Müller-Stewens & Stonig (2020) 1
Ecosystems and Platforms:
Towards a Shared Understanding (Version 2.0)
Günter Müller-Stewens and Joachim Stonig
Working Paper, University of St. Gallen, March 24, 2020
Corresponding author’s email: firstname.lastname@example.org
This is an updated version of: Müller-Stewens G./Stonig, J. (2019): Business Ecosystems and Platforms: Towards
a Shared Understanding. Die Unternehmung, Special Issue “Business Ecosystems”, Vol. 73, Nr. 4, 381-386.
The term ecosystem is currently attracting enormous interest among managers, consultants and
researchers. Unfortunately, in some areas it has also become a fashionable "buzzword" used in
an almost inflationary way. As a result, our understanding of what exactly an ecosystem is—
and what it is not—in a business context has expanded considerably as well. And this process
of consolidating and condensing our understanding of ecosystems is continuing, which is why
we have published this update of the first version of this article. We see this document as a basis
for ongoing discussion and therefore welcome any thoughts and feedback.
Our aim is to improve our common understanding of the key terms used in the context of
Ecosystems are an organizational form that enables coordination between independent
but complementary players to provide a shared customer value proposition, thus creating
Based on this definition, the following characteristics of an ecosystem emerge:
• “organizational form”: An ecosystem is an organizational form that enables and structures
economic transactions between different players and is therefore an entrepreneurial entity
on its own.
This does not necessarily mean that an ecosystem must act in a profit-oriented
manner – primarily socially oriented value propositions can also be provided by an
• “provide a shared customer value proposition”: The basis of the ecosystem is a value
proposition shared by the members of the ecosystem that addresses a specific customer
This customer value proposition represents the mission of the ecosystem, as it defines
its purpose. Ecosystems are therefore configured on the basis of a customer need to be
A special feature of ecosystems is that they usually serve very complex customer needs
through the coordinated interaction of the various ecosystem members. The value
proposition of an ecosystem offers a solution to a customer problem, drawing on an
integrated and comprehensive offering building on the contributions of different players,
but at the same time has a high degree of user-friendliness (convenience).
This distinguishes ecosystems, for example, from “virtual organizations”, which are
configured more from the point of view of firms, such as a group of industrial companies
that have expanded their spread of offerings through a network.
In the same way, a network
of stakeholders is usually configurated from the perspective of the company and does not
have to be an ecosystem, because an ecosystem is configurated outside-in from the
perspective of the customer’s demand.
Müller-Stewens & Stonig (2020) 2
• “enables coordination”: Performance in an ecosystem is based on the possibility and ability
of breaking down a complex customer need into independent activity modules in such a way
that these can function independently, but at the same time interact in an organizationally
and technologically coordinated and integrated manner. The activities on which the value
proposition is based on are thus configured
in a new way led by the customer need –
sometimes spanning industry boundaries.
The coordination of activities in an ecosystem is managed by an orchestrator that assumes
a central role for the players cooperating in the ecosystem (therefore sometimes also referred
to as a lead organization). The orchestrator, for example, ensures the development of the
common value proposition and defines the rules of interaction (via technological interfaces,
monetization mechanisms, etc.). The orchestrator thus exerts a certain degree of control
over the ecosystem. The role of the orchestrator is not always assumed by one single
organization. Often there is a technical operator as well as an owner of the ecosystem – as
in the case of smartphones, where some companies control the software platform of the
ecosystem, but the technical infrastructure is produced by device manufacturers like
• “complementary”: The activities underlying an ecosystem’s value proposition are not fully
controlled and executed by the orchestrator. Many of them are performed by independent
complementors. The term complementor (in contrast to terms like ecosystem partner)
emphasizes that the components in the ecosystem are complementary, i.e., without them the
ecosystem does not function, functions worse, or is incomplete. The complementors are
selected and coordinated by the orchestrator, such that their offerings fit together in a
modular way. To achieve these complementarities, the complementors have to co-specialize
their activities to the ecosystem. In some ecosystems, complementors of the same activity
compete with each other, which creates redundancies. However, value creation in an
ecosystem is not always based on having as many complementors as possible; success often
depends on the availability of the complementors that are essential from the customer's point
of view. The interaction between customers and complementors is direct, the orchestrator
only facilitates these interactions. However, this does not rule out the possibility that the
orchestrator itself may offer certain complements. For the orchestrator it is an important
strategic question which parts of the value proposition it wants to address via
complementors and which ones it chooses to integrate internally.
