Foreign market expansion of ecosystems: a process model

ABSTRACT

Ecosystems of actors that need to interact to create value for end users are becoming an integral part of firms’ strategic realities as they reach new markets. However, the phenomenon of ecosystem internationalisation has not been explored in comparison to its practical prevalence. We conceptualise this phenomenon as agent-led structuration on new markets. We build on the structural ecosystem literature and literature on internationalisation challenges to create our recursive process model of ecosystem creation on foreign markets. Creating ecosystems on foreign markets necessitates new ecosystem structures that are adjusted to value propositions by orchestrators through interaction with foreign market actors. The model explains how ecosystem orchestrators drive the international expansion of ecosystems through blueprinting and aligning mechanisms and how these mechanisms help manage liabilities of foreignness. The model accounts for the agency of involved actors, and with the help of the bottleneck concept bridges international business and ecosystem literatures.

KEYWORDS:

• Ecosystem structure
• ecosystem process
• ecosystem internationalisation
• bottlenecks
• liabilities of foreignness
• complementors

JEL CLASSIFICATION:

• F23
• M10
• M16
• O31
• O36

1. Introduction

Ecosystems are becoming an integral part of firms’ strategies and operational realities to deliver value to customers and facilitate growth (Freundt et al. 2023; Pidun, Reeves, and Schüssler 2020). A growing stream of literature has begun exploring the influence of ecosystem strategies, particularly digital platforms, on the internationalisation strategies of firms (e.g. Brouthers et al. 2022; Coviello, Kano, and Liesch 2017; Jones and Pitelis 2015; Monaghan, Tippmann, and Coviello 2020; Nambisan, Zahra, and Luo 2019; Ojala, Evers, and Rialp 2018). However, the inverse relationship has seen less attention within the ecosystem literature, i.e. how internationalisation affects ecosystems. This is surprising since ecosystems are considered open systems, meaning their structure is influenced by the characteristics of the market. The entering ecosystem may incur unfamiliarity, discrimination costs, and costs of weaker embeddedness because of cultural and institutional differences (Eden and Miller 2004; Lu, Ma, and Xie 2022; Zaheer 1995). In this paper, we address this limitation by explicitly considering how the structure of an ecosystem changes through the expansion to new markets. We draw on the structural perspective on ecosystems. From this perspective, ecosystems are structures of hierarchically independent actors that together realise ecosystem value propositions by performing different activities (Adner 2017; Kapoor 2018; Shipilov and Gawer 2020). These value propositions are integrated offerings, consisting of user-facing products and services that rely on complementary products and services to be provided by independent firms (Lingens, Seeholzer, and Gassmann 2022; Stonig, Schmid, and Müller-Stewens 2022; Thomas and Autio 2020), and are often enabled by digitalisation and modular architectures (Jacobides, Cennamo, and Gawer 2018; Nambisan, Zahra, and Luo 2019).¹ Because of this reliance on complementary products and services, the activities of independent ecosystem actors need to be aligned for value to reach the end-user (Adner 2017), but also exposes the ecosystem to bottlenecks. Bottlenecks are concentrations in the flow of activities that limit the ecosystem value creation and capture through technological or strategic inefficiencies, and to which there is a lack of alternatives (Baldwin 2018; Hannah and Eisenhardt 2018; Kapoor 2018; Masucci, Brusoni, and Cennamo 2020).

Recent developments in ecosystems literature have begun to explore how ecosystem structures are created and aligned (c.f., Ansari, Garud, and Kumaraswamy 2016; Dattée, Alexy, and Autio 2018; Snihur, Thomas, and Burgelman 2018). These accounts emphasise how ecosystem orchestrators actively create and change ecosystem structures through various mechanisms that add up to two distinct processes: blueprinting and aligning. Blueprinting is the process through which orchestrators create a blueprint of the structure for value creation and capture based on the underlying ecosystem value offering (Adner 2017; Dattée, Alexy, and Autio 2018). Aligning is the ongoing coordination of actors to perform these required activities. Previous results show that these are dynamic processes that result in continuous updating, rather than top-down specification of the blueprint and alignment of the structure (Adner 2017; Autio 2021; Dattée, Alexy, and Autio 2018; Snihur, Thomas, and Burgelman 2018). Technology has been shown to be a critical catalyst for changes in ecosystem structures (e.g. Adner and Kapoor 2010, 2016; Dattée, Alexy, and Autio 2018). But like the structure of any organisation, which are seen as open system, ecosystem structures must also be reconfigured when contexts change (Adner 2017; Brouthers et al. 2022), which is becoming an increasingly important question as more firms are relying on ecosystem creation with the aim of internationalising value propositions, i.e. providing and capturing value from different markets (Coviello, Kano, and Liesch 2017; Nambisan, Zahra, and Luo 2019; Ojala, Evers, and Rialp 2018; Tatarinov, Ambos, and Tschang 2022; Yildirim, Clarysse, and Wright 2022). Expanding to international markets can expose the structure to new bottlenecks which limit the ability to create value, and new markets have different complementors to which the ecosystem structure also needs to be reconfigured (Li et al. 2019; Nambisan and Luo 2021; Nambisan, Zahra, and Luo 2019; Tatarinov, Ambos, and Tschang 2022; Yildirim, Clarysse, and Wright 2022). For these reasons, we need to develop an understanding of how internationalisation affects the processes that create and shape the structure of an ecosystem.

Therefore, we argue that the effects of internationalisation on ecosystem structure is not currently understood. From the structural perspective on ecosystems (Adner 2017), ecosystem entry into new countries requires change to the ecosystem structure through the development of new ecosystem blueprints and alignment of new actors ‘to comply with local legislation, gain legitimacy, attract local complementors, and create value for customer/users’ (Brouthers et al. 2022, 2096; Nambisan and Luo 2021). Empirical work exemplifies that ecosystem structures need to be adapted to various markets (e.g. Adner 2013; Noel and Sovacool 2016; Pidun, Reeves, and Schüssler 2020; Tatarinov, Ambos, and Tschang 2022; Yildirim, Clarysse, and Wright 2022), but we do not yet have an understanding of how ecosystem actors use the blueprinting and aligning processes to structure an ecosystem to create value on a new market.

