Your firm has a competitive advantage when it generates and appropriates more economic value than your competitors. You can gain a competitive advantage by selling innovative new products or services, by rebranding the products or services you sell, by cutting your costs and reducing the prices of your products or services, or in countless other ways.

But how long will your competitive advantage last? Most competitive advantages are quickly competed away when current or new competitors duplicate the sources of your advantage. Indeed, this competitive intensity has accelerated to the point that some observers have suggested that sustained competitive advantage is no longer possible—that the best you can hope for is to develop a string of short-lived advantages that are rapidly competed away.

Our research shows that competitive advantage can still be sustainable—even in a globally competitive, technologically accelerated world—when the resources and capabilities you use to conceive of and implement your strategies are costly or difficult for other firms to imitate. When you leverage these kinds of resources and capabilities in choosing and implementing your strategy, they can be at least partially immune to imitation, and thus be a source of sustained competitive advantage.

So, what kinds of strategically relevant resources and capabilities are costly to imitate? Traditionally, scale and its associated cost advantages played such a role in physical industries. The importance of sheer scale has diminished, however, and it can even be a source of inertia in dynamic environments. Nowadays, inimitable resources are generally socially complex in nature, develop over long periods of time in a “path-dependent” way, or are bundled together to enable strategic choice and implementation. While we will describe each of these sources separately, sustainably competitive companies often have resources with two or even all three of these attributes.

Are Your Resources and Capabilities Socially Complex?

The creation and implementation of your strategies sometimes require the active and creative involvement of many of your employees and other stakeholders. In these settings, it is not enough for you to simply tell your critical employees and stakeholders what your strategy is and what their role in implementing that strategy must be. Instead, your critical employees and stakeholders must work with you to help conceive and implement your strategy.

But when will your employees and stakeholders be willing and able to engage in mutually creating and implementing your strategy? This kind of cooperation requires the development of trust and reciprocity—a sense that you and your senior management team are committed to not only bettering yourselves but also bettering them too. In short, your critical employees and stakeholders must trust your intentions.

This kind of trust develops over time and is often reflected in your firm’s culture—in the stories your employees and stakeholders tell about you and your top management team. Your culture can thus ensure that critical employees and other stakeholders—including the customers who trust your brand promise—know that you and your top management team will value their input in developing and implementing a strategy.

It is this socially complex mix of trust, reciprocity, and commitment that your competitors may have a hard time imitating, especially if they do not have a history with this kind of organizational culture. If the strategies you choose and implement depend critically on these resources and capabilities, then those strategies can be costly for your competitors to imitate.

Consider Toyota’s lean production system. The quality advantages of this system have been known since at least the mid-1960s. And yet, many competitors still lag behind Toyota in many measures of quality, including reliability. And this is the case, even though Toyota has let both academics and competitors tour their factories, study their manufacturing processes, and analyze their supply chains.

Of course, these tangible aspects of Toyota’s production system are important in understanding its competitive advantage in quality, and they have been largely imitated by Toyota’s competitors. But what has been more difficult to imitate has been the complex social relationship between Toyota’s management, its employees, some of its suppliers, and its customers.

Maybe someday, competitors will be able to transform their cultures and fully imitate Toyota’s lean production system. If that happens, then Toyota’s competitive advantage will no longer be sustained. In the meantime, despite intense worldwide competition, Toyota has been able to sustain a competitive advantage.

Are Your Resources and Capabilities Path Dependent?

Sometimes, it is possible to plug in new resources and capabilities instantly simply by investing in a new machine, a new distribution system, new IT, or some other tangible assets. However, at other times, the only way a firm can develop new resources and capabilities is to have traveled a particular path through time, a path that has enabled a firm to accumulate experience and judgment—something that might be called know-how—that cannot be obtained in any other way. These path-dependent capabilities can be very costly to imitate.

Consider ASML, the Dutch firm that has a virtual monopoly in the world market for machines that make the most sophisticated semiconductor chips. Formed as a joint venture between ASM and Philips in 1984, ASML became an independent firm in 1988. In 2004, it formed an alliance with Taiwan Semiconductor Manufacturing (TSMC) to develop and make immersion lithography machines that could compete with Nikon and Canon in making machines that produce semiconductor chips at scale.

During the 1990s, ASML built on the problem-solving skills it had developed in immersion lithography machines to introduce a deep ultraviolet (DUV) lithography machine to the market. When fully developed, these machines produced, at scale, semiconductors with elements down to 20 nanometers in width. These machines are still widely used around the world to manufacture less sophisticated semiconductor chips, with elements between 70 and 120 nanometers in width. (For comparison, the width of a human hair is approximately 100,000 nanometers.)

To go to the next level of semiconductor sophistication, ASML had to develop extreme ultraviolet (EUV) lithography machines. These machines required entirely new technology, compared to the DUV machines, but were still built on the problem-solving skills ASML had developed in bringing the immersion and DUV technologies to market. For example, extreme ultraviolet light does not naturally occur on earth—only in outer space. To produce this light, ASML fires a laser onto microscopic droplets of tin, which are then turned into a plasma that generates extreme ultraviolet light. That light is then used to produce, at scale, semiconductors with elements down to 13 nanometers in width.

