Ecosystems as an engine for innovation and learning

Bill Fischer

The acceleration of industry change, as measured in shorter S-curve periods, and the profound changes that are taking place in industry definitions as well, are sure signs that the possession of knowledge stocks is not likely to be as beneficial to an organization in the future as it was to prior generations. Knowing things will be less valuable than knowing how to learn and transforming learning into responses will quickly become the complementary part of a new formula for organizational success: rapid and experimental learning and responding. Most contemporary organizations are not well set-up for either learning or quick response. All too often, they have been built for the quieter, mid-section of longer S-curves, where the game is all about the elimination of variance through greater efficiencies.

Over the past year, working with ecosystem development in diverse industries – laundry (Haier’s Internet of Clothing), food (Haier’s Internet of Food), hospital functions (Haier’s Smart Hospital Internet), recreational vehicles (Haier’s RV ecosystem), and ESports (Copenhagen’s ESports Ecosystem1) – has led to a number of insights into the role of ecosystems as an engine for innovation and learning and the managerial choices that might be necessary to do this in an effective manner. The purpose of this paper is to present several of these insights and discuss them in a way that moves ecosystem engagement from the abstract into a more tangible form.

A key element is the belief that as we move more into the unknown (as opposed to the uncertain2, we need to be able to learn better and more effectively than in the past, and to be able to do this at scale. John Hagel has spoken about the move from mastering “scalable efficiency” to mastering “scalable learning.”3 My conclusion from the work described below is that ecosystems can be a source of abundant innovative ideas and scalable learning, but they are not necessarily easy to see by experienced managers, and once they are identified they need to be engaged with in a manner different from how value chains of the past were managed. Finally, engaging with ecosystems in a serious manner inevitably involves profound changes in your existing organization architecture and management style.

Ecosystems are not mere super-charged, muscular value chains.

As is so often the case, the present heightened increase in interest in ecosystems is occurring under the watch of a corporate leadership generation that was born and bred in a prior era that prized value chain ascendancy. While value chains have been extraordinarily useful as a recognized strategic concept for several decades, they are not ecosystems and, more dangerously, a familiarity with value chains can suggest conceptual limitations that could be constraining as we move into a less predictable future. Understanding the difference between value chains and ecosystems is vitally important if we are to be able to fully appreciate the opportunities inherent in ecosystems, especially as a means of achieving more powerful learning.

Michael Porter first introduced the concept of value chains as part of his approach to industry analysis. The linearity and sequentiality of Porter’s value chains are well represented in the horizontal axis that links Supplier Power to Customer Power in his five forces framework. More than a map, this axis is a mindset that all too often holds value chain partners frozen in place regarding present skill sets and future possibilities. Linear and sequential also inevitably imposes a “convoy reality” on its members; you move at the speed and imagination of the slowest member. Learning and sharing of ideas and insights, fast response and adaptability, are rendered irrelevant by the convoy mentality, which prefers reliability over innovation and prejudges all contributions by the position in the chain that the member presently maintains.

Ecosystems, on the other hand, are anything but linear or sequential and as a result can offer the energy and surprise of non-linearity; done right, they have the power to liberate their members from stereotyping based on chain-rigidity. This is no longer about the regularity of convoys as much as it is about spontaneity and daring. No one link in the chain is necessarily locked into place, nor is it dependent upon the multiple approvals of other, often uninformed and disinterested fellow chain members. Ecosystems offer surprise over reliance. I have repeatedly marveled that the very first microenterprise to be born out of Haier’s Internet of Clothing was neither an extension, nor an adjacency of present offerings or markets, but a rather unlikely venture into the laundering of exotic fabrics, such as scarves and high-end bags. No value chain-dominated management team would ever have made such a strategic arabesque, but fast-moving ecosystem partnerships sensed the opportunity and made it a natural choice for an experiment.

Value chains and ecosystems are completely different. They require different approaches to engage, and different roles to play. Power, be it commercial, technical or financial, dominates value chains; nothing could be less desirable in ecosystems if we are to entertain the abundant possibilities that are inherent in surprise relationships. Value chains prize efficiency and reliability; ecosystems flourish on serendipity and offer spontaneity. Value chains fulfill promises; ecosystems flirt with possibilities. Value chains are about delivery and performance; ecosystems are about dreams and the unthinkable.

For me, one of the early key lessons of the ecosystem era is that for innovation and learning, value chains suffer from being legacy constructs that tend to constrain rather than liberate the firm, while ecosystems celebrate spontaneity and responsiveness and ultimately have the power to take the firm in directions it would have never thought about on its own.

Ecosystems often exist in plain view but are difficult to see.