Thus, the purpose of this paper is to develop a conceptual process model of how ecosystem structures are affected by internationalisation. We begin by establishing the ontological foundation on which we conceptualise this process. The structural ecosystem literature is based on a structuration ontology (Adner 2017; Giddens 1984), which offers an integrated view on structure and agency. Structuration theory emphasises recursive interaction between the agency of actors and the social structures that enable and constrain this agency (Giddens 1984). By exercising their agency, actors both recreate the social structures that constrain and enable them and create new social structures. This ontological perspective is well-suited for developing process models, as it is compatible with a strong process view (Langley 2009). A strong process view is concerned with ‘the ongoing micro-processes that contribute to constituting and reproducing the organization as a stable entity’ (Langley 2009, 415; Tsoukas and Chia 2002). By drawing on a structuration ontology and strong process view, we ask the question: how do the blueprinting and aligning processes contribute to constituting and reproducing the ecosystem as a stable structure within a new market? We take a theory adaptation approach (Jaakkola 2020) and utilise insights from structural ecosystems and the literature on liabilities of foreignness (LOF; Lu, Ma, and Xie 2022; Zaheer 1995) to develop a process model of ecosystem creation on international markets (c.f., Cloutier and Langley 2020). We primarily draw on the conception of LOF as institutional differences (Lu, Ma, and Xie 2022), which we operationalise as institutional voids (c.f., Doh et al. 2017; Khanna and Palepu 1997) and the availability and ownership of both infrastructural and technological complementary assets (Cuervo-Cazurra, Maloney, and Manrakhan 2007; J.-F. Hennart 2009; Meyer et al. 2023; Teece 1986, 2018). From our structuration ontology, institutional differences correspond to the different social structures that enable and constrain individual and organisational agency and grant them access to resources. By informing the structural perspective on ecosystems with these international business perspectives, the scope of ecosystem literature is expanded to account for structural changes catalysed by the interaction between actors that draw on different social structures. Our starting point is that the goals (Van de Ven and Poole 1995) of the orchestrators is to create an activity structure suitable for the value proposition to reach the local market, and thus adjust the ecosystem blueprint and alignment. We make the base assumption that these goals are related to an existing value proposition, the development of which we do not include in the model to avoid the ‘chicken-and-egg’ problem common to ecosystems (c.f., Caillaud and Jullien 2003; Thomas and Ritala 2021).² For our process theorising we employ the narrative-based style (Cornelissen 2017) with focus on key conceptual linkages and set of mechanisms as the underlying storyline of our process model.

We contribute to ecosystem literature by developing a new model of ecosystem internationalisation that explores how international expansion requires the creation of new ecosystem structures. Our model highlights the processual nature of the structural ecosystem perspective. While we distinguish between blueprinting and aligning processes and associated mechanisms available to orchestrators during ecosystem creation, we were able to explicate the recursive, rather than linear, nature of interaction between orchestrators and complementors. By drawing on the liabilities of foreignness concept, we specify relevant challenges for ecosystems during international expansion, and explain how these become sources of bottlenecks within the ecosystem. We then explain the role played by each mechanism in resolving these bottlenecks. Thus, we show how institutional differences between countries do matter for ecosystems as open systems. Based on the conceptual developments in this paper we put forward several important directions for future research.

2. Defining ecosystems as structure and bottlenecks

Ecosystems are open organisational systems whose hierarchically independent members are interdependent in the realisation of focal value propositions or ecosystem outputs (Adner 2017; Gomes et al. 2021; Shipilov and Gawer 2020; Thomas and Autio 2020). This perspective takes a structuralist view (Adner 2017; Giddens 1984) and has emerged as the dominant way of conceptualising ecosystems (Gomes et al. 2021; Jacobides, Cennamo, and Gawer 2018). In this perspective, emphasis is placed on interdependencies as they relate to a specific value proposition or ecosystem output and how that value proposition is realised on the market (Adner 2006, 2017; Kapoor 2018; Shipilov and Gawer 2020; Thomas and Autio 2020). Here, ecosystems are defined as ‘the alignment structure of the multilateral set of partners who need to interact in order for a focal value proposition to materialise’ (Adner 2017, 42). Focal value propositions are integrated, user-facing offerings that rely on complementary products and services to be provided (Lingens, Seeholzer, and Gassmann 2022; Stonig, Schmid, and Müller-Stewens 2022; Thomas and Autio 2020). Examples include residential solar panels or electric vehicle battery exchange services in different markets (Hannah and Eisenhardt 2018; Pidun, Reeves, and Schüssler 2020) or automation within the maritime industry (Tsvetkova and Hellström 2022).

The ecosystem structures that surround these value propositions are composed of four elements: activities, positions, links, and actors (Adner 2017). The specific activities that are required to materialise a value proposition, and how these activities are positioned in relation to each other, which also specifies the links, are collectively referred to as the ecosystem blueprint (Adner 2017; Lingens, Miehé, and Gassmann 2021). It is the actors that perform these activities that need to be aligned to create ecosystem value (Adner 2017; Linde et al. 2021). Examples of activities include the supply of components and complements or the integration of these, whereas position could refer to whether these components and complements are integrated by an ecosystem actor, or by the end user themselves (Adner and Kapoor 2010; Jacobides, Cennamo, and Gawer 2018; Tatarinov, Ambos, and Tschang 2022). Modular design and complementarity are often what enables ecosystem value propositions to reach end users and deliver value (Jacobides, Cennamo, and Gawer 2018; Shipilov and Gawer 2020). The value creation is thus dependent on activities occurring outside the hierarchical control of any one firm (Altman, Nagle, and Tushman 2021; Jacobides, Cennamo, and Gawer 2018), and instead alignment of these activities is needed (Adner 2017). The focus on multilateral structures of activities, rather than the dyadic relationships between firms, means that many different actors can perform the same activity (Adner 2017), which helps ecosystems develop economies of scope (Shipilov and Gawer 2020). The lack of hierarchical control between the different ecosystem actors also means that the individual actors that decide to contribute to the value creation retain control of their own activities.

However, this separation of activities into an interdependent system is also exposed to constraints in the form of bottlenecks. Bottlenecks result from technological or strategic limitations where the flows of activities are concentrated and where there is a lack of suitable alternatives (Adner 2006; Baldwin 2018; Baldwin and Clark 2000; Kapoor 2018). Technological bottlenecks can emerge if important components or infrastructure is of low quality, for example batteries in electric vehicles, a national electrical grid, or the absence of a bridge crossing a river (Adner 2013; Baldwin 2018; Kapoor 2018). Strategic bottlenecks instead emerge when important activities are controlled by only one or a few actors or require unique capabilities or resources, for example Microsoft in the software ecosystem, Nikon in the semiconductor ecosystem, or the owner of the aforementioned bridge (Baldwin 2018; Kapoor 2018; Ojala, Evers, and Rialp 2018). To support our later theorising, it is important to highlight that bottlenecks exist at the system-level and can therefore emerge in both institutionally- and privately controlled activities in the system.

