Good Strategy/Bad Strategy: The difference and why it matters

by Richard P. Rumelt

Despite the roar of voices wanting to equate strategy with ambition, leadership, “vision,” planning, or the economic logic of competition, strategy is none of these. The core of strategy work is always the same: discovering the critical factors in a situation and designing a way of coordinating and focusing actions to deal with those factors.

The best answer to this puzzle is that the real surprise was that such a pure and focused strategy was actually implemented. Most complex organizations spread rather than concentrate resources, acting to placate and pay off internal and external interests. Thus, we are surprised when a complex organization, such as Apple or the U.S. Army, actually focuses its actions. Not because of secrecy, but because good strategy itself is unexpected.

Bad strategy, I explained, is not the same thing as no strategy or strategy that fails rather than succeeds. Rather, it is an identifiable way of thinking and writing about strategy that has, unfortunately, been gaining ground. Bad strategy is long on goals and short on policy or action. It assumes that goals are all you need. It puts forward strategic objectives that are incoherent and, sometimes, totally impracticable. It uses high-sounding words and phrases to hide these failings.

An interesting aspect of this language is the idea that leadership teams must share common beliefs and values. This is now a frequent demand in education circles. One would hope that the experience of North Korea would have cured people of the idea that forcing everyone to believe in and value the same things is the road to high performance. Yet, within politically correct edu-speak, this impossible state of affairs is continually sought as the path to “transformational change.”

There has been a lot of ink spilled on the inner logic of competitive strategy and on the mechanics of advantage. But the essential difficulty in creating strategy is not logical; it is choice itself. Strategy does not eliminate scarcity and its consequence—the necessity of choice. Strategy is scarcity’s child and to have a strategy, rather than vague aspirations, is to choose one path and eschew others. There is difficult psychological, political, and organizational work in saying “no” to whole worlds of hopes, dreams, and aspirations.

Strategies focus resources, energy, and attention on some objectives rather than others. Unless collective ruin is imminent, a change in strategy will make some people worse off. Hence, there will be powerful forces opposed to almost any change in strategy. This is the fate of many strategy initiatives in large organizations. There may be talk about focusing on this or pushing on that, but at the end of the day no one wants to change what they are doing very much. When organizations are unable to make new strategies—when people evade the work of choosing among different paths into the future—then you get vague mom-and-apple-pie goals that everyone can agree on. Such goals are direct evidence of leadership’s insufficient will or political power to make or enforce hard choices. Put differently, universal buy-in usually signals the absence of choice.

To see this displacement, consider Peter Senge’s enormously successful book The Fifth Discipline, published in 1990. One of Senge’s most influential ideas was the vital importance of “shared vision”: “It is impossible to imagine the accomplishments of building AT&T, Ford, or Apple in the absence of shared vision.… What is most important is that these individual’s visions became genuinely shared among people throughout all levels of their companies—focusing the energies of thousands and creating a common identity among enormously diverse people.”12 This statement is very appealing to many people and, at the same time, is quite obviously untrue. Ascribing the success of Ford and Apple to a vision, shared at all levels, rather than pockets of outstanding competence mixed with luck, is a radical distortion of history. Apple did not invent the personal computer—the technology was in the air and hundreds of entrepreneurs sought to design and build “computers for everyman.” Apple’s success derived, in large measure, from Steve Wozniak’s ability to cleverly trick the Motorola microprocessor at the center of the Apple II into directly driving the video and a floppy disk—using straight outputs from the CPU—rather than building expensive controllers. And, from the advent of VisiCalc, which gave people other than hobbyists a reason to buy an Apple II.13 Similarly, Ford’s “cars for everyman” vision was hardly unique and not necessarily shared by the five-dollar-a-day assembly line work...

A great deal of strategy work is trying to figure out what is going on. Not just deciding what to do, but the more fundamental problem of comprehending the situation.

The diagnosis for the situation should replace the overwhelming complexity of reality with a simpler story, a story that calls attention to its crucial aspects. This simplified model of reality allows one to make sense of the situation and engage in further problem solving. Furthermore, a good strategic diagnosis does more than explain a situation—it also defines a domain of action. Whereas a social scientist seeks a diagnosis that best predicts outcomes, good strategy tends to be based on the diagnosis promising leverage over outcomes. For instance, we know from research that K–12 student performance is better explained by social class and culture than by expenditures per student or class size, but that knowledge does not lead to many useful policy prescriptions. A very different strategic diagnosis has been provided by my UCLA colleague Bill Ouchi. His book Making Schools Work diagnoses the challenge of school performance as one of organization rather than as one of class, culture, funding, or curriculum design.3 Decentralized schools, he argues, perform better. Now, whether the organization of a school system explains most of the variations in school performance is not actually critical. What is critical, and what makes his diagnosis useful to policy makers, is that organization explains some part of school performance and that, unlike culture or social class, organization is something that can be addressed with policy.

