Kati Järvi
ECOSYSTEM ARCHITECTURE DESIGN:
ENDOGENOUS AND EXOGENOUS STRUCTURAL PROPERTIES
Acta Universitatis Lappeenrantaensis 511
Thesis for the degree of Doctor of Science (Technology) to be presented with due permission for public examination and criticism in the Auditorium of the Student Union House at Lappeenranta University of Technology, Lappeenranta, Finland on the 23rd of February, 2013, at noon.
Supervisor Professor Marko Torkkeli
Faculty of Technology Management Department of Industrial Management Lappeenranta University of Technology Reviewers Dr. Anne‐Laure Mention
Innovation Economics & Service Valuation Unit Centre de Recherche Public Henri Tudor Luxembourg
Associate Professor Marko Seppänen
Center for Innovation and Technology Research Department of Industrial Management
Tampere University of Technology Finland
Opponent Professor Helena Forsman School of Management University of Tampere Finland
ISBN 978‐952‐265‐378‐9 ISSN‐L 1456‐4491
ISSN 1456‐4491
Lappeenranta University of Technology Yliopistopaino 2013
ABSTRACT Kati Järvi
Ecosystem architecture design: endogenous and exogenous structural properties Lappeenranta 2013
67 p.
Acta Universitatis Lappeenrantaensis 511 Diss. Lappeenranta University of Technology ISBN 978‐952‐256‐378‐9
ISSN‐L 1456‐4491 ISSN 1456‐4491
The aim of this dissertation is to bridge and synthesize the different streams of literature addressing ecosystem architecture through a multiple‐lens perspective. In addition, the structural properties of and processes to design and manage the architecture will be examined. With this approach, the oft‐neglected actor‐structure duality is addressed and both the position and structure, and action and process are under scrutiny. Further, the developed framework and empirical evidence offer valuable insights on how firms collectively create value and individually appropriate value.
The dissertation is divided into two parts. The first part comprises a literature review, as well as the conclusions of the whole study, and the second part includes six research publications. The dissertation is based on three different reasoning logics: abduction, induction and deduction;
related qualitative and quantitative methodologies are utilized in the empirical examination of the phenomenon in the information and communication technology industry. The results suggest firstly that there are endogenous and exogenous structural properties of the ecosystem architecture. Out of these, the former ones can be more easily influenced by a particular actor whereas the latter ones are taken more or less for granted. Secondly, the exogenous ecosystem design properties influence the value creation potential of the ecosystem whereas the endogenous ecosystem design properties influence the value appropriation potential of a particular actor in the ecosystem. Thirdly, the study suggests that there is a relationship between endogenous and exogenous structural properties in that the endogenous properties can be leveraged to create and reconfigure the exogenous properties whereas the exogenous properties prose opportunities and restrictions on the use of endogenous properties. In addition, the study suggests that there are different emergent and engineered processes to design and manage ecosystem architecture and to influence both the endogenous and exogenous structural properties of ecosystem architecture.
This study makes three main contributions. First, on the conceptual level, it brings coherence and direction to the fast growing body of literature on novel inter‐organizational arrangements, such as ecosystems. It does this by bridging and synthetizing three different streams of literature, namely the boundary, design and orchestration conception. Secondly, it sets out a framework that enhances our understanding of the structural properties of ecosystem architecture; of the processes to design and manage ecosystem architecture; and of their influence on the value creation potential of the ecosystem and the value capture potential of a particular firm. Thirdly, it offers empirical evidence of the structural properties and processes.
Keywords: architecture; ecosystem; design; inter‐organizational relationships UDC: 65.012.6:65.011.8
ACKNOWLEDGEMENTS
Thank you, Helena Forsman, for being my opponent.
Thank you, Marko Torkkeli, for being my supervisor.
Thank you,
Anne‐Laure Mention and Marko Seppänen,
for your extremely valuable comments that helped me finalize the dissertation.
Thank you, my co‐authors,
for your contributions and collaboration on the publications.
Thank you, Paula Haapanen,
for making the dissertation understandable.
Thank you,
the Research and Training Foundation of Telia‐Sonera Finland Oyj,
the Finnish Doctoral Program in Industrial Engineering and Management, and Tekniikan Edistämissäätiö,
for providing financial support.
Helsinki, February 2013 Kati Järvi
TABLE OF CONTENTS
LIST OF PUBLICATIONS OF PART II AND AUTHOR’S CONTRIBUTION PART I: OVERVIEW OF THE DISSERTATION
1 INTRODUCTION ... 13
1.1 Research background and motivation ... 13
1.2 Research gaps ... 14
1.3 Research objectives and research questions ... 16
1.4 Key concepts and boundary conditions ... 16
1.5 Outline ... 20
2 RESEARCH DESIGN ... 21
2.1 Ontology, epistemology and methodology ... 21
2.2 Multiple‐lens perspective ... 22
2.3 Reasoning and methods ... 25
2.3.1 Abduction ... 28
2.3.2 Induction ... 29
2.3.3 Deduction ... 31
3 ECOCYSTEM ARCHITECTURE: A MULTIPLE‐LENS PERSPECTIVE ... 32
3.1 Boundary conception ... 33
3.2 Design conception ... 36
3.3 Orchestration conception ... 37
3.4 A summary of the conceptions ... 38
4 OVERVIEW OF THE RESULTS OF THE PUBLICATIONS ... 42
4.1 Publication I ... 42
4.2 Publication II ... 43
4.3 Publication III ... 44
4.4 Publication IV ... 44
4.5 Publication V ... 45
4.6 Publication VI ... 46
5 CONCLUSIONS ... 49
5.1 Review of the results of the study and answering the research questions ... 49
5.2 Theoretical implications ... 52
5.3 Managerial implications ... 53
5.4 Limitations ... 53
5.5 Suggestions for further research ... 55
REFERENCES ... 57
PART II: PUBLICATIONS ... 67
LIST OF PUBLICATIONS OF PART II AND AUTHOR’S CONTRIBUTION
The dissertation comprises an introductory part and six individual publications. The individual publications are listed below, together with a description of their role in the dissertation. Further, as the publications are co‐authored, the author’s contribution to each one is described.