This also means bottlenecks exist in relative rather than absolute terms (Jacobides, Knudsen, and Augier 2006), and change over time as technologies develop and ownership rights shift (Baldwin 2018). Bottlenecks can limit the overall value creation potential of the entire ecosystem by stifling technological efficiency or by the creation of ‘convenient rules of the game’ by owners of strategic bottlenecks (Jacobides, Knudsen, and Augier 2006; emphasis in original;, p. 1209; Masucci, Brusoni, and Cennamo 2020). But strategic bottlenecks can also offer an attractive position for capturing value from the ecosystem output (Hannah and Eisenhardt 2018). Hannah and Eisenhardt (2018) explain how firms within the residential solar panel ecosystem captured greater value from ecosystems by holding bottleneck positions in financing and components and adjusting their activities to remain in these positions as bottlenecks shifted. However, bottlenecks can also emerge from the social structure of a market, which includes its institutions, or the lack thereof, and available resource endowments from which actors can draw.

Bottlenecks pose a challenge for realising the ecosystem value, since actors with an eye for the overall value of the ecosystem will want to resolve bottlenecks, while actors benefitting from the bottlenecks will want to maintain them. This may then require changes in the structure of the ecosystem, either by changing the blueprint or the alignment of actors (Hannah and Eisenhardt 2018; Kapoor 2018; Masucci, Brusoni, and Cennamo 2020).

As this review has revealed, ecosystem structures are not static constructs and the creation and change of such structures are agent-driven processes. The activities required for value propositions to materialise can change, as can the positions of activities and the links between them. Actors performing the activities can also change, and changes in any of these structural elements result in different structures. Structural outcomes like bottlenecks can also catalyse change by trying to resolve them. This means that blueprints are redefined and continuously developed as the surrounding conditions change, which subsequently requires realignment of actors to meet the new conditions (Adner 2017; Dattée, Alexy, and Autio 2018).

Ecosystem orchestrators (often labelled leaders, lead firms, hubs, keystones, or focal firms) are the agents that take the responsibility of designing ecosystem blueprints and exercise effort to align other actors to these blueprints (Dattée, Alexy, and Autio 2018; Foss, Schmidt, and Teece 2023; Lingens, Miehé, and Gassmann 2021). Orchestrators are either individual agents or collectives that are comprised of agents filling multiple roles within the ecosystem. In other words, ecosystem actors can orchestrate at the system level, while providing complementary assets within the system. What is important is that taking the role of orchestrator requires that the actors both have the capabilities to hold a system-level view of the ecosystem (e.g. Foss, Schmidt, and Teece 2023; Helfat and Raubitschek 2018) and have sufficient knowledge of the value proposition to design appropriate blueprints (Lingens, Miehé, and Gassmann 2021). Not every complementor has system-level benefits in mind or sufficient knowledge of the value proposition to orchestrate, but performing orchestration activities is not necessarily the purview of single agents. This is more common in platform ecosystems, where ownership of the platform often implies power and capabilities for orchestration.

3. Clarifying ecosystem creation processes

Since ecosystem structures are not static and agents create and respond to changes, we can deduce two key processes: blueprinting and aligning. The blueprinting process is undertaken by ecosystem orchestrators as they attempt to create a blueprint for value creation and capture based on the underlying ecosystem value offering, codified into activities, positions and links (Adner 2017; Dattée, Alexy, and Autio 2018; Linde et al. 2021; Thomas and Autio 2020). Aligning then refers to the ongoing coordination of actors to perform the activities as they emerge in the blueprint, induce commitment to the ecosystem and often involves resolving emergent structural issues (Adner 2017; Lingens, Miehé, and Gassmann 2021). Actors need specific capabilities, competencies, and resources to perform ecosystem activities, and can rarely perform several of the activities (Altman, Nagle, and Tushman 2021; Helfat and Raubitschek 2018). Because of this, the blueprint needs to be reconstituted according to the accessible actors within a market to achieve the goal of value creation and capture. But the characteristics of the market itself can also have an impact since ecosystems are open system, meaning their structures is also open to influence from contextual factors such as institutions and culture (Brouthers et al. 2022; Doh et al. 2017; Giddens 1984; J.-F. Hennart 2009). This will be expanded below but is important to keep in mind while deriving the ecosystem creation processes (Tables 1 and 2). The following two subsections will focus on each of these processes and their constituent mechanisms, paying special attention to the role of agency in each (Giddens 1984; Langley 2009; Van de Ven and Poole 1995).

Table 1. Mechanisms of the Blueprinting Process.

Mechanism and definition	References
Monitoring the interaction between the ecosystem structure and the environment	Adner (2017); Ansari et al., (2016); Dattée et al., (2018); Lu et al., (2022); Snihur et al., (2018); Yildirim et al., (2022)
Specifying the activities that need to be completed in the environment	Adner (2017); Baldwin and Clark (2000); Jacobides et al., (2018); Williamson and De Meyer (2012)
Creating a common vision for the ecosystem to converge on a common value creating path and facilitate the resolution of bottlenecks	Adner (2017); Foss et al., (2023); Thomas and Ritala (2021); Williamson and De Meyer (2012)

Table 2. Mechanisms of the Aligning Process.

Mechanism and definition	References
Incentivising commitment to the ecosystem by negotiating value distribution and knowledge sharing schemes	Dhanaraj and Parkhe (2006); Foss et al. (2023); Linde et al., (2021); Ritala et al., (2013); Williamson and De Meyer (2012)
Influencing the development of the ecosystem in line with the vision through revealing proprietary knowledge and signalling the vision	Dattée et al., (2018); Linde et al., (2021); Snihur et al., (2018); Thomas and Ritala (2021); Williamson and De Meyer (2012)
Stimulating investment in bottleneck activities to facilitate their resolution through negotiation, communication, negotiation, and sharing knowledge	Adner (2017); Baldwin (2018); Hannah and Eisenhardt (2018); Masucci et al., (2020)

3.1. The blueprinting process

This is the process by which ecosystem orchestrators clarify how value is intended to be created and captured by the actors participating in the ecosystem (Adner 2017; Dattée, Alexy, and Autio 2018). To create this blueprint, orchestrators need to rely on three distinct, yet parallel, mechanisms: monitoring the environment, specifying activities, and creating common vision, which we now detail.