ASML’s most recent technology—so-called High NA (numerical aperture) EUV machines—can produce, at scale, semiconductors with elements down to 8 nanometers in width, and ASML can see a path to manufacturing logic and memory chips with elements down to 2 nanometers in width.

ASML has been developing these machines, and the engineering and problem-solving skills needed to build them, for more than 30 years. During this time, it has developed relationships with over 5,000 suppliers, including the German optics firm Zeiss, and has acquired several other firms, including Cymer, the firm that designed and made the machines that expose droplets of tin to a laser light. Introducing these advanced technologies often required additional development when they were put into operation in customers’ factories. ASML has worked closely with its customers—including Samsung and TSMC—to improve the efficiency of these machines once they are installed.

The technical and organizational know-how ASML has developed while building these four generations of semiconductor manufacturing machines—that is, its unique path through time—has given ASML a costly to imitate competitive advantage in building machines to manufacture the most sophisticated semiconductors in the world.

In the mid-2000s, ASML competitors, including Nikon and Canon, chose not to invest in EUV machines, thinking that the technology was too uncertain, too complex, too expensive, and that it did not have a clear position in the semiconductor manufacturing marketplace. Now that ASML has demonstrated the potential of EUV and High NA EUV, they and other firms are scrambling to build their own machines for these segments. However, most observers think it will take at least ten years to catch up with ASML’s technology.

Is Your Advantage Based on Interdependent Systems of Resources and Capabilities?

Sometimes, your ability to choose and implement a strategy requires only a single resource or capability—a new factory, a new enterprise resource planning (ERP) system, a new supplier. Such strategies are relatively easy to implement. However, their very simplicity and wide availability can make them easy to imitate.

Other times, the choice and implementation of a strategy requires many interrelated resources and capabilities, all of which need to work together to realize the full value of your strategies.

Strategies that require these interrelated bundles of resources and capabilities to choose and implement can be costly to imitate for at least two reasons. First, to do so, your competitors must identify and imitate all the different resources and capabilities you use to choose and implement your strategies. This can be a difficult task, especially when there are many such resources and capabilities involved.

Second, it can be difficult to tell which of the many resources deployed are particularly important for your competitors to imitate. They may have guesses about what needs to be imitated first and what can be delayed, but these can easily be wrong.

So, while the need to bundle interdependent systems of resources and capabilities may complicate the implementation of your strategic choices, this can also increase the cost to your competitors of imitating your strategies. Complexity comes at a cost, but also has strategic advantages in this respect.

Consider, for example, Amazon’s retail strategy. Which are the most important elements of this strategy—its ability to manage hundreds of thousands of supply relationships’ its warehousing operations; its delivery systems; its software for searching among products; its marketing efforts to brand its services; its knowledge about the purchasing decisions of all its customers; the complexity of its multiple and interrelated IT systems; or its culture, which demands high accountability from all its employees? Or is it the interplay of all these resources and capabilities? If the latter, this greatly increases the cost of imitating Amazon’s strategy, since to imitate Amazon, you would have to imitate all the many things they do to gain a competitive advantage.

Choosing a Strategy That Generates a Sustained Advantage

Thus, it is possible to choose strategies that generate a sustained competitive advantage, but only if those strategies leverage your unique, socially complex, path-dependent, or multiple, interrelated systems of resources and capabilities. Strategies that do not leverage these kinds of resources can be sources of only temporary advantage, and the easier it is for your competitors to imitate the resources and capabilities you used to choose and implement these strategies, the shorter this temporary advantage.

However, even sustained competitive advantages do not last forever. While they are not competed away through imitation, new technologies or changes in consumer preferences can make what were sources of sustained advantage sources of disadvantage. For example, Toyota could have lost its quality advantage if it had fallen behind in hybrid or electric cars; ASML could have lost its innovation advantage if it had failed to invest in the next generation of microchip production technology (that is, the High NA EUV machine); Amazon could lose its advantage if it fails to anticipate changes in consumer shopping preferences.

The very complexity of resources that is the basis for competitive advantage is also an obstacle to creating an advantaged company in new circumstances. Companies that are sustainably advantaged by complex resources, therefore, need to take the additional step of making their business systems evolvable. This is why Alibaba adopted the mantra that everything in a business needs to be evolvable under changing market conditions, and defaults to change and experimentation to maintain evolvability.

Interestingly, this ability to evolve can itself be a socially complex, path-dependent, and complex bundle of resources, and thus can be a source of sustained competitive advantage.

When Sustained Competitive Advantage Is Not Possible

Perhaps you have concluded that your firm does not possess the kinds of resources it needs to choose and implement a strategy that can generate a sustained advantage. Or perhaps you have costly to imitate resources and capabilities, but they cannot be leveraged in choosing and implementing a value-creating strategy. What are your strategic options?

First, your firm can survive even if it only gains competitive parity from implementing its strategies. But you may have little margin for error. Any mistakes you make in choosing or implementing parity-generating strategies can lead to economic losses that could put your firm’s survival at risk.

Second, while the best time to have developed the kinds of resources and capabilities needed to generate a sustained competitive advantage from choosing and implementing your strategies was years ago, the second-best time to do so is now. Build a culture that can be a source of competitive advantage later; invest in innovations that may generate path-dependent learning that can pay off in the future; learn how to bundle your resources and capabilities to choose and implement your next strategies. In short, do the things now to generate a competitive advantage later.