The users or customers of an ecosystem are usually not just passive consumers. In order to
fulfill an ecosystem’s complex value proposition, users can be highly involved and interact
intensively with the ecosystem, e.g., by delegating responsibility or providing data. They
often take on the role of complementors themselves (for instance, by creating content). Just
like complementors, users therefore also contribute to the ecosystem value.
• “independent players”: The players involved join the ecosystem on their own initiative and
can leave it again of their own accord, taking possible contracts into account. Of course,
there may be implicit constraints, but there is still freedom of choice. This is one of the
differences to diversified corporations to which the subsidiaries are under full control of the
In contrast to more vertically organized value chains (“pipelines”), ecosystems are based on
horizontal, network-like cooperation in which the participants meet on eye level. Thus, there
is no tendency towards hierarchical subordination. Since not all details of the cooperation
can be specified using contracts, the establishment of an underlying basis of long-term trust
between the participants is important. This demands a considerable degree of collaboration
competence from the participating organizations. If this is not forthcoming, it will not be
possible to attract suitable complementors for one's own ecosystem and keep them in the
Müller-Stewens & Stonig (2020) 3
system. This means that the ability to interact in partnerships is critical for the ecosystem
participants, especially for orchestrators.
• “creating added value”: The economic goal of an ecosystem is that the value of the
ecosystem is greater than the sum of its parts (the activities of orchestrators and
complementors), i.e., an ecosystem surplus is achieved. This surplus is generated by various
value drivers, in particular network effects and learning effects. Network effects occur when
a larger or broader range of complementary partners has a positive influence on customers'
willingness to join and pay. Learning effects mean that the ecosystem can offer a better
value proposition through a deeper and longer interaction between the different participants,
which in turn positively influences willingness to pay.
This added value can manifest itself as financial profit for the players. However, it is not
certain that the orchestrator and complementors will appropriate this added value; it can
also be the customers (e.g., in the case of Uber, which is currently not making a profit
despite an attractive customer offering). The definition of the ecosystem boundaries is also
crucial for measuring added value. For example, a bike sharing system, as such, cannot
create financial added value, but it can indirectly create added value for a city's mobility
ecosystem through spill-over effects.
In the context of ecosystems, there is often talk of "platforms" or "platform-based business
models". It should be noted here that not every platform-based business model is also an
ecosystem. Conversely, not every business model of an ecosystem is technologically
implemented via a platform, even though this is frequently the case (as platforms considerably
accelerate the scalability of the business model and thus also lead to the realization of network
Platforms are technological infrastructures connecting players and objects.
• “technological infrastructure”: In its simplest form, a platform is nothing more than a
technological foundation that brings together different activities (of players and objects) in
a coordinated way. The field of application is extremely broad and can range from a
motherboard in a computer, through a dating website, up to a platform construction kit in
the automotive industry. New technologies, in particular the internet, have led to a much
higher prevalence of digital platforms in all domains of business.
• “connecting”: A platform must create the conditions for platform participants to interact
efficiently (cheap, fast, convenient, seamless etc.) together. For this purpose, the platform
operator defines interfaces that ensure that the various activities of all players mesh together
as seamlessly as possible. However, networking is not only based on an efficient
technological infrastructure (hardware and software), but also on social norms and rules of
interaction (e.g., entry and exit rules, conflict escalation mechanisms, revenue mechanics
and calculation modes, etc.).
Digital platforms are particularly suitable for the implementation of ecosystems because
many elements of value creation processes can now be networked via digital interfaces,
which significantly extends the reach of platforms. This makes it easier for ecosystems to
coordinate participants beyond organizational and industry boundaries and significantly
reduces transaction costs between those involved. IT platforms have thus become a key
enabler of many of today's ecosystems.