To ensure that the blueprint will achieve the intended goal of creating value for a defined set of end users, orchestrators rely on monitoring the internal functioning of the ecosystem in interaction with its environment. Through monitoring, orchestrators can discover the need for complementary assets and potential partners that could perform the activities necessary to materialise the value proposition and gain an information advantage (Dattée, Alexy, and Autio 2018). Effective monitoring depends on orchestrators collecting various forms of feedback by directly interacting with actors in the environment, by observing their activities, and by considering the social structure in which the ecosystem is embedded (Ansari, Garud, and Kumaraswamy 2016; Dattée, Alexy, and Autio 2018; Snihur, Thomas, and Burgelman 2018). By monitoring, orchestrators can recognise gaps in the ecosystem structure, where additional activities are needed or recognise the need for changing their approach to aligning other actors (Adner 2017; Dattée, Alexy, and Autio 2018; Yildirim, Clarysse, and Wright 2022).

Recursively with monitoring, the orchestrators employ two other mechanisms of blueprinting. Specifying activities focuses on the structure and its potential modification as response to contextual factors/new market characteristics. Creating common vision is needed for the actors to converge on a promising value creating path.

Orchestrators begin specifying the activities that need to be performed for the value to be created (Adner 2017; Jacobides, Cennamo, and Gawer 2018). Because ecosystem value propositions require input from the surrounding environment, orchestrators need to be able to take a system level view (Baldwin and Clark 2000; Foss, Schmidt, and Teece 2023). By taking this system level view, orchestrators can further specify where these activities should be placed in relation to each other, meaning their positions, and what should flow between each position (Adner 2017; Baldwin and Clark 2000; Jacobides, Cennamo, and Gawer 2018; Williamson and De Meyer 2012). However, specifying activities is difficult task, especially if there are uncertainties regarding the underlying value, technology, or surrounding environment (Dattée, Alexy, and Autio 2018), any of which may lead to the emergence of technological or strategic bottlenecks (Baldwin 2018).

Creating a common vision involves imagining and ideating diverging potential futures for the ecosystem, with the purpose of converging on a promising value creating path for the ecosystem (Adner 2017; Dattée, Alexy, and Autio 2018; Foss, Schmidt, and Teece 2023; Thomas and Autio 2020). This common vision should convey the value the ecosystem creates for both partners and end users, and the roles of these actors within it to anchor the blueprint and facilitate the resolution of emergent bottlenecks (Foss, Schmidt, and Teece 2023). Orchestrators draw from the ongoing monitoring and specifying to facilitate convergence towards this common vision embedded within the social structure (Dattée, Alexy, and Autio 2018).

3.2. The aligning process

Achieving the required alignment of actors to realise the ecosystem value proposition is the goal of the aligning process. Aligning is focused on how orchestrators fill activity gaps in the blueprint through reciprocal interaction with actors in the environment (Adner 2017). Three things need to happen through the aligning process; ecosystem actors need to commit their contributions to the value proposition, the development trajectory of the ecosystem needs to be maintained, and potential bottlenecks need to be overcome. Orchestrators generally have three distinct non-contractual mechanisms in their toolbox when aligning hierarchically independent actors: incentivising commitment, influencing development, and stimulating investment. Although these mechanisms rely on similar means, their purposes for the materialisation of the ecosystem value proposition are different. In essence, all these mechanisms relate to the orchestrators ability to induce and maintain commitment to the ecosystem blueprint.

The first mechanism that orchestrators can utilise is incentivising commitment by negotiating value distribution and knowledge sharing schemes that ensure value capture by ecosystem members (Dhanaraj and Parkhe 2006; Linde et al. 2021; Williamson and De Meyer 2012). Sharing intellectual property and revealing knowledge affects the commitment decisions of other actors by reducing their uncertainty and providing value (Alexy, George, and Salter 2013; Masucci, Brusoni, and Cennamo 2020; Williamson and De Meyer 2012). Depending on the blueprint, orchestrators can achieve value capture through the creation of markets (e.g. Apple or Android ecosystems), which allows complementors to freely compete for their share of value capture. In other cases (e.g. smart city initiatives), the ecosystem orchestrators must make more explicit incentive schemes that prospective ecosystem members find agreeable (Foss, Schmidt, and Teece 2023; Linde et al. 2021; Ritala et al. 2013). These value distribution schemes may also need to change alongside changes in the blueprint or other external factors (Linde et al. 2021).

Orchestrators also need to influence the development of the ecosystem in the trajectory laid out by the vision (Dattée, Alexy, and Autio 2018; Snihur, Thomas, and Burgelman 2018). By revealing proprietary knowledge directly to established and/or potential ecosystem actors, orchestrators can influence the trajectory that align with the envisioned future development of the ecosystem (Alexy, George, and Salter 2013; Dattée, Alexy, and Autio 2018; Linde et al. 2021; Williamson and De Meyer 2012). The revealed proprietary knowledge limits the scope of potential trajectories and reduces the need for other actors to expend resources on their own knowledge creation (Dhanaraj and Parkhe 2006; Williamson and De Meyer 2012). Orchestrators can also use indirect communication of the vision and knowledge to signal trajectories that align with the vision for the ecosystem (Gurses and Ozcan 2015; Snihur, Thomas, and Burgelman 2018; Thomas and Ritala 2021).

Lastly, stimulating investment is a key mechanism related to the resolution of bottlenecks. Orchestrators can rely on their system-level view to highlight the emerging bottleneck to direct activity and stimulate investment in its resolution (Baldwin 2018; Masucci, Brusoni, and Cennamo 2020). Hence, stimulating investment is differentiated from incentivising commitment and influencing development by its purpose, rather than its means. Orchestrators can use the same means of negotiating, communicating, and sharing knowledge to direct and stimulate investment towards bottlenecks and their resolution. Orchestrators can create conditions for resolving technological bottlenecks (Baldwin 2018) by, for example, communicating bottlenecks as value capture opportunities to direct other actors to invest in its resolution (Hannah and Eisenhardt 2018), or resolve strategic bottlenecks by sharing bottleneck-related IP to encourage partner investment (Masucci, Brusoni, and Cennamo 2020).