The dominant view at the company and among Wall Street analysts was that IBM was too integrated. The new industry structure was fragmented and, it was argued, IBM should be broken up and fragmented to match. As Gerstner arrived, preparations were under way for separate stock offerings for various pieces of IBM. After studying the situation, Gerstner changed the diagnosis. He believed that in an increasingly fragmented industry, IBM was the one company that had expertise in all areas. Its problem was not that it was integrated but that it was failing to use the integrated skills it possessed. IBM, he declared, needed to become more integrated—but this time around customer solutions rather than hardware platforms. The primary obstacle was the lack of internal coordination and agility. Given this new diagnosis, the guiding policy became to exploit the fact that IBM was different, in fact, unique. IBM would offer customers tailored solutions to their information-processing problems, leveraging its brand name and broad expertise, but willing to use outside hardware and software as required. Put simply, its primary value-added activity would shift from systems engineering to IT consulting, from hardware to software. Neither the “integration is obsolete” nor the “knowing all aspects of IT is our unique ability” viewpoints are, by themselves, strategies. But these diagnoses take the leader, and all who follow, in very different directions.

You may correctly observe that many other people use the term “strategy” for what I am calling the “guiding policy.” I have found that defining a strategy as just a broad guiding policy is a mistake. Without a diagnosis, one cannot evaluate alternative guiding policies. Without working through to at least the first round of action one cannot be sure that the guiding policy can be implemented. Good strategy is not just “what” you are trying to do. It is also “why” and “how” you are doing it.

On the other hand, the potential gains to coordination do not mean that more centrally directed coordination is always a good thing. Coordination is costly, because it fights against the gains to specialization, the most basic economies in organized activity. To specialize in something is, roughly speaking, to be left alone to do just that thing and not be bothered with other tasks, interruptions, and other agents’ agendas. As is clear to anyone who has belonged to a coordinating committee, coordination interrupts and de-specializes people. Thus, we should seek coordinated policies only when the gains are very large. There will be costs to demanding coordination, because it will ride roughshod over economies of specialization and more nuanced local responses. The brilliance of good organization is not in making sure that everything is connected to everything else. Down that road lies a frozen maladaptive stasis. Good strategy and good organization lie in specializing on the right activities and imposing only the essential amount of coordination.

In 1981, I was fortunate enough to spend a week with Pierre Wack at a Shell Group Planning retreat in Runnymede, England. Talking about scenarios, he told me: If you do standard “scenario” forecasting, you wind up with a graph with three lines labeled “high,” “medium,” and “low.” Everyone looks at it and believes that they have paid attention to the uncertainty. Then, of course, they plan on “medium”! But they are missing the risk. The risk is not that the price of oil may be high or may be low. The risk is that it will go high, suckering you into a major investment, and then turn and dive to low, leaving you with useless assets.

Murata explained that the company had come to the conclusion that Japanese customers were extremely sensitive to variations in local tastes and fond of both newness and variety. “In Japan,” he told me, “consumers are easily bored. In soft drinks, for example, there are more than two hundred soft-drink brands and lots of new ones each week! A 7-Eleven displays fifty varieties with a turnover of seventy percent each year. The same holds true in many food categories.” To create leverage around this pattern, 7-Eleven Japan has developed a method of collecting information from store managers and employees about local tastes and forming quick-response merchandising teams to develop new product offerings. To further leverage this information and team skills, the company has developed relationships with a number of second- and third-tier food manufacturers and found ways to quickly bring new offerings to market under its own private-label brand, at low prices, using the food manufacturers’ excess capacity.

For example, there seems to be a threshold effect in advertising. That is, a very small amount of advertising will produce no result at all. One has to get over this hump, or threshold, to start getting a response to advertising efforts.5 This means it may pay companies to pulse their advertising, concentrating it into relatively short periods of time, rather than spreading it evenly. It may also make sense for a company to roll out a new product region by region, concentrating its advertising where the product is new so as to spur adoption. Due to similar forces, business strategists will often prefer to dominate a small market segment over having an equal number of customers who represent only a sliver of a larger market. Politicians will often prefer a plan that delivers a clear benefit to a recognizable group over one that provides larger benefits spread more thinly across the population. Within organizations, some of the factors giving rise to concentration are the substantial threshold effects in effecting change and the cognitive and attention limits of the senior management group. Just as an individual cannot solve five problems at once, most organizations concentrate on a few critical issues at any one time. From a psychological perspective, there can be returns to focus or concentration when people ignore signals below a certain threshold (called a “salience effect” in psychology) or when they believe in momentum—that success leads to success. In either case, the...