The activities connected via the platform through technological interfaces become modules
that can be easily exchanged and changed without affecting the entire platform. This results
Müller-Stewens & Stonig (2020) 4
in a platform architecture, usually with a stable platform core and flexible connected
With regard to the networking and openness of the platform in relation to the modules, a
distinction is made between one-sided and two-sided platforms. Two (or more) sided
platforms bring together the modules of different sides, e.g., providers and users. A classic
example here is eBay or a dating platform like Tinder. The presence on one side has an
effect on the other side, which is not completely internalized by the participants.
contrast, there are one-sided platforms where platform participants do not interact with
another side – like a supply chain platform in the automotive industry, where the OEM only
interacts with one side: the supplier.
• “players”: In its simplest form of connecting players, a platform is a virtual electronic
marketplace. The players and users of a platform are then the sellers and their
buyers/customers. For some platforms, like social networking sites, players can take both
roles, depending on the transaction. Here it is irrelevant whether networking takes place at
the consumer-to-consumer (C2C), peer-to-peer (P2P), business-to-consumer (B2C) or
business-to-business (B2B) level.
The platform assumes a mediating function between the players. Information that enables
more efficient networking is therefore critical to the success of the platform. Bringing
together supply and demand gives rise to transactions that generate data. Platforms that
have a larger number of participants and a higher volume of usage data are usually more
successful, provided they can make sense of the data (for example through algorithmic
pattern recognition, machine learning, etc.). A platform can also selectively pass on data to
participants, e.g., to developers who receive access to user data, advance information on
interface changes or shares of sales from the use of the app.
• “objects”: Platforms can network not only players, but also objects. In industrial platforms,
for instance, the Internet of Things (IoT) forms the technological basis for networking.
Physical products, such as machines or semi-finished products, become information carriers
that generate data from built-in sensors in order to control processes via the platform. One
step further, however, they may also be able to initiate follow-up activities in production
processes independently or automatically (smart devices). For system manufacturers (such
as lift systems), communication with the systems after delivery can also be ensured in this
way, e.g., in order to identify maintenance requirements at an early stage before there is a
defect (predictive maintenance).
Platform-based business models are mechanisms building on platforms for value creation
and value distribution.
• “mechanisms building on platforms”: Digital platforms enable new business models with
unprecedented scalability and scope. Thanks to extensive standardization and automation
of the transaction process between the participants, only very low transaction costs are
incurred. The computing capacity required is not a real bottleneck, so the marginal costs of
growth are often close to zero. The essential question is to what extent investors are prepared
to finance the investment costs of the first few years associated with high negative cash
flows. And as we know in the meantime, not all “winners” in a platform market are
profitable in the long run.
• “value creation”: If the focus of the platform is primarily on establishing transactions, then
value arises here (1) through the transactions of the players involved, which are brought
Müller-Stewens & Stonig (2020) 5
together, and (2) through the provision of information that improves the quality of
transactions, for instance large scale data analytics.
The platform can also generate additional value, e.g., (3) through services to support the
execution of transactions (such as payment processing or mutual evaluation of players), and
(4) by providing and safeguarding the legal and regulatory framework conditions.
A key aspect of value generation (for multi-sided platforms) are the network effects to be
expected with the use of a platform. This means that with a growing number of participants,
the benefit for the players on both sides of the platform increases proportionally (or even
exponentially). This results from feedback effects: The more transactions there are, the more
attractive it is to participate—whether because of greater choice or higher availability. This
means that the greater the number of participants on the platform and the more data
generated, the greater the value of the platform. This applies both to network effects among
users of the same side (same-side effect) and to network effects between different sides,
because the platform is all the more attractive for providers the more buyers there are and
vice versa (cross-side effect). The network effects, however, only materialize above a
critical mass of users and providers.
Network effects can be enhanced by the skillful use of data. In the case of transactions on
the platform, the players involved generate data which they (consciously or unconsciously)
make available to the operator of the platform. Additional value is created if knowledge
about user behavior is gained from this transaction data and passed on to the providers, as
appropriate. Using algorithms, often based on artificial intelligence methods, tailored offers
and recommendations can be created for individual customers.
For example, a fintech company can use a platform to submit personalized investment
proposals to the user via its own app after linking to the user's bank account. As bank data
also provides information on other preferences and behaviors, personalized offerings
beyond banking (e.g., mobility services) can be made using additional apps based on the
available user data. Often users can also download such additional functions via a "store".
This can also take place in a form in which the user does not directly recognize that the offer
comes from third parties (white label solution).