When considering the blueprinting and aligning processes together, one helpful distinction is their orientation. The blueprinting process focuses inward, as orchestrators use feedback from external actors to update its own blueprint that enables value creation and capture from the ecosystem value proposition. The focus of aligning is outward for inducing desired behaviour of existing and potential ecosystem actors over whom there is no hierarchical control. We have previously explained that ecosystem structure is not static, thus the two processes have an inherent updating dynamic (Ansari, Garud, and Kumaraswamy 2016; Dattée, Alexy, and Autio 2018; Linde et al. 2021; Snihur, Thomas, and Burgelman 2018). Orchestrators can adjust their specification of activities to accommodate environmental contingencies or structural issues. They can also choose to update their approaches to aligning other actors by, for example, adjusting incentive structures or moving from indirect to direct influencing. The aligning mechanisms that the orchestrators choose to employ also provides opportunities to monitor the environment via its interactions with other actors (Dattée, Alexy, and Autio 2018).

4. Liabilities of foreignness and ecosystem expansion

By defining ecosystems as open systems (Adner 2017; Shipilov and Gawer 2020), we acknowledge the influence of the surrounding market and institutional environment on ecosystem structures. However, the differences between foreign markets are not explicitly treated by the ecosystem literature, so it follows that we do not understand how these differences affect the structure of ecosystems and how orchestrators can use the blueprinting and aligning mechanisms for restructuring. Hence, we draw on core international business literature on the liabilities of foreignness to enrich our process model of ecosystem expansion on international markets. LOF can be considered an umbrella concept that covers a wide variety of differences between contexts and the associated challenges and costs that internationalising firms incur because of these differences (Eden and Miller, 2004; Zaheer 1995). It has been used to refer to, among other things, differences in available resources between contexts, the perceived image of a country, and differences in both formal and informal institutions (Denk, Kaufmann, and Roesch 2012; Lu, Ma, and Xie 2022; Meyer et al. 2023). The liabilities faced by firm as they internationalise are contextual and have varied dynamics, depending on distance, familiarity, and legitimacy (Cozza et al. 2021; Lu, Ma, and Xie 2022; Zaheer 1995). Due to practical considerations, we are unable to fully treat this broad literature on foreignness and have chosen to delimit ourselves to certain aspects of foreign contexts to include in our process model. We pay particular attention to the role of institutional differences in the process of ecosystem expansion as these are a key driver of LOF (Lu, Ma, and Xie 2022). First, institutional voids, defined as non-functional, missing, or weak institutions (Khanna and Palepu 1997), can be considered a key difference between institutional settings (Doh et al. 2017). Institutional voids increase the costs associated with entering a foreign market as entering firms needs to address relevant voids (Doh et al. 2017). When value creating ecosystem activities rely on institutions enabling flows, the lack of such institutions constitutes a bottleneck that needs to be resolved. Recent empirical work has emphasised the consequences of weak or lacking institutions for the international expansion of ecosystems and how this impacts the strategies of orchestrators and complementors (e.g. Parente et al. 2019; Rong et al. 2015; Tatarinov, Ambos, and Tschang 2022; Yildirim, Clarysse, and Wright 2022). While these studies have addressed structural outcomes and MNE-centric strategies that rely on ecosystems, these have yet to be integrated into an ecosystem-centric process model. LOF are addressed by restructuring the ecosystem, which incurs costs (Brouthers et al. 2022; Nambisan and Luo 2021; Nambisan, Zahra, and Luo 2019). Thus, we draw on this growing body of work and integrate it into our process model, which grounds our discussion about the role of institutions in ecosystem expansion.

Second, institutional differences also have implications for the resource endowments within a context (c.f., Lu, Ma, and Xie 2022; Meyer et al. 2023), which impacts availability of complementary assets for the ecosystem (Cuervo-Cazurra, Maloney, and Manrakhan 2007; J.-F. Hennart 2009; Teece 1986, 2018). These complementary assets are under dispersed ownership which varies with the institutional arrangement of the specific context (c.f., J.-F. Hennart 2009; Li et al. 2019; Lu, Ma, and Xie 2022; Meyer et al. 2023; Teece 1986, 2018). As we have previously argued, the success of ecosystems depends on the alignment of appropriate complementary assets necessary to materialise the value proposition within a context. The dispersed ownership of complementary assets and quality of institutional resource endowments, therefore, create two separate forms of LOF that impact the alignment of the ecosystem. The differences in institutional systems result in different ownership structures of complementary assets (Cuervo-Cazurra, Maloney, and Manrakhan 2007; J.-F. Hennart 2009; Meyer et al. 2023), which from an ecosystem perspective results in different challenges related to the blueprinting and aligning of these actors. Institutional systems further impact contexts on a broader scale through the development, provision and upkeep of common resources, such as physical infrastructure, electrical grids, and digital infrastructure (c.f., Andrews et al., 2018; Meyer et al. 2023; Noel and Sovacool 2016; Ojala, Evers, and Rialp 2018).

These differences in institutional systems result in the emergence of diverse technical and strategic bottlenecks that need to be accounted for to successfully align the ecosystem (Baldwin 2018; Kapoor 2018; Ojala, Evers, and Rialp 2018). To reiterate, bottlenecks represent a major hurdle for materialising ecosystem value propositions by limiting the performance of certain activities or limiting flows between positions in the ecosystem structure. As a result, when expanding to different foreign contexts, ecosystem structures need to be adjusted through the blueprinting and aligning processes to accommodate and resolve the bottlenecks that emerge from institutional differences (c.f., Jacobides, Knudsen, and Augier 2006; Ojala, Evers, and Rialp 2018). These adjustments can vary in their broader implications for the ecosystem structure. For example, bottlenecks that emerge from institutional voids or infrastructural issues generally require the introduction of new activities into the ecosystem blueprint, whereas strategic bottlenecks, primarily related to the ownership of assets (c.f., Baldwin 2018), can be managed through adjustments in aligning mechanisms. This will be explored in more detail in the next section.

The integration of the literature on LOF with the ecosystem literature is anchored in our structuration ontology (Giddens 1984). Based off the assumption that institutional differences arise from the varying structure of the social systems within different contexts and its impact on agency (McPhee et al. 2013), it follows that ecosystem expansion into foreign contexts requires reconstitution of the ecosystem structure to account for the agency of local actors that emerges from their social context. The orchestrator can rely on the agency provided (and constrained) by the home context social structure to begin the reconstitution process, through which the ecosystem structure is blueprinted and aligned in accordance with the agency and constraints in the host context. The exercise of agency from both orchestrators and complementors implies a recursive and reciprocal process of mutual adjustment (c.f., Adner 2017; Linde et al. 2021; Stonig, Schmid, and Müller-Stewens 2022), where the orchestrators’ knowledgeability of the ecosystem value proposition, blueprint, and alignment in the home context is brought to bear together with the knowledgeability of complementors to accommodate institutional differences (Giddens 1984, 91, 191; McPhee et al., 2013, p. 76–78).