PUBLICATION I: Järvi, K., & Kortelainen, S. 2011. The dynamism of the two‐sided application store market. International Journal of Technology Marketing, 6: 57‐71.
Publication I locates the research problem, creates a pre‐understanding of the dimensions of ecosystem architecture design, and sets out relevant directions for future studies and theory development. The paper was accepted to the International Journal of Technology Marketing following a double blind review process. The author was responsible for planning, designing, and coordinating the research. She also coordinated the writing process. She conducted the literature review and was responsible for the conceptual development. Analysis with system dynamics was done in collaboration with the co‐author. Discussion and conclusions were written with the help of the co‐author.
PUBLICATION II: Järvi, K., & Pellinen, A. 2010. Redefinition of the business model concept with the value co‐creation aspect. Proceedings of the 10th Annual EBRF Conference, September 15‐17, 2010, Nokia, Finland.
Publication II is a conceptual paper that proposes a taxonomy of variation in ecosystem architecture based on two structural properties. The paper was accepted to the 10th Annual EBRF Conference after the double blind review of an extended abstract and was presented at the conference session. The author was responsible for planning, designing, and coordinating the research. She also coordinated the writing process. She conducted the literature review. Conceptual development was done in collaboration with the co‐author. Discussion and conclusions were written with the help of the co‐author.
PUBLICATION III: Pellinen, A., Ritala, P., Järvi, K., & Sainio, L‐M. 2012. Taking initiative in market creation – a business ecosystem actor perspective.
International Journal of Business Environment, 5: 140‐158.
Publication III expands the literature on management and governance of innovation networks to the ecosystem level and examines, with qualitative, exploratory case study, multi‐actor collaboration and the role of initiative taking from different actors in a business ecosystem. The paper was accepted to the International Journal of Business Environment following a double blind review process. The author planned and designed the research with the co‐authors. She conducted the literature review in collaboration with the second and fourth co‐author. She conducted the data collection (semi‐structured interviews) and analysis. She wrote the case description.
She wrote the results together with the first co‐author (after the analysis the results and interpretations were discussed with co‐authors and firm
representatives). She wrote the discussion and conclusions together with the second co‐author.
PUBLICATION IV: Järvi, K., Sainio, L‐M., Ritala, P. & Pellinen, A. 2010. Building a framework for partnership business model. International Journal of Management Concepts and Philosophy, 4: 100‐117.
Publication IV develops a construct of a partnership business model and with a qualitative, exploratory case study it examines the usability of such a model in examining complementarity as a property of the structural template defining multi‐actor collaboration. The paper was accepted to the International Journal of Management Concepts and Philosophy following a double blind review process. The author was responsible for planning, designing, and coordinating the research. She also coordinated the writing process. She conducted the literature review, formulated the theoretical framework, and conducted the data collection (semi‐structured interviews) and analysis. The results and interpretations were discussed with co‐
authors and firm representatives after the analysis. Discussion and conclusions were written together with the help of the co‐authors.
PUBLICATION V: Järvi, K., Kortelainen, S., Puumalainen, K., & Ritala, P. 2012. Entry timing strategies and competition in platform‐based markets.
Proceedings of the 23rd ISPIM Conference, June 17‐20, Barcelona, Spain.
Publication V examines different market entry timings and a range of strategic decisions in competition between ecosystems. The paper was accepted to the 23rd ISPIM Conference after the double blind review of an extended abstract and was presented at the conference session. The author was responsible for planning and coordinating the research. She also coordinated the writing process. She conducted the literature review. She formulated the hypotheses with the help of the fourth co‐author. She designed the simulation model in collaboration with the second co‐author.
She wrote the results together with the third co‐author. She was responsible for writing the discussion and conclusions.
PUBLICATION VI: Uzunca, B., Järvi, K. & Tscherning, H. 2012. Sources of architectural advantage: platform competition in multi‐sided markets. Proceedings of the 32nd Annual International SMS Conference, October 7‐9, 2012, Prague, Czech Republic.
Publication VI proposes a conceptual model of the structural properties which can be used and leveraged to change the focal firm’s architecture to appropriate more value. The paper was accepted to the 32nd Annual International SMS Conference after the double blind review of an extended abstract and was presented at the conference session. The author planned and designed the research with the co‐authors. She coordinated the research process and writing of the manuscript. She conducted the literature review, developed the conceptual model and formulated the hypotheses in collaboration with the co‐authors. Discussion and conclusions were written together with the first co‐author.
PART I: OVERVIEW OF THE DISSERTATION
13 1 INTRODUCTION
1.1 Research background and motivation
It is quite common, especially in the field of information and communications technology (ICT) for organizations to divide labour (and surplus) in novel inter‐organizational arrangements, such as ecosystems. Furthermore, these ecosystems compete against each other. At the heart of these ecosystems are hub firms, who design and manage both the architecture of the technology and the ecosystem. Through operating a technological platform, hub firms enable the creation of complementary and interoperable products. In addition, the hub firms orchestrate a wide array of activities aimed at collective value creation between the hub firm and complementors. Examples of these types of arrangements include the video game ecosystem, led by game console manufacturers, such as Microsoft and Sony, and complemented by game developers; and application store ecosystems, led by mobile phone manufacturers, such as Apple and Nokia, and complemented by third party application developers.