If it is possible to make the platform attractive for particularly efficient providers, this could
open up great potential for innovation. Companies such as Microsoft, Google and Apple
hold large developer conferences every year. At the Apple Worldwide Developers
Conference (WWDC), for example, an outlook will be given on the next version of the
operating systems for Apple devices. This should motivate them to develop innovative new
functions for the devices. In certain regions, such as Silicon Valley, there are real innovation
clusters of such developers.
• “value distribution”: The distribution of the value creation of a platform-based business
model takes place between the platform and the participating sides on the one hand, and
between competing platforms, on the other. The former depends on the distribution of
power between the different sides, which influences the collection of fees and the
distribution of data. The possibility of players being active on several platforms at the same
time (multi-homing) or circumventing them plays a role in the distribution of power. The
latter is based on the results of platform competition. The presence of network effects often
leads to monopolization (or at least oligopolization) of global platform markets (winner-
Müller-Stewens & Stonig (2020) 6
If one follows the above explanations, it is easy to see that ecosystems and platforms have many
things in common, but are still not the same. Ecosystems are an inter-company organizational
form that addresses a value proposition; platforms are a technical infrastructure that can be
used, but does not have to be used, for the operational implementation of ecosystems. An
ecosystem can, for example, consist of several (nested) platforms.
The emergence of ecosystems gives rise to a new level of competition, where multiple
orchestrators compete to best serve a complex value proposition. Often, ecosystem competition
is not only a competition for customers (i.e., market share), but also for complementors.
Ecosystem orchestrators have to build competitive advantage at the ecosystem level and
generate high ecosystem surplus. However, besides generating high ecosystem surplus, each
participant in the ecosystem (the orchestrator, complementors, and customers) also has to be
able to appropriate a sufficient share of value. Finding the appropriate strategy for competition
among and within ecosystem will be a major challenge for many companies in the years to
Adner, R. (2017): Ecosystem as Structure: An Actionable Construct for Strategy, in: Journal of
Management, Vol. 43, No. 1, p. 39-58.
Baldwin, C. Y./Woodard, C. J. (2009): The Architecture of Platforms: A Unified View, in
Gawer, A. (ed.) Platforms, Markets and Innovation. Cheltenham, U.K. and Northampton,
Gawer, A. (2014): Bridging differing perspectives on technological platforms: Toward an
integrative framework, in: Research Policy, Vol. 43, No. 7, p. 1239-1249.
Jacobides, M. G./Cennamo, C./Gawer, A. (2018): Towards a theory of ecosystems, in: Strategic
Management Journal, Vol. 39, No. 8, p. 2255-2276.
Kapoor, R. (2018): Ecosystems: broadening the locus of value creation, in: Journal of
Organization Design, Vol. 7, No. 1, p. 12.
Nambisan, S./Sawhney, M. (2011): Orchestration Processes in Network-Centric Innovation:
Evidence From the Field, in: Academy of Management Perspectives, Vol. 25, No. 3, p.
Rochet, J.-C./Tirole, J. (2006): Two-sided markets: a progress report, in: RAND Journal of
Economics, Vol. 37, No. 3, p. 645-667.
Wareham, J./Fox, P. B./Giner, J. L. C. (2014): Technology Ecosystem Governance, in:
Organization Science, Vol. 25, No. 4, p. 1195-1215.
Zott, C./Amit, R./Massa, L. (2011): The business model: Recent developments and future
research, in: Journal of Management, Vol. 37, No. 4, p. 1019-1042.
Also see Jacobides/Cennamo/Gawer (2018).
Also see Kapoor (2018).
See Müller-Stewens (editor, 1997).
Also see Adner (2017).
Also see Nambisan/Sawhney (2011) or Wareham/Fox/Giner (2014).
Also see Baldwin/Woodard (2009) and Gawer (2014).
Also see Rochet/Tirole (2006). A participant from one side does not consider its own contribution to the value
of the platform when making a purchase decision. For example: A man who joins a dating platform takes only
the ratio of the supply of women to the cost of the platform as a basis for his purchase decision. However, he
does not take into account the fact that his purchase decision increases the value of the platform for the women's
side and so he should therefore pay less.
For the definition of business models, see Zott/Amit/Massa (2011).