At that time, I worked for Phyllis Buwalda, who directed Future Mission Studies at JPL. Homeschooled on a ranch in Colorado, Phyllis had a tough, practical intellect that could see to the root of a problem. She was best known for her work on a model of the lunar surface.3 With this specification in place, JPL engineers and subcontractors were able to stop guessing and get to work. The lunar surface Phyllis described was hard and grainy, with slopes of no more than about fifteen degrees, scattered small stones, and boulders no larger than about two feet across spaced here and there. Looking at this specification for the first time, I was amazed. “Phyllis,” I said, “this looks a lot like the Southwestern desert.” “Yes, doesn’t it?” she said with a smile. “But,” I complained, “you really don’t know what the moon is like. Why write a spec saying it is like the local desert?” “This is what the smoother parts of the earth are like, so it is probably a pretty good guess as to what we’ll find on the moon if we stay away from the mountains.” “But, you really have no idea what the surface of the moon is like! It could be powder, or jagged needles.…” “Look,” she said, “the engineers can’t work without a specification. If it turns out to be a lot more difficult than this, we aren’t going to be spending much time on the moon anyway.” Her lunar specification wasn’t the truth—the truth was that we didn’t know. It was a strategically chosen proximate objective—one the engineers knew how t...

Phyllis’s insight that “the engineers can’t work without a specification” applies to most organized human effort. Like the Surveyor design teams, every organization faces a situation where the full complexity and ambiguity of the situation is daunting. An important duty of any leader is to absorb a large part of that complexity and ambiguity, passing on to the organization a simpler problem—one that is solvable. Many leaders fail badly at this responsibility, announcing ambitious goals without resolving a good chunk of ambiguity about the specific obstacles to be overcome. To take responsibility is more than a willingness to accept the blame. It is setting proximate objectives and handing the organization a problem it can actually solve.

The proximate objective is guided by forecasts of the future, but the more uncertain the future, the more its essential logic is that of “taking a strong position and creating options,” not of looking far ahead. Herbert Goldhamer’s description of play between two chess masters vividly describes this dynamic of taking positions, creating options, and building advantage: Two masters trying to defeat each other in a chess game are, during a large part of the game, likely to be making moves that have no immediate end other than to “improve my position.” One does not win a chess game by always selecting moves that are directly aimed at trying to mate the opponent or even at trying to win a particular piece. For the most part, the aim of a move is to find positions for one’s pieces that (a) increase their mobility, that is, increase the options open to them and decrease the freedom of operation of the opponent’s pieces; and (b) impose certain relatively stable patterns on the board that induce enduring strength for oneself and enduring weakness for the opponent. If and when sufficient positional advantages have been accumulated, they generally can be cashed in with greater or less ease by tactical maneuvers (combinations) against specific targets that are no longer defensible or only at terrible cost.

Quality matters when quantity is an inadequate substitute. If a building contractor finds that her two-ton truck is on another job, she may easily substitute two one-ton trucks to carry landfill. On the other hand, if a three-star chef is ill, no number of short-order cooks is an adequate replacement. One hundred mediocre singers are not the equal of one top-notch singer. Keeping children additional hours or weeks in broken schools—schools that can neither educate nor control behavior—does not help and probably increases resentment and distrust.

What is especially fascinating is that both excellence and being stuck are reflections of chain-link logic. In the case of excellence, like IKEA, a series of chain-linked activities are all maintained at a high level of quality, each benefiting from the quality of the other and the whole being resistant to easy imitation. On the other hand, when a series of chain-linked activities are of low quality, as in General Motors circa 2007, the system can be stuck, because there is little gain to improving only a fraction of the activities. Marco Tinelli’s success demonstrates that to unstick a stuck chain-linked system, a strong leader must possess the insight and fortitude to make the necessary investments in each link of the chain.

I am describing a strategy as a design rather than as a plan or as a choice because I want to emphasize the issue of mutual adjustment. In design problems, where various elements must be arranged, adjusted, and coordinated, there can be sharply peaked gains to getting combinations right and sharp costs to getting them wrong. A good strategy coordinates policies across activities to focus the competitive punch.