5. A process model of ecosystem international expansion

Next, as we use the narrative-based style, we provide the underlying storyline of our process model with which we specify how the set of blueprinting and aligning mechanisms can be utilised by the orchestrators to reconstitute the ecosystem structure on a foreign market, thereby answering our research question. Tables 1 and 2 define the key mechanisms in our process model which are further detailed in the text with contingent variation to strengthen their explanatory power (Cornelissen 2017).

Our process model depicts a recursive process of stabilising and restructuring an ecosystem within a new context (Cloutier and Langley 2020), rather than linearly describing the individual steps the orchestrators go through. Because of this, time becomes implicit in our model, but establishing the necessary relationships and making the necessary structural adjustments is a time-consuming process. This recursive process is initialised by the agency of orchestrators as they attempt to expand to new markets and recreate the ecosystem structure on that market. The process continues through interactions with local actors representing the pre-existing social structures on that market. These interactions then continue recursively over time as orchestrators identify and integrate insights into the ecosystem blueprint and adjust their aligning approach to exploit local complements and resolve emergent bottlenecks. We narratively trace these processes in this section, and have exemplified the connection between LOF, bottlenecks and empirical examples of mechanisms in play in Tables 3 and 4.

Table 3. Institutional differences, technology bottlenecks, and blueprinting mechanisms. Examples are drawn from published cases and explicated in more detail in the main text. ARM is a semiconductor and software design company (Rong et al., 2015). SR refers to screen reader software brought to the Turkish market (Yildirim, Clarysse, and Wright 2022). BP refers to Better Place, the now bankrupt battery replacement (Adner 2013; Noel and Sovacool 2016).

LOF from institutional differences	Institutional void Common resource endowment
Bottleneck	Institutional hinder of activities and flows Low quality infrastructure
Mechanisms during international expansion	To monitor institutional differences and associated bottlenecks within the host context and continuously monitor the functionality of the ecosystem	To specify the activities, positions, and links that comprise the ecosystem structure in the host context, and resolve bottlenecks in the ecosystem	Create common vision to guide various development trajectories for the ecosystem structure so to provide flexibility in blueprints and prevent potential bottlenecks
Example	Monitoring for relevant assets and demand for the value proposition (ARM) Monitor for infrastructure/energy supply (BP) Monitor for educational infrastructure (SR) Continuous monitoring of the results of choices made (ARM, SR) or failure to monitor (BP)	New activities specified: developed and published educational material (ARM) Integrated local OEMs (ARM) Specified distribution of activities according to environmental differences (BP) Utilized complementary assets by involving local entrepreneurs (SR)	A vision of a viable ecosystem in China (ARM) From economic viability to rapid deployment of the value proposition in other markets (BP) Flexible structure able to account for local entrepreneurs as complementors (SR)

Table 4. Institutional differences, strategic bottlenecks, and alignment mechanisms. Examples are drawn from published cases and explicated in more detail in the main text. ARM is a semiconductor and software design company (Rong et al., 2015). SR refers to screen reader software brought to the Turkish market (Yildirim, Clarysse, and Wright 2022). BP refers to Better Place, the now bankrupt battery replacement (Adner 2013; Noel and Sovacool 2016).

LOF from institutional differences	Different ownership structure
Bottleneck	Availability of complementary assets
Mechanism during international expansion	To incentivise the commitment of complementary asset owners to the ecosystem structure and materialisation of the ecosystem value proposition	To influence developments of complementary asset owners within the ecosystem towards the desired trajectory for the ecosystem structure	To stimulate investment in the resolution of bottlenecks caused by institutional voids, missing complementary assets, or lacking infrastructure necessary for the materialisation of the ecosystem value proposition
Examples	Sharing knowledge and IP (ARM) Offering attractive value capture for complementors through dealership rights (SR)	Educating students and potential complementors (ARM) Communicating the vision broadly (ARM) Framing the value proposition to fit each market (BP) Visible high-profile founder (BP)	Valuable bottleneck position for local entrepreneurs to fill (SR) Incubated local complementors to fill missing complementarities (ARM) Sharing IP to establish demand (ARM) Brought in global partners in the design of complementary assets to overcome local bottlenecks (BP)

The process begins with the decision to expand the value proposition to a new market. This is a decision made by ecosystem orchestrators, with the goal of creating value for end users and exploit the value proposition by designing and realising an ecosystem structure (Adner 2017; Dattée, Alexy, and Autio 2018; Foss, Schmidt, and Teece 2023).

Initially when entering a new market, ecosystem orchestrators need to monitor the host context being entered to identify relevant liabilities that will impact the ecosystem structure. Institutional differences are all potentially affecting the ecosystem structure. We argue that central among these are the availability of necessary infrastructure (Cuervo-Cazurra, Maloney, and Manrakhan 2007; Meyer et al. 2023), and institutional voids (Doh et al. 2017; Khanna and Palepu 1997) and access to complementary assets (J.-F. Hennart 2009; Jacobides, Cennamo, and Gawer 2018; Teece 1986, 2018). Complementary assets are central for the value creation of ecosystems (Jacobides, Cennamo, and Gawer 2018), but entering foreign contexts may limit the access to these assets through ownership, pre-existing contracts (c.f., J.-F. Hennart 2009), or the lack of available resources all together (c.f., Rong et al., 2015).

As monitoring reveals institutional differences, which may manifest in technological or strategic bottlenecks, orchestrators need to specify suitable activities, positions, and links to account for these differences. Host contexts have characteristics and features that fix certain activities and positions and bottleneck the links between activities. Considering these differences is crucial for re-constituting the ecosystem within the new market (Li et al. 2019; Nambisan and Luo 2021; Tatarinov, Ambos, and Tschang 2022). While many of the activities related to the core value proposition are likely to remain the same, their positions and links are likely to change as some activities will be performed by cross-border partners, and other performed by local partners (Tatarinov, Ambos, and Tschang 2022).

During the blueprinting process, orchestrators need to maintain a future orientation to remain flexible and adaptable to changing social structures. By being in the new environment, the ongoing interactions between the ecosystem and its surroundings impact the liabilities to which it is exposed (Lu, Ma, and Xie 2022). But the environment itself is also changing through processes not directly related to the ecosystem. Envisioning facilitates the creation of a vision to guide the ecosystem through these changes (Foss, Schmidt, and Teece 2023). A common vision for the ecosystem allows the blueprint to remain agile and adaptable to emergent bottlenecks and the alignment of actors.