Many of the ecosystems we are working with were already in place for some time without being recognized for the potential that they offered to generate ideas. The basic foundational elements of the Copenhagen ESports ecosystem existed in the same small part of Skåne, Sweden and Denmark for many years without feeling or acting like an ecosystem. They already had in place major players in game design and ESports was well represented by Astralis, the world champion Counterstrike team. What was missing was a catalytic event or actor that could give this community of fellow travelers a real identity. Ultimately, it was a common problem of talent attraction to the region, which resonated with the Copenhagen municipality’s desire to build strength in coding and software design, that led to an intervention by Louise Juhl, director of marketing and communication at Copenhagen Capacity, whose mission is for the greater Copenhagen region to “become a leading international hub for talents, investments and knowledge capable of competing with the most successful metropolises in Europe.” The efforts of her and her team succeeded in encouraging the disparate members of what would eventually become an ecosystem, to think of themselves as a group defined by more than simply geographic proximity. This being said, it should also be acknowledged that maximum fulfillment of an ecosystem’s promise is the acceptance that no one player can ever be allowed to dominate an ecosystem; to do so is the first step on the slippery slope to transforming an ecosystem into a value chain.

In 2004, Marco Iansiti and Roy Levien wrote a path-breaking book entitled The Keystone Advantage: What the New Dynamics of Business Ecosystems Mean for Strategy, Innovation, and Sustainability.4 They proclaimed: “Strategy is becoming, to an increasing extent, the art of managing assets that one does not own,” and argued that every value chain/ecosystem has roles that need to be played, including that of the keystone, who “shape and coordinate the ecosystem, largely by the dissemination of platforms that form a foundation for ecosystem innovation and operations (such as eBay, IBM and Microsoft).” What we have seen in our research with ecosystems is that “managing assets that one does not own” is probably too strong a sentiment, and that, instead, ecosystems respond best to light touches. Having a keystone might well be an advantage, but only if they don’t use it to dominate their partners. An ecosystem’s members are engaged rather than conscripted or contracted, and they would probably object to the notion of someone else managing their assets; in fact, we think that ecosystems work best the closer they get to “centerless.” In this regard, Iansiti and Levien recognized that “the health of an ecosystem is appraised by the extent to which the ecosystem as a whole is durably creating opportunities for each of its domains.” This implies a lot of self-determination among members, each of which we find characterized by a unique objective function that is normally oblique to all others. One way to think of successful ecosystems is as a collection of independent, and often ornery partners, occasionally working in concert for the benefit of some, or even all, while always in pursuit of their own individually selfish objectives.

With centerless ecosystems as the goal, it is not far-fetched to describe ecosystems as galaxies of networks,5 where each member sees itself in a relationship web with a unique set of fellow actors. As a result, centerless ecosystems lead to an infinite number of polycentric ecosystems, existing in parallel and each defined, or imagined, by one or more actors, without the others even recognizing these different expressions of potential; there is never just one ecosystem in play at one time. As a result, the role of an ecosystem animateur, such as Louise Juhl, who could see what others could not, and encourage even a slight bit of coordination, becomes extremely important in the recognizing, legitimizing and orchestration of what could ultimately lead to durable, yet fragile partnerships.

What does it take to make spontaneity a competitive differentiator?

The potential magic of ecosystems is realized in the relationships between organizations, not necessarily forever, but for the moment at hand. Spontaneity, the ability to seize a passing collision of new ideas and turn it into a customer experience that has never been thought of before, is something that has traditionally never occurred to managers, either because they have not been pursuing unique customer experiences or because their organizations have prevented them from acting on such a notion, had it ever occurred. Such relationships are profoundly different from what is found in the middle of an industry S-curve where commoditization is first appearing, where everything is familiar, where costs and efficiency are the only objectives, and where repetition rules supreme.

But, times are changing. Zhang Ruimin, the CEO of Haier, speaks about the company formerly being a “solutions” provider, selling products to fulfill customer needs; solving their problems. Today, however, Haier offers those solutions as part of a more complex set of unique customer experiences. The solutions (washing machines, refrigerators, etc.) alone are no longer enough; refrigerators without the ability to search recipes, or track food provenance, for each and every customer, are merely “dumb” solutions. To remain relevant in the lives of sophisticated customers, Haier needs the capacity to enter into any conversation, at any time, and this is something that most traditional organizations are not prepared for; Haier needs spontaneity in the way in which it engages with customers and learns with any ecosystem partner, and this means a new type of organization.

One of the most interesting outcomes of Haier’s latest organizational transformation6 is how Haier has made it easier to rapidly adjust to the situation at hand, to be spontaneous. This is the outcome of several managerial choices, none of which are intuitively obvious in a world that has long been concerned with predictable mass production, price competition driving efficiency as the competitive differentiator, and variance reduction as the overall objective. These new managerial choices include:

Betting on many unknowns, rather than predicting the right one

Haier believes that it will remain relevant in the future, which is unknown (because it has never had to serve smart homes before), not because it will be better at predicting likely outcomes, but because it will have placed more bets on possible outcomes than anyone else; its secret to success will be that it winds up being at the right place, at the right time, because it took more chances. Yes, this puts Darwin in the Chief Strategy Officer’s seat, but what more could strategy really be when you can’t predict, extrapolate, or assume?

