturally prioritize innovations that promise improved profit margins relative to their current cost structure. For Schwab, therefore, it was far more straightforward to create a new business model that could view $29.95 as a profitable proposition than it would have been to hack enough cost out of the original organization so that it could make money at the disruptive price point. This is the best way to change values because the new, disruptive game almost always must begin while the established business still has substantial, profitable sustaining potential.

  What does autonomous mean? Our research suggests that geographical separation from the core business is not a critical dimension of autonomy. Nor is ownership structure. There is no reason why a disruptive venture cannot be wholly owned by its parent. The key dimensions of autonomy relate to processes and values. The disruptive business needs to have the freedom to create new processes and to build a unique cost structure in order to be profitable as it makes and sells even its earliest products. Making the judgment calls about which of the mainstream businesses’ processes and overhead costs the new venture should and should not accept is a key role of the CEO in building new-growth businesses. We will return to this in chapter 10.

  Buying Resources, Processes, and Values

  Managers often think that acquiring rather than developing a set of capabilities makes competitive and financial sense. Unfortunately, companies’ track records in developing new capabilities through acquisition are frighteningly spotty. The RPV framework can be a useful way to frame the challenge of integrating acquired organizations. Every time one company acquires another, it buys its resources
, its processes, and its values. Acquiring managers therefore need to begin by asking, “What is it that really made this company that I just bought so expensive? Did I justify the price because of its resources—its people, products, technology, or market position? Or was a substantial portion of its worth created by its processes and values—its unique ways of working and decision making that have enabled the company to understand and satisfy customers; develop, make, and deliver new products in a timely way; and to do so within a cost structure that gave it disruptive potential?”

  If the acquired company’s processes and values are the real drivers of its success, then the last thing the acquiring manager wants to do is to integrate the company into the new parent organization. Integration will vaporize many of the processes and values of the acquired firm as its managers are required to adopt the buyer’s way of doing business and have their new-growth proposals evaluated according to the decision criteria of the acquiring company. If its processes and values were the reason for its historical success, a better strategy is to let the acquired business stand alone, and for the parent to infuse its resources into the acquired firm’s processes and values. This strategy, in essence, truly constitutes the acquisition of new capabilities.

  If, on the other hand, the company’s resources were the primary rationale for the acquisition, then integrating the firm into the parent makes a lot of sense—essentially plugging the acquired people, products, technology, and customers into the parent’s processes as a way of leveraging the parent’s existing capabilities.