Multiple streams of research have noted the rise of novel inter‐organizational arrangements, such as networks and ecosystems, where organizations focus on their own strengths while leveraging those of their partners (i.e., resources and market positions). Discussion on where a firm draws its boundaries in general (Williamson, 1991; Parmigiani & Mitchell, 2009) and in the context of networks and ecosystems (Jacobides & Hitt, 2005) has been extensive. Further, research on inter‐
organizational value creation and appropriation suggests that the focal firm must rely on its ties to partners in order to enhance its performance and to profit from innovation (Ozcan & Eisenhardt, 2009), and research on business models extends the locus of value creation outside of firm boundaries (Zott & Amit, 2008).
When Nokia invented the first 3G supported mobile device (the 6650) in 2002, it was doing everything right in terms of innovation. However, what Nokia did not understand was that the success of the new product would be dependent on many complementary products and services, such as infrastructure or the content providers. Other parts of the ecosystem that would have supported Nokia in its efforts were not ready. It was Apple who succeeded with a similar effort in 2007 by launching iPhone when, this time, the mobile digital services partners in the ecosystem were ready to support the fledgling 3G technology (Adner, 2012).
This example illustrates the phenomenon of shrinking the core while expanding the periphery (Gulati & Kletter, 2005; Gulati et al., 2012), which the aforementioned streams of literature, the emerging literature on inter‐organization design and also this dissertation seek to understand. The contemporary business world is full of examples like this (AMR, P&G, Toyota, Apple, and Wikipedia (Gulati et al. (2012)), especially in the information and communication technology industry due to the systemic nature of the industry (i.e., Ethiraj & Puranam, 2004) and its technologies (i.e., Bartnett, 1990). Firms are increasingly pushed towards adapting to competitive pressures that originate from interconnected environments. Multiple parties contributing to new product designs or co‐creating with partners while leveraging knowledge from multiple industries are some of these pressures that companies have to face. This and shrinking the core is a natural result of greater specialization (Gulati et al., 2012), and firms are more aware than ever that individual efforts are not enough for success. In this sense, collaboration between firms and understanding the outcomes of asymmetries among complementary players’ interdependences becomes of crucial importance (Gulati & Singh, 1998; Gulati & Sytch, 2007).
14
The complementary nature of resources, which are an important rationale for partnership formation and criteria in partner selection (i.e., Das & Teng, 2000), can be illustrated with the following example: “an IT firm launching a new software platform may seek to include established OEMs in its ecosystem of partners because they can lend legitimacy to the new software and provide access to a large potential customer base” (Gulati et al., 2012, p. 576). Complementary resources are the distinctive resources of partners which collectively generate greater rents than the sum of those obtained from the individual endowments (Dyer & Singh, 1998; Hitt et al., 2000).
Previous research, applying for example the resource‐based view of the firm to strategic alliances, has discussed resource and asset complementarities as determinants of the structure of the inter‐
organizational collaboration in dyadic (Das & Teng, 2000) and industry‐wide settings (Jacobides et al., 2006). More extensively, the structure of collaboration has been examined in social network theory, and on its embeddedness, and positional (Granovetter, 1985; Uzzi, 1997) and structural patterning (Burt, 1992; Kilduff & Tsai, 2003) arguments.
Most recently, the emerging literature on the design and dimensions of inter‐organizational structures (Fjeldstad et al., 2012; Gulati et al., 2012; Jacobides et al., 2006) has addressed the several unique theoretical puzzles which the shrinking the core and expanding the periphery phenomenon poses, such as how organizations coordinate and distribute rents among themselves, what dimensions guide the design choices of the structures, how the dimensions or structural properties are interrelated, and how they affect the desired outcomes (i.e., system level goals or value creation) of the ecosystems.
The dissertation is motivated by the shrinking the core and expanding the periphery phenomenon taking place particularly in the context of the information and communication technology industry and the emergence of the application store market. In this context, the mobile operators who have traditionally controlled the customer interface and produced or at least delivered the content are now playing the second fiddle to new market entrants such as Google, Microsoft, and Apple, creating more open ecosystems. The heavily vertically‐integrated, operator‐centric walled garden model has gradually been replaced with the multi‐actor collaboration structure responsible for value creation. This structure has emerged as the template to divide labour and surplus between the multiple actors and referred to as ecosystem architecture. The dissertation focuses on producing a better understanding of the phenomenon in question and of the aforementioned theoretical puzzles. In addition to the practical importance and relevance of the topic, three theoretical gaps have been identified. These gaps, stressing the theoretical importance and relevance, are elaborated next.
1.2 Research gaps
Inter‐organizational collaboration is evolving from control‐based, hierarchical and contractual linkages to more novel arrangements, such as ecosystems. To that end, different streams of literature have risen to address the design of collaboration between organizations. For example, the recently introduced meta‐organization design stream states that coalitions of formally independent firms now collaborate multilaterally as if one meta‐organization (Fjeldstad et al., 2012; Gulati et al., 2012), inducing cooperation through mutual dependence, sometimes even in the absence of binding contracts. Building on organization design literature, the extension in the unit of analysis from intra‐firm design to inter‐firm design relies of the approach of vertical, cross‐level theory borrowing, a tradition widely utilized in the studies on inter‐organizational relationships. In general, the theoretical and empirical examination of managing inter‐organizational arrangements has comprised the dyadic level or the so‐called relational view (Dyer & Singh, 1998; Teece, 1986), network or portfolio level (Lavie, 2007; Ozcan & Eisenhardt, 2009) and most recently the
15
ecosystem level (Adner & Kapoor, 2010; Iansiti & Levien, 2004) as the unit of analysis. The same shift in the unit of analysis is taking place in the examination of the design of inter‐organizational structures. Whereas the organization design literature at the firm level is quite explicit in explaining the properties and design principles of organizational structure and architecture (Galbraith, 1973), the internal fit between strategy and structure, and the external fit between structure and environment (Donaldson, 2001), the research on the new level of analysis is still in its infancy. Past research has indicated some properties of inter‐organizational design (Fjeldstad et al., 2012; Gulati et al., 2012; Jacobides et al., 2006), but no coherent framework of such design properties exists.