In the published case by Rong et al. (2015), the process of ARM started with monitoring of the Chinese market, which revealed that few assets existed that could support its value proposition (Table 3). They also met limited demand from the local market as there were no partners to licence its technology (Rong et al., 2015). After ARM’s initial monitoring of the Chinese context, they specified the need to develop and publish educational material for engineering schools, and contract local OEMs and design firms. The specification of activities by ARM was driven by a vision of a functional ecosystem within the Chinese market, which it used to incubate many complementary asset owners. ARM continuously monitored the effects of their efforts to integrate their technology in the surrounding market.

The monitoring of the Danish and Israeli markets by Better Place identified substantial differences in the supporting infrastructure central to their value proposition (Table 3). Where Denmark had a stable electrical grid supported by wind, the cold weather strained the performance of their batteries, whereas the Israeli market had a grid dependent on foreign oil and worries of energy security were pervasive (Adner 2013; Noel and Sovacool 2016). In the two countries, Better Place specified the distribution of their charging stations to account for the adverse effect of the weather. However, Better Place failed to continue monitoring their integration and acceptance in the different markets, instead chose to extend their monitoring to other market, effectively stretching themselves too thin (Adner 2013; Noel and Sovacool 2016). This misguided monitoring by Better Place was driven by a change in vision from ‘economic viability’ to rapid global deployment of the value proposition (Adner 2013).

Lack of absorptive capacity within educational institutions in Turkey was identified as a limiting factor for the introduction of screen readers (Yildirim, Clarysse, and Wright 2022; Table 3). The orchestrators in this case specified the need to bridge their gap of substandard educational institutions by utilising complementary assets owned by local entrepreneurs. In this case, the common vision facilitated the creation of flexible blueprints that allowed local entrepreneurs to adapt their activities to match the institutional needs of the context. The monitoring of institutional voids transitioned into monitoring of how the solutions to these provided by the aligned local entrepreneurs were proceeding.

As entry into these various markets progressed, orchestrators needed to monitor the development of the ecosystem within the context, maintaining a view for emerging bottlenecks. The content of the three blueprinting mechanisms naturally differed, but the illustrative cases also differed in their monitoring approach, and how flexible the pursued common vision was.

As the blueprint is specified in accordance with the ongoing monitoring and envisioning, orchestrators concurrently need to initiate the process of aligning potential partners. This includes the decision of whether to align local market complementors or global complementors, which is a choice which will depend on the blueprint and bottlenecks emerging from institutional differences in the new context. Monitoring may reveal that orchestrators are unable to access the specified assets necessary to materialise the value proposition within the new context. This would require orchestrators to align partners from other markets, by providing the necessary incentives for international operations (Parente et al. 2019). But as mentioned above, the local market may have crucial complementarities that need to be integrated to reach the market (Li et al. 2019; Parente, Geleilate, and Rong 2018).

The primary aligning mechanism is incentivising commitment to the ecosystem via appropriate value distribution and sharing knowledge (Table 4). Within the new market, negotiation for value distribution needs to consider the institutional environment, demands from complementary asset owners, and the severity of the technological and strategic bottlenecks. Depending on the efficiency of local markets, orchestrators may prefer to rely on more explicit incentives to align other actors to the ecosystem. During negotiation, orchestrators can rely on key knowledge regarding the ecosystem value proposition and associated key technologies. Thus, they are often able to share intellectual property with local complementors to incentivise their commitment to the ecosystem.

Orchestrators may also need to influence other actors via communication in order to maintain the envisioned trajectory for the ecosystem (Table 4). This is generally a more time-consuming, but potentially valuable approach to aligning ecosystem partners in uncertain markets (c.f., Dattée, Alexy, and Autio 2018). This communication is also facilitated by utilising both global and local partners that can support and amplify the claims of orchestrators (Tatarinov, Ambos, and Tschang 2022).

The means used by orchestrators to incentivise other actors and influence their development trajectories can also be utilised to resolve various forms of bottlenecks by stimulating investment (Table 4). As explained, institutional differences can introduce new bottlenecks into the ecosystem structure that require that new activities are performed in that market, which can offer attractive value capture opportunities for local complementors (Baldwin 2018; Hannah and Eisenhardt 2018; Yildirim, Clarysse, and Wright 2022). It could also necessitate reliance on global partners that bring their capabilities and resources to the market alongside orchestrators (Tatarinov, Ambos, and Tschang 2022). Without their resolution, bottlenecks hamper, or in worst cases, prevent, ecosystem value from reaching the end user, meaning their resolution needs to be a priority for orchestrators as they enter new markets.

ARM shared substantial knowledge of their core technology and shared IP with local OEMs to ‘guarantee manufacturing feasibility’ (Rong et al., 2015, p. 301) so to incentivise commitment of the local complementors. Influencing was the mechanism as ARM began incubating potential partners by creating educational materials for both engineering students and their potential complementors (Rong et al., 2015). This was done concurrently with the indirect communication of their vision through broad channels. Thus, it used a mix of direct and indirect means of influencing the adoption of the core technology within the market (Rong et al., 2015). The mechanisms employed by ARM to incubate local complementors also served the dual purpose of resolving the technological bottleneck on the demand-side by sharing IP to establish demand for their product.

Better Place used a mix of direct and indirect communication to influence the development towards its envisioned trajectory. It worked directly with the framing of its value proposition to fit each market it entered (Noel and Sovacool 2016). At the same time, Better Place and its founder was a high-profile case and the subject of many news articles, TedTalks, and case studies (Adner 2013; Noel and Sovacool 2016). However, Better Place relied on global partners in the design of complementary assets for their core technology and faced rising strategic bottlenecks in the Israeli market as the government’s attention shifted from electrification to gas (Noel and Sovacool 2016).

In Turkey, the lack of developed educational institutions resulted in bottlenecks that proved to be valuable positions for local entrepreneurs to hold within the ecosystem (Yildirim, Clarysse, and Wright 2022). Local entrepreneurs were incentivised to enter the ecosystem through dealership rights, and provided the ecosystem with the means to bypass and fill institutional voids present on the market that prevented value creation and capture within the ecosystem (Yildirim, Clarysse, and Wright 2022).