Implicit in these principles is the notion that tight integration comes at some cost. That is, one does not always seek the very highest level of integration in a design for a machine or a business. A more tightly integrated design is harder to create, narrower in focus, more fragile in use, and less flexible in responding to change. A Formula 1 racing car, for example, is a tightly integrated design and is faster around the track than a Subaru Forester, but the less tightly integrated Forester is useful for a much wider range of purposes. Nevertheless, when the competitive challenge is very high, it may be necessary to accept these costs and design a tightly integrated response. With less challenge, it is normally better to have a bit less specialization and integration so that a broader market can be addressed.

The problem with diving into the growing PET industry was that growth in a commodity—such as cement or aluminum or PET containers—is an industry phenomenon, driven by an increase in overall demand. The growing demand pulls up profit, which, in turn, induces firms to invest in new capacity. But most of the profits of the growing competitors are an illusion because they are plowed back into new plant and equipment as the business grows. If high profits on these investments can be earned after growth slows, then all is well. But in a commodity industry, as soon as the growth in demand slows down, the profits vanish for firms without competitive advantages. Like some sort of economic black hole, the growing commodity industry absorbs more cash from the ordinary competitor than it ever disgorges. The proposition that growth itself creates value is so deeply entrenched in the rhetoric of business that it has become an article of almost unquestioned faith that growth is a good thing.

Defining “sustainability” is trickier. For an advantage to be sustained, your competitors must not be able to duplicate it. Or, more precisely, they must not be able to duplicate the resources underlying it. For that you must possess what I term an “isolating mechanism,” such as a patent giving its holder the legally enforceable right to monopolize the use of a technology for a time.2 More complex forms of isolating mechanisms include reputations, commercial and social relationships, network effects,* dramatic economies of scale, and tacit knowledge and skill gained through experience.

I could see that Stewart’s approach to the nut business was a complex coordinated maneuver over a decade of time. His original large-scale holdings enabled him to capture the lion’s share of benefits from investments in research, market development, advertising, and promotion. The seven-to-ten-year lag in competitive response provided both the financing and a window of opportunity to build large-scale nut-processing facilities. The economies of scale in processing have so far prevented smaller competitors from achieving equivalent costs. It must have taken iron nerve to wait years for the strategy to work. “You are still looking five to ten years ahead?” I query. “It is one of the big benefits of being a private company. When I first bought these lands from major oil companies, they were looking ahead one quarter or one year. They wanted to get the assets ‘off their books’ to make their financial ratios look better. We can do more with these businesses because we don’t suffer the crazy pressures that are put on a public company.”

Extensions based on proprietary know-how benefit from the fact that knowledge is not “used up” when it is applied; it may even be enhanced. By contrast, extensions based on customer beliefs, such as brand names, relationships, and reputation, may be diluted or damaged by careless extension. Although great value can sometimes be created by extending these resources, a failure in the new arena can rebound to damage the core.

When change occurs, most people focus on the main effects—the spurts in growth of new types of products and the falling demand for others. You must dig beneath this surface reality to understand the forces underlying the main effect and develop a point of view about the second-order and derivative changes that have been set into motion. For example, when television appeared in the 1950s it was clear that everyone would eventually have one and that “free” TV entertainment would provide strong competition to motion pictures. A more subtle effect arose because the movie industry could no longer lure audiences out of their homes with “just another Western.” Traditional Hollywood studios had been specialized around producing a steady stream of B-grade movies and did not easily adapt. By the early 1960s, movie attendance was shrinking rapidly. What revived Hollywood film was a shift to independent production, with studios acting as financiers and distributors. Independent producers, freed from the nepotism and routines of the traditional studio, could focus on assembling a handpicked team to make a film that might be good enough to pull an audience off of their family-room sofas. Thus, a second-order effect of television was the rise of independent film production.

Sloan’s product policy is an example of design, of order imposed on chaos. Making such a policy work takes more than a plan on a piece of paper. Each quarter, each year, each decade, corporate leadership must work to maintain the coherence of the design. Without constant attention, the design decays. Without active maintenance, the lines demarking products become blurred, and coherence is lost. If the company is fully decentralized, this blurring of boundaries is bound to happen and the original design, based on brands defined around price ranges, becomes buried under a clutter of new products. For example, executives running the Chevrolet division knew that they could increase the Chevrolet division’s sales and profit by offering some higher-priced models. This move might take some business away from Chrysler; it would also take business away from Pontiac and Oldsmobile. Conversely, executives running the Pontiac division saw that they could increase their division’s revenues if they offered some lower-priced models, so they pushed in that direction. Like a parent who must resist fourteen-year-olds’ pushing for beer at the party, corporate management’s job is to resist these imprecations and preserve the design. If the design becomes obsolete, management’s job is to create a new way of coordinating efforts so that the competitive energy is directed outward instead of inward.