Further, it is unclear how such properties are interrelated and how they affect the desired outcomes (e.g., system‐level goals or value creation) of the ecosystems. This represents the first identified gap.
Previous research on the network or ecosystem level suggests that leading hub firms are often needed in order to orchestrate broad networks of complementary actors (e.g., Belussi & Arcangeli, 1998; Dhanaraj & Parkhe, 2006; Ritala et al., 2009). However, in sociologically rooted inter‐
organizational research, the larger socioeconomic context and the structure of the network have been emphasized over the actions (such as orchestration activities) of a particular actor. Similarly, in the industrial economic literature, the interactions, transactions or building blocks are emphasized. The deficit of such a focus is the neglect of firm‐level strategies, processes, and behaviours of individual actors. Not to lose sight of the forest for the trees (Jacobides & Hitt, 2005) or vice versa (Salancik, 1995), both the trees (the actors) and the forest (the structure of the network) should be paid equal attention. The previously unexplored actor‐structure duality (Dhanaraj & Parkhe, 2006) represents the second identified gap.
The third identified gap arises from the unequal theoretical and empirical examinations of the value creation and appropriation phenomena. For example, the boundary conception related to ecosystem level value creation and appropriation emphasizes value appropriation more due to the path dependency of this particular theorizing; the boundary conception builds upon transaction cost economics and evolutionary economics, as firm level theories traditionally have the tendency to emphasize profiting from innovation more (i.e., Teece, 1986). Another point of inequality rises from the strategic actions or moves which firms can employ to increase their individual share of the collectively created value. Again, burdened by history, the boundary conception states that in order to increase value appropriation and architectural advantage (proportion of the value created), the strategy and strategic action to emphasize is the modification of the boundaries of the firm (vertical integration or disintegration). Orchestration conception goes further in describing different processes through which the hub firm can influence the ecosystem’s innovation output (Dhanaraj &
Parkhe, 2006).
In line with the formulated research question, the purpose of this study from the theory building point of view is to produce a clear base on which scholarly work on ecosystem architecture design can ascend coherently. As a reflection on scholarly specialization, there is a tendency, similar to many other management fields, to create isolated silos of knowledge (Pfeffer, 1993) in ecosystem research as well. By utilizing a multiple‐lens perspective to organize quite discrepant references and to create intersecting areas between three acknowledged and developed research programs, this dissertation combines various theoretical conceptions, such as the boundary, design and orchestration conception, to develop a more comprehensive ecosystem architecture perspective.
16 1.3 Research objectives and research questions
The main objective of the dissertation is to increase the understanding of ecosystem architecture design and to examine it both from exogenous and endogenous perspectives. The dissertation firstly looks at the architecture of ecosystems, and the structural properties of the architecture.
Secondly, it discusses the implications of such structural properties on value creation and appropriation. Thirdly, the scholarly inquiry addresses the processes for designing and managing the architecture. Given the objective of the study, the main research question is formulated as follows:
How does the design of the ecosystem architecture affect the value creation and appropriation within a particular ecosystem?
To structure the research effort and to facilitate the analysis of the different aspects that are relevant to the main research, two separate sub‐questions are formulated. The first one concerns the examination of variation in different ecosystem architectures and the structural properties as the dimensions according to which variation takes place. Hence the first sub‐question is:
Sub‐question 1: What are the structural properties of ecosystem architecture?
The second sub‐question addresses the often neglected actor‐structure duality and is formulated as follows:
Sub‐question 2: How are ecosystem architectures designed and managed?
1.4 Key concepts and boundary conditions
This chapter clarifies and defines the central constructs of the dissertation, and elaborates their scope conditions and interrelations. The central constructs of the dissertation are ecosystem, ecosystem architecture, value creation, and value appropriation. Borrowed from biology, the term ecosystem appears frequently both in the jargon used by business practitioners and academics.
Morton (2012) explains that in 1935, opposing the dominant view of holism, British ecologist Arthur Tansley coined the term ecosystem. Tansley defined that ecosystems are like the human mind; dynamic and shaped by circumstances. In biology, an ecosystem is not composed only of plants and animals, but also of their mineral substrates and the energy they use. Unlike the communities of holism, which have a preordained endpoint, ecosystems are the product of the forces and flows that made them. While there are typical ecosystems, there can also be novel ones;
wherever living things come together there has to be an ecosystem of some sort, whether nature had envisaged it or not.
The ecosystem construct was introduced to organization and management studies by Moore (1993). With the term, he wanted to draw attention to the following issues. Firstly, he suggests that a firm is not only a member of a single industry but a part of a business ecosystem which might cross a variety of industries. Secondly, firms do not exist in a vacuum; they attract resources from various sources and they are subjected to changes in the external environment. Lastly, they co‐
evolve together with their partners and confront environmental changes together.
Building upon Moore (1993), industrial economics, organization design and sociologically rooted inter‐organizational research have tried to examine and understand systems where organizations come together. In terms of theoretical proliferation, it is not merely raining, it is starting to pour.