In sum, we propose that the liabilities to which an entering ecosystem is exposed results in the emergence of bottlenecks via institutional voids, inaccessible complementary assets, and necessary infrastructure. When technological bottlenecks emerge (Table 3), the blueprinting process is engaged in the first hand because new activities are needed to account for these bottlenecks, which can also affect positions and links. This subsequently engages the aligning process to bring actors into the adjusted blueprint. Conversely, when attempting to gain access to complementary assets, which comprises a strategic bottleneck (Table 4), the aligning process is engaged in the first hand. This is because appropriate complementary assets that fit to an existing blueprint can be aligned without changing the blueprint. However, if the attempt at aligning fails, the complementary asset instead needs to be treated as a technological bottleneck, which engages the blueprinting process.

6. Discussion and conclusion

Our process theorising of ecosystem creation on foreign markets, with a theory adaptation approach integrated the structural perspective on ecosystem and relevant insights from the liabilities of foreignness. The creation of ecosystems on foreign markets necessitates the creation of new ecosystem structures that are adjusted to the value proposition so to overcome bottlenecks emerging from liabilities of foreignness through interaction with foreign market actors. The model explains how orchestrators can reconstitute the ecosystem as a stable structure, integrate it into the existing social structure of the foreign market, and create ecosystem value for local end users.

First, our model contributes to ecosystems literature by explicating the processes and mechanisms associated with ecosystem creation and how the agency of orchestrators drives these processes through goal-oriented actions and interaction with complementors. There have been criticisms levied towards the structural ecosystem perspective for being static and not accurately portraying how ecosystems are created (c.f., Autio 2021; Dattée, Alexy, and Autio 2018; Thomas and Autio 2020). By deriving process from the structural definition of ecosystems, we demonstrate that the structural perspective is inherently processual. Our model shows how the blueprinting and aligning processes result in constituting the ecosystem as a stable social entity through recursive interactions between actors both internal and external to the ecosystem and the social structures in which they operate (Giddens 1984; Langley 2009).

Second, while the ecosystem literature has recognised the importance of complementarities for value creation (Jacobides, Cennamo, and Gawer 2018; Lingens, Seeholzer, and Gassmann 2022), fewer studies have explicitly considered the broader scope of liabilities introduced by international expansion. Therefore, we contribute by drawing on the liabilities of foreignness. While it is an umbrella concept, it emphasises a wide spectrum of challenges that are relevant for ecosystem expansion (c.f., Lu, Ma, and Xie 2022). In our model, we consider institutional differences, which we operationalise as institutional voids, availability of complementary assets, and infrastructure. These enabled us to expand the considered sources of bottlenecks to incorporate specific institutional differences. Expanding on the concept of bottlenecks is our third contribution. Bottlenecks are typically considered from technological or strategic perspectives (Baldwin 2018; Hannah and Eisenhardt 2018; Kapoor 2018). However, institutions play a more or less important role in the realisation of ecosystem value, and their deficiencies can be characterised as bottlenecks from the structural perspective. This aligns with insights from by Adner (2017) and Ojala et al. (2018). By developing our model, we offer practical insights into the resolution of bottlenecks that emerge from these institutional differences.

As a first attempt to provide a processual explanation of how ecosystems can expand on foreign markets, our model has limitations that give directions for future research that can contribute to strengthening its explanatory potential. The first is the empirical substantiation of our model. Even though we have relied on published case studies to develop our arguments, no empirical studies have, to our knowledge, explicitly focused on the processes of ecosystem expansion to foreign market. Longitudinal case studies within varied institutional and industrial contexts seem appropriate to substantiate the processes our model explicates. Such studies may also reveal additional mechanisms that deserve a place within the model.

Second, we have emphasised ecosystem orchestrators as the drivers of the ecosystem creation processes in interaction with local actors. However, future research should deepen the understanding the agency of local market actors within this process. Following such instances longitudinally would also allow for the evaluation of outcomes and could potentially reveal insight into the relative impact of local vs global actors on ecosystem value creation and capture.

Third, our model focuses on the expansion of the ecosystem to a foreign market and the associated processes. However, it does not consider the effect of expansion to a foreign market on the home market. Since internationally expanding ecosystems may align a mix of both local and global complementors (Tatarinov, Ambos, and Tschang 2022), how cross-border flows impact the materialisation of ecosystem value seems a relevant direction for future research.

Fourth, while our model does not assume single orchestrators, the prevailing view is that ecosystem leadership is the purview of single firms (c.f., Foss, Schmidt, and Teece 2023). Hence, future research could offer insight into alternative leadership modes that are better suited for international expansion, which would be interesting to link to how orchestrators acquire and develop the required capabilities. This could potentially lead to integrations between ecosystem creation processes and internationalisation models, such as the Uppsala model (c.f., Johanson and Vahlne 2009), international new ventures (Oviatt and McDougall 1994), or born globals (J. Hennart 2014; Monaghan, Tippmann, and Coviello 2020). Many of the base assumptions regarding firms are consistent across these models. Integrating a deeper perspective on firm evolution into the understanding of ecosystem structures and process outcomes would offer interesting insights that could create further bridges between the ecosystem and internationalisation literatures.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

¹ Our conception of ecosystems follows the structural perspective, which does not distinguish between ecosystem ‘types’ based on labels, but rather based on the nature of the value propositions and related complementarities (Adner 2017). This most closely relates to what Thomas and Autio (2020) label modular ecosystems and Adner and Kapoor (2010), Jacobides et al. (2018), and Shipilov and Gawer (2020) label innovation ecosystems. These ecosystems are characterised by non-hierarchical orchestration and primarily unique complementarity (Jacobides, Cennamo, and Gawer 2018; Teece 2018), rather than supermodular complementarity (e.g. network effects; Shipilov and Gawer 2020; Teece 2018). Typically, these ecosystems have less implicit hierarchy than, for example, platform ecosystems due to the lack of a strong central actor that controls the platform. Instead, these ecosystems are often characterised by contestable and/or shared leadership where platforms and digitalisation are tools for materialising value rather than the core value proposition (Adner 2017; Foss, Schmidt, and Teece 2023; Lingens, Miehé, and Gassmann 2021).

² We establish this basic assumption to establish a taking off point for our theorising. We acknowledge that the value proposition could change during internationalisation but argue that this is not a necessary condition for our process model. Rather, we claim that the ecosystem blueprint and alignment will change regardless of whether the value proposition changes. Therefore, we take the value proposition as given ex ante to facilitate a parsimonious model. See Dattée et al. (2018), Snihur et al. (2018), and Thomas and Ritala (2021) for more details regarding the development of ecosystem value propositions.