Treating strategy like a problem in deduction assumes that anything worth knowing is already known—that only computation is required. Like computation, deduction applies a fixed set of logical rules to a fixed set of known facts. For example, given Newton’s law of gravity, one can deduce (calculate) the period of Mars’s orbit around the sun. Or given the costs and capacities of tankers, pipelines, and refineries, one can optimize the flow of oil and refined product within an integrated oil company. If everything worth knowing is already known, the problem of action reduces to crank winding. The presumption that all important knowledge is already known, or available through consultation with authorities, deadens innovation. It is this presumption that stifles change in traditional societies and blocks improvement in organizations and societies that come to believe that their way is the best way. To generate a strategy, one must put aside the comfort and security of pure deduction and launch into the murkier waters of induction, analogy, judgment, and insight.

Joe Santos’s comments imply that incumbents had difficulty understanding Starbucks because it was vertically integrated—because it roasted, branded, and served its own coffee in its own company restaurants. Starbucks did not vertically integrate to purposefully confuse the competition. It did so in order to be able to mutually adjust multiple elements of its business and to capture the information generated by each element of its business operations. Integration is not always a good idea. When a company can buy perfectly good products and services from outside suppliers, it is usually wasteful to go through the expense and trouble of mastering a new set of business operations. However, when the core of a business strategy requires the mutual adjustment of multiple elements, and especially when there is important learning to be captured about interactions across business elements, then it may be vital to own and control these elements of the business mix.

The problem is that there might be better ideas out there, just beyond the edge of our vision. But we accept early closure because letting go of a judgment is painful and disconcerting. To search for a new insight, one would have to put aside the comfort of being oriented and once again cast around in choppy waters for a new source of stability. There is the fear of coming up empty-handed. Plus, it is unnatural, even painful, to question our own ideas. Thus, when we do come up with an idea, we tend to spend most of our effort justifying it rather than questioning it. That seems to be human nature, even in experienced executives. To put it simply, our minds dodge the painful work of questioning and letting go of our first early judgments, and we are not conscious of the dodge.

The introductory topic taught in any modern course on business strategy is the connection between industry structure and profit. This topic is usually called the “Five Forces,” following Michael Porter’s pioneering analysis of industry structure, published in 1980. A quick summary is that a terrible industry looks like this: the product is an undifferentiated commodity; everyone has the same costs and access to the same technology; and buyers are price sensitive, knowledgeable, and willing to switch suppliers at a moment’s notice to get a better deal.

For centuries, mathematicians believed that within any axiomatic system—such as geometry, arithmetic, or algebra—every statement was either true or false. In 1931, Viennese mathematician Kurt Gödel proved that this intuition was wrong. He showed that sufficiently complex logical systems are “incomplete.” That is, they contain statements and propositions that cannot be judged true or false within the logic of the system. To judge their truth one most look beyond the system to outside knowledge.4 I believe this idea applies, metaphorically, to human systems. Specifically, when executives and experts invest and manage solely based on recent stock market signals, the flow of information from corporate investment decisions to traders to stock prices and then back to corporate decisions becomes a closed circle. The “axiom” of market accuracy anchors the circle’s center. In the fiber-optic business, this closure was sealed because analysts measured “growth” by looking at increases in installed capacity. Within that closed system, a question such as “Are transatlantic cables an overcapacity commodity business?” became what Kurt Gödel would call an un-decidable proposition. To answer this question one would have to look outside the closed system and make an independent estimate of the situation.

In America, the nineteenth-century credit crises grew out of initiatives aimed at settling western lands and at the concomitant extension of the railroads. During most of the twentieth century, the focus switched to home ownership. In housing, a myriad of government policies, tax breaks, regulations, oversight organizations, and other instruments have, over generations, promoted an ever-increasing rate of home ownership. The original Jeffersonian ideal was a nation of citizen farmers, each owning the means of his or her own support. Today, this vision has morphed into one of a nation of homeowners, each working 100 days a year to pay their taxes and another 125 days a year to pay their mortgages.