17
Different streams of literature emphasize different areas and so far, no stream has been able to offer a holistic overview of the architecture of an ecosystem and how to design and manage such architecture. As self‐contained theoretical islands seldom lead to integrated streams of empirical research that provide a source of continuity in the discipline, this dissertation offers an effort to produce an integrative examination of ecosystem architecture.
An ecosystem comprises interdependent, interconnected and interacting coevolving actors;
customers, agents and channels, sellers of complementary products and services, suppliers, and the firm itself (Moore, 1996; Gossain & Kandiah, 1998). The construct is based on the ecological metaphor that firms participate in a larger economic community, an ecosystem, where each actor occupies a contributing role and forms symbiotic relationships with other actors (Moore, 1993;
Moore, 1996; Iansiti & Levien, 2004; Basole & Karla, 2011). Definitions of ecosystems, such as the one of an innovation ecosystem tend to emphasize the interdependence: “a given innovation often does not stand alone; rather it depends on accompanying changes in the firm’s environment for its own success. These external changes which require innovation on the part of other actors embed the focal firm within an ecosystem of interdependent innovations.” (Adner & Kapoor, 2010, p. 306) Besides being called as innovation ecosystems, these communities are also more generically labelled business ecosystems.
In a related research stream, Gulati et al. (2012, p. 573) offer a strikingly similar definition of their
“competing” construct, meta‐organization, by stating that “meta‐organizations resemble biological super‐organisms, a multitude of individual organism that coexists, collaborate, and coevolve via a complex set of symbiotic and reciprocal relationships”. They further specify that meta‐
organizations comprise networks of firms or individuals not bound by authority based on employment relationships, but characterized by a system‐level goal (Gulati et al., 2012). A similar approach is taken by Fljedstad et al. (2012, p. 737) in their definition of the architecture of inter‐
organizational collaboration as the “use of flexible assembly of firms with specialized capabilities to achieve economies of scale and experience”.
The industry sector construct offers an alternative viewpoint in order to define multi‐actor collaboration. Malerba (2002) defines an industry sector as a network of firms in different sub‐
industries that supply complementary goods for making a set of system products. The systemic nature of a product refers to for example components of a technological system valued as complements by the customer (Ethiraj & Puranam, 2004). In a sector, various types of relationships, such as alliances, joint ventures, and competition may connect the firms. An industry sector is a system that emerges from the participation and interactions of self‐organized firms, coordinated by technical and economic forces. Thus it is not engineered by anyone (i.e., a hub).
Further, literature on two‐sided and multi‐sided markets (Evans & Schmalensee, 2007) identifies the structure of collaboration in terms of participants (intermediary or platform and two or more sides of the market). This stream of literature has however been focusing mostly on the nature and characteristics of such structure. Issues such as what are the functions performed by the intermediary or platform are left aside and the literature on two‐sided and multi‐sided markets is not full‐fledged enough in terms of defining who does what and who gets what in terms of value creation and appropriation.
Given these definitions, ecosystems can be differentiated from alliances, alliance portfolios, networks, clusters and other aggregates of co‐operating firms in several dimensions. Firstly, as traditional inter‐organizational forms, such as strategic alliances, are usually based on closed membership, ecosystems can include open memberships. Secondly, although interdependence and
18
reciprocity is a shared characteristic between an ecosystem and other aggregates of co‐operating firms, the notion of coevolution is unique to the ecosystem. Thirdly, the dissertation subscribes to the view that ecosystem architecture can both emergent and engineered. Whereas the definition of the industry sector (Malerba, 2002) states that no actor engineers the architecture, that it emerges based on multi‐actor interaction, the dissertation discards this view by turning to the emerging literature on agency and strategic action stating that inter‐organization collaboration can be given birth to (i.e., Ozcan & Eisenhardt, 2009). Further, this offers a point of departure from the deterministic view of inter‐organizational collaboration.
Architecture is defined as “the synthesis of form in response to function” (Alexander, 1964;
Fjeldstad et al., 2012); it is the configuration of a system taking into account how its components relate, work, and integrate to and with each other and the environment. Thus, structure should be consistent with purpose, meaning that form must follow function1 or vice versa (Fjeldstad et al., 2012). The concept of architecture has been used in many levels including products (Sanchez &
Mahoney, 2003), organizations (Ethiraj & Levinthal, 2004), and industries (Jacobides, 2005;
Jacobides et al., 2006). In addition, previous research includes multi‐level examination of architecture and exploration of how the architecture of one system may affect the architecture of other systems, namely how product architecture may influence organization architecture or vice versa. This was labelled the mirroring hypothesis (Langlois & Robertson, 1992; Sanchez &
Mahoney, 1996; Baldwin & Clark, 1997, 2000; Schilling, 2000; Hoetker 2006; Fixson & Park, 2008;
Tiwana, 2008a, 2008b; Colfer & Baldwin, 2010).
Regarding the definition of ecosystem architecture, this dissertation builds upon the literature on industry architecture (Miller, 2001; Jacobides et al., 2006). Industry architecture represents “a structure of co‐specialized2 agents and assets” (Jacobides et al., 2006, p. 1202) and as a template, it describes the distribution of labour and assets among a set of co‐specialized firms across a set of industries (Jacobides et al., 2006). Pisano and Teece (2007) further define industry architecture as the characterization of the nature and degree of specialization of industry players (or organizational boundaries and the structure of relationships between those players).
Synonymously to industry architecture, sector structure (Luo, 2010) and value creation architecture (Dietl et al., 2009) have been used to describe the division of labour between different actors. Luo (2010), adopting Malerba’s (2002) definition of the industry sector, defines sector structure as the transactional network of firms among the complementary industries in a sector and as the patterns of transactional relationships between all the firms in a sector comprising several levels of industries. By introducing value creation architecture, Dietl et al. (2009, p. 26) define it as the “structure and relationships of all the value‐adding activities that are carried out by various actors and companies to bring a particular product or service to market”.
Previous research has noted that ecosystem architectures are not static. Ecosystem architectures are shaped by legal and regulatory authority and also by industry participants who stand to benefit from a particular architecture. These stakeholders usually fight the introduction of new alternatives through legislative or regulatory means (Jacobides et al., 2006; Ferraro & Gurses, 2009). However, this study does not focus on these types of macro‐level exogenous factors affecting
1 Originally unidirectional causality was announced, for example in structural contingency theory, in terms of form follows function or structure follows strategy. This view encountered considerable criticism (Drazin & Van de Ven, 1985;
Peteraf & Reed, 2007).
2 Jacobides et al. (2006) use the term co‐specialization as their framework builds upon Teece’s (1986) framework for co‐
specialization and profiting from innovation. For a more detailed definition of industry architecture, see Jacobides et al.
(2006), p. 1203.
19
ecosystem architecture. Thus, industry evolution or the industry life cycle model is out of the scope of the theoretical examination of the dissertation and the definition of the ecosystem architecture construct. In contrast, the study does contribute to those literatures as depicted in the theoretical implications section.
As an ecosystem is defined to comprise several different types of interacting organizations, previous research has come up with a multitude of terms to address the positions which organizations can hold in an ecosystem. In a loosely‐coupled inter‐organizational system characterized by high‐centrality and low‐density (Dhanaraj & Parkhe, 2006) such as in an ecosystem one can usually distinguish between at least entities occupying more central positions and those occupying peripheral positions. The vocabulary referring to the entity occupying a more central position includes hub firm (Jarillo, 1988), key actor (Knoke, 1994), triggering entities (Doz et al., 2000), strategic centres (Lorenzoni & Baden‐Fuller, 1995), flagship firms (Rugman & D’Cruz, 2000), network orchestrator (Hacki & Lighton, 2001), keystone firm (Iansiti & Levien, 2004), platform leader (Gawer & Cusumano, 2002; Gawer & Cusumano, 2008), relationship‐centred organization (Gulati & Kletter, 1995), and ecosystem architect (Gulati et al., 2012). Given the scope of this study and taking rhetorical consistency into account, it would be anticipated from the dissertation to adopt the term ecosystem architect to refer to the actor in the central position in a given architecture. However, as architect has a strong connotation only to structure and design, the dissertation adopts the more connotation‐free, neutral and frequently used term of ecosystem hub3.
The notion of architectural advantage, a favourable position in certain ecosystem architecture, builds upon Teece’s (1986) co‐specialization framework and incorporates cooperative game theory (Brandenburger & Stuart, 1996; Brandenburger & Stuart, 2004), resource‐based analysis (Lippman and Rumelt, 2003; Winter, 1995), industry evolution (Langlois and Robertson, 1995; Jacobides and Winter, 2005), and theoretical economics (Deardorff and Stern, 1994). Jacobides et al. (2006, p.
1200) define that such “architectural advantage comes about when firms can enhance both complementarity and mobility in parts of the value chain where they are not active”. Thus architectural advantage can lead to high levels of value appropriation without the need to engage in vertical integration (Pisano and Teece, 2007). This study builds upon this notion, but instead of discussing the implications of the ecosystem architecture to architectural advantage, this study discusses the implications of ecosystem architecture for value creation and capture in more general.
Value creation and appropriation are not focal concepts only for innovation networks, but also for the whole field of strategy and management (see e.g., Pitelis, 2009), and therefore multiple definitions of the concepts exist depending on the chosen perspective. In general, value creation refers to mechanisms related to innovating, producing, and delivering products or services to markets, and value appropriation refers to mechanisms focused on extracting profits from the marketplace (Lepak et al., 2007). More specifically in the context of innovation networks or ecosystems (Adner & Kapoor, 2010), value creation has been portrayed as a collective activity among the collaborating actors, aiming to create value for the customer with new or improved offerings (Ritala & Hurmelinna‐Laukkanen, 2009). From the economics perspective, value is
3 Nambisan and Sawhney (2011) make a distinction between a hub firm as an innovation integrator and as a platform leader. According to their definition, a hub firm as an innovation integrator defines the basic architecture of the core innovation, invites network members to provide components to this core innovation, integrates the different components and then markets it. A hub firm as a platform leader also defines basic innovation architecture, and invites network members to build their own complementary innovation. With complementary innovations, the reach and range of basic innovation architecture can be extended and/or enhanced.
20
defined as the eventual willingness‐to‐pay for a certain product from the viewpoint of the end customer (Brandenburger & Stuart, 1996; see also Bowman & Ambrosini, 2000). Thus, in the end customers define what is economically valuable or not, and the value creation efforts – in both the organization and the ecosystem level – are eventually directed towards this goal. Value appropriation, on the other hand, has been defined as the individual share that an individual actor is able to capture from the value created in innovation activities (Ritala & Hurmelinna‐Laukkanen, 2009). Both of these processes are fundamental parts of the logic of innovation networks and business ecosystems – without value creation, there is nothing to appropriate and without appropriable value in sight, there are no incentives to create the value in the first place. Aligned with this thinking, it has been suggested that in the inter‐organizational level, value creation processes are often collective by nature, while value appropriation processes are fundamentally organizational‐level processes (Adegbesan & Higgins, 2010; Ritala & Hurmelinna‐Laukkanen, 2009).
Furthermore, the outcomes of these processes are defined through the lenses of value creation potential and value appropriation potential. The potential is emphasized, rather than the realized value created or the realized value appropriated, since the identified structural properties cannot fully explain whether something is eventually realized or not. However, they provide a framework for the potential to either create value or to appropriate value. Realized value creation or appropriation means that the resources used in these processes have been used and have been performing (e.g., Das & Teng, 2000). Therefore, value creation/appropriation potential suggests boundary conditions for such phenomena, without suggesting the realization of activities.
Some researchers see value creation as a process that precedes value appropriation (e.g., Bowman
& Ambrosini, 2000), while some suggest that these are more or less parallel processes (Ritala &
Hurmelinna‐Laukkanen, 2009). The study focuses on the latter category and subscribes to the view that in innovation networks, value is created and appropriated simultaneously and continuously (while it is certainly true that value appropriation cannot take place if it is not created first). In a very specific innovation project, the value creation phase could be separated from value appropriation – which is commonly shown for instance in stage‐gate models (Cooper, Edgett &
Kleinschmidt, 2002) – but in a multi‐actor context with a long time span and various motivations, value creation and appropriation do take place in parallel. Therefore, the aim is not to build a linear process model of value creation and appropriation, but rather an actor‐structure model of how these distinct processes work in ecosystems.
1.5 Outline
The thesis comprises an introduction and six individual, complementary publications. The introduction is organized as follows. Following the discussion on the purpose of the study, and the identification of research gaps and research questions in the first chapter, the second chapter describes and justifies the ontological, epistemological, and methodological choices, and adopted reasoning logics of the dissertation. In addition, the second chapter includes an illustration of the utilized vehicle to build theory, the multiple‐lens perspective. The third chapter introduces and reviews three different conceptions of ecosystem architecture and synthesizes these to develop a framework of the structural properties and design dimensions of ecosystem architecture. The fourth chapter illustrates a summary of the publications and a review of the results. Finally, the fifth chapter closes by answering the research questions, by providing suggestions regarding the collective value creation potential of the ecosystem and individual value capture potential of a particular actor firm and by commenting upon the implications, limitations, and suggestions for further research.
21 2 RESEARCH DESIGN
The three cornerstones of research are theory, method and the empirical phenomena. The purpose of this chapter is to firstly, describe, discuss, and justify the choice of the philosophy underlying the scientific practice of this dissertation. The ontological assumptions and epistemological stance of the study are introduced which have guided the theoretical and empirical work. Secondly, this chapter introduces the multiple‐lens perspective as a vehicle to build the theoretical framework.
Thirdly, the methods of inquiry are introduced with a description of the data collection and analysis methods applied in each publication.
2.1 Ontology, epistemology and methodology
The philosophy of science, which informs us about the nature of the phenomenon examined and the methods for understanding it, underlies all forms of research. Understanding the implication of this choice is integral for any reflective and responsible scientific inquiry (Bechara & Van de Ven, 2007).
Too often, this choice is described only in terms of methods, and not in terms of ontology, epistemology, and methodology. The inquirer and the ontological and epistemological foundations of research are guided by a paradigm, which refers to the basic belief system of worldview (Greene, 2007). The inquiry paradigm defines the ontological questions (what is the form and nature of reality); the epistemological questions (what is the nature of the relationship between the knower and would‐be knower and what can be known); and the methodological questions (how can the inquirer go about finding out whatever they believe can be known), thus leaving the questions of methods secondary to questions of paradigms (Guba & Lincoln, 1994).
The historical and epistemological foundations of organization theory are grounded in integrating sociological theory with engineering discourse (Shenhav, 2003). This statement already implies that researchers in the field of organization and management studies subscribe to different inquiry paradigms. This was especially highlighted in the beginning of the 1990s and during the raging paradigm war between organization positivists or realists (Pfeffer 1982; Pfeffer 1993; Pfeffer 1995;
Donaldson 1985, 1996; Bacharach 1989), and constructivists (Lincoln 1985; Lincoln & Guba 1985;
Reed & Hughes 1992; Perrow, 1994; Van Maanen 1995; Alvesson & Deetz, 1996; Burrell 1996; Chia 1996).
Most typically, scholars in the field of organization and management science are categorized as positivists or constructivists. The former is said to have a “hypothetico‐deductive, particularistic, objective, outcome‐oriented, and natural science world view” and the latter is said to “subscribe to phenomenological, inductive, holistic, subjective, process‐oriented, and social anthropological world view” (Reichardt & Cook, 1979, p. 9‐10). Moving beyond the dichotomy, scholars such as Burrel and Morgan (1979) and Deetz (1996) have offered more detailed categorizations of paradigms and today, the field of organization science can be characterized as positive, interpretive, critical or postmodern science (Donaldson, 2003; Hatch & Yanow, 2003; Willmott, 2003; Chia, 2003; Hatch & Cunliffe, 2006).
The epistemological and ontological foundations of this study represent the post‐positivist paradigm. The ontological foundation of post‐positivism utilizes critical realism (Guba & Lincoln, 1994) or Campbellian realism (McKelvey, 1997). Both critical realism and Campbellian realism assumes realist ontology. The ontological problem that scholars in the field of organization and management science face is that most of the time we are simultaneously trying to accept the existence of idiosyncratic events and still pursuing the essential elements of justification logic
22
which requires non‐idiosyncratic events (Hempel, 1965; Suppe, 1977; McKelvey, 1997). As idiosyncrasy will not disappear, prediction, generalization, and falsification cannot take place. In effect, we are wearing a normal science straightjacket (Daft & Lewin, 1990). Thus, this dissertation subscribes to Campbellian realism, which allows social scientists to accept real‐world phenomena as criterion variables against which theories may be tested without denying the reality of individual interpretation and social construction (McKelvey, 1997). Put in other words, there is “real” reality out there but only imperfectly and probabilistically apprehendable (Guba & Lincoln, 1994); a critical realist is critical of the ability to know reality with certainty. The epistemological foundation of post‐positivism and this study is modified objectivism, meaning that the notion of complete distinction between the investigation and the phenomenon being investigated is not required.
Organization theory as a positive science aims at producing theories which explain how the world works (rather than being normative) and explanation takes place primarily in terms of causes that determine effects (Donaldson, 2001). Furthermore, the theory building and knowledge accumulation of the post‐positivist paradigm is based on the accretion of and addition to the edifice of knowledge (Guba & Lincoln, 2005). The following section explains the multiple‐lens perspective, which was adopted in the attempt to synthetize the relevant literature, the design of the overall research process and the research process of individual publication, all of which have been guided by the post‐positivist paradigm.
2.2 Multiple‐lens perspective
The name of organization theory suggests that there is a single, integrated, and overarching explanation what comes to the existence of organizations and organizing in general (Hatch &
Cunliffe, 2006). This might have been the case during the early years of organization theory (from the beginning of 1960s to mid‐1970s) when the field was characterized by a relatively high degree of consensus. In the early 1990s, the first concerns about knowledge fragmentation in the field of organization and management studies were raised. During the so‐called paradigm wars, Pfeffer (1993) was among the first to argue that due to scholarly specialization, knowledge creation tends to take place in silos (Pfeffer, 1993), leading to “Towers of Babel” (for opposing arguments and responses to Pfeffer (1993), see for example Cannella & Paetzold (1994), Perrow (1994), and Van Maanen (1995)). Ever since, scholars have drawn increasing attention to this danger which may limit the development of organization and management studies as a field (Pfeffer, 1993; Burrell, 1996; Okhuysen & Bonardi, 2011). Stated by McKinley et al. (1999, p. 643), “even a brief review of contemporary organization theory suggests that the discipline is composed of multiple, largely incommensurable theoretical frameworks or schools of thought”. Much of this holds true still today.
The pluralism in theories has been explained with various reasons, the most common one being the newness of the field. As many of the paradigms are quite recent, time is needed “for the weight of incremental refutations to finally scuttle strongly entrenched claims, paradigms, or schools”
(McKelvey, 1997, p. 354). Further (McKelvey, 1997) explains that there is reason to believe that the testability criterion has failed in organization science for two reasons: 1) prediction, falsification, and generalization do not work with organizational phenomena because much of it is idiosyncratic, and 2) subjectivist thinking is increasingly prevalent. In addition, pluralism has been explained with the presence of theory‐building approaches, such as borrowing and blending. As organization theory is an inter‐disciplinary field, having its roots in sociology, in the improvement of management control over work processes, and in economics (Fligstein, 2001), the scholars of this discipline are accustomed of borrowing and applying concepts and theories from its underlying disciplines. Whetten et al. (2009) distinguish between two types of borrowing: vertical and horizontal borrowing. Vertical, cross‐level borrowing refers to application of concepts and theories
23
within one discipline but on a different unit of analysis, whereas horizontal, cross‐context borrowing refers to the application of concepts and theories from one discipline to another.
More recently, McKinley (2010) discussed the trend of migrating from theory validation to new theory development, calling this the displacement of ends. Further, McKinley et al. (2011) label this as low heed in organization theory. They argue that organization theory operates as a low heed discipline, in which scholars take minimal heed of the contributions of their fellows. This condition of low heed is revealed in several specific aspects of the discipline: lack of attention to testing previously published theories, lack of emphasis on the replication of published empirical research, low standardization of construct definition and measurement, and a minimally‐developed division of labour between theorists and empirical researchers.
In an attempt to elevate the heed in organization theory, this dissertation adopts an integrative perspective in building the theoretical framework with an aim to create a comprehensive approach to ecosystem architecture. The literature on inter‐organizational design issues has been growing in recent years with few signs of convergence. The vehicle to build the ecosystem architecture concept is to combine lenses to build theory. This multiple‐lens perspective produces synthetized coherence by citing and drawing connections between works and investigative streams not typically cited together to suggest the existence of undeveloped research areas (for a further discussion on the textual acts and rhetorical practices to construct synthetized coherence, see Locke & Golden‐Biddle (1997)). The aims of such an approach are: 1) to bring attention to, and invent or reinvent as topics for inquiry, subjects that are implicit in other works, 2) synthetize recent advances and ideas into fresh theory, 3) to integrate theories or theoretical perspectives to provide a theoretical structure that was not there before, and 4) to clarify a central construct and to position the construct within a constellation of antecedents and outcomes in ways that not only generate communication among scholars but also spark research aimed at resolving the theoretical puzzles and empirical questions illuminated in the study (Gioia & Pitre, 1990; Locke & Golden‐Biddle, 1997; LePine & King, 2010;
Okhuysen & Bonardi, 2011).
Okhuysen and Bonardi (2011) discuss the different types of multiple‐lens perspectives in more detail. According to their definition and in order to build theory, different theoretical lenses are combined. These theories can be either compatible or incompatible in regard to their underlying assumptions, and they can either address substantially similar (or even the same) phenomena or different ones. More specifically, the degree of compatibility between the underlying assumptions of the lenses being combined is “the degree to which the theories brought together rely on similar or dissimilar individual decision‐making processes, organizational mechanisms, or other properties in the development of their explanations” (ibid., p. 