Haier has made it easier for its employees to place bets on the future

Who best to trust regarding an unknown future: veteran employees who have grown up with their market’s transformation journey, and who are close to the customer, or C-suite strategists, fresh out of business school who have good models? For more than 35 years, Haier has consistently trusted the knowledge of its employees to move the organization forward. At this point, when the advent of the smart home threatens to change much about what we know about home appliances, who better to steer the company into the future than the very employees who have established “zero-distance” between themselves and their customers? They may not have all the right answers, but they are so close to the market that they more than likely have the right questions.

Haier has also made it easy for those same employees to launch new ideas. Only three colleagues are required to endorse a new idea, and invest their own money, in order to propose a possible new start-up (microenterprise). If the logic of the proposal makes sense, they are off and running, and these new founders of the microenterprise are given responsibility for hiring, big decisions, and distributing rewards. Along with the “zero distance to the customer,” there is also “zero approvals” in operating decisions, and so there should be “zero delays” in responsiveness.

Bet on advantages of smallness

Microenterprises are essentially startups on the interface between the unknown and Haier’s traditional large organizational advantages. Typically, no more than ten people in size, they are able to adjust quickly to changes in the world around them, act in an end to end manner with everyone involved, and are too small to ever believe that they can do everything themselves, especially when there is an admittedly high mortality rate. One way to think of them is as provocations of an unknown future, trying something, many things, and seeing what works; they are always “— on the edge — in between the known and the unknown and you have to keep pushing it towards the unknown otherwise it and you die.”7 The words are those of Steve Lacy, a jazz musician, who knew quite a bit about improvisation, but the spirit is very much that of the microenterprises we have watched.

Redesign the mother ship so that scalable learning takes place

What really drives Haier’s ecosystem-enabling organization8 , however, is neither the microenterprises, which move fast across an ecosystem to suggest relationships that might lead to novel and delightful customer experiences, nor the platforms upon which these microenterprises rest. Instead, it is what Marshall Meyer calls a symbiotic relationship that creates an essential dynamic between these two key architectural elements, so imperative if an ecosystem engagement is to work. Part of this, Art Kleiner has suggested, is that the platforms, which provide launching pads for microenterprises that share a common affinity characteristic (mostly markets such as hospitals, clothing, or food), can act as mediators between the chaotic Brownian motion of the outside ecosystem, and the more regulated cadence of life within a much larger, complex organization. It would be difficult, if not impossible, for conversations to move from the wild frontier of Haier’s ecosystem partners directly into the organization’s conference rooms. At the same time, platforms can also serve as energy absorbers, so that the rough edges in ecosystem relationships are less stressful. Finally, having such platforms that are constantly in touch with a wide range of microenterprise ecosystem interactions across an industry should provide an excellent opportunity for “scalable learning.”

So, in conclusion, here are three simple thoughts regarding ecosystems and their potential:

  1. They are so much more than value chains;
  2. They are already more or less in place, needing a catalytic agent to animate them; and
  3. Engaging with an ecosystem requires speed and This, in turn, needs management to apply a light touch, reassess the suitability of its architecture, and be prepared to lose control.

Bill Fischer is a Professor of Innovation Management at IMD. He cofounded and codirects the IMD programme on Driving Strategic Innovation, in cooperation with the Sloan School of Management at MIT. His books include: Reinventing Giants: How Chinese Global Competitor Haier has Changed the Way that Big Companies Transform (with Umberto Lago & Fang Liu, 2013), The Idea Hunter (with Andy Boynton, 2011), and Virtuoso Teams (with Andy Boynton, 2005). He is a member of the Thinkers50 Hall of Fame.

This paper is the result of a multi-year consulting relationship with the Haier Culture Platform, Haier’s RDHY Model Research Institute, and a research partnership with Simone Cicero. It has also benefited from case research with Anouk Lavoie on Copenhagen’s Esports ecosystem. I am indebted to all for any insights I may have gained, and I accept all errors as my own.


1 Anouk Lavoie and Bill Fischer, “How Copenhagen launched an Esports ecosystem,” IMD Case IMD-7-2102.

2 Bill Fischer, “The unknown is not uncertain,” com, May 5, 2016: the-unknown-is-not-uncertain/

3 John Hagel, “Scalable learning in an exponential world,” Edge Perspectives, October 4, 2016:

4 Marco Iansiti and Roy Levien, The Keystone Advantage: What the New Dynamics of Business Ecosystems Mean for Strategy, Innovation, and Sustainability, Harvard Business School Press, 2004.

5 Greg Satell has used the term “networks of networks,” in “How to build an ecosystem strategy,” com, August 3, 2019.

6 More than three decades of prior transformations are addressed in Bill Fischer, Umberto Lago and Fang Liu, Reinventing Giants, Jossey-Bass, 2013.

7 American soprano saxophonist Steve Lacy, quoted in Derek Bailey, Improvisation, Da Capo, 1992, p. 54.

8 This is Simone Cicero’s term and I am continually impressed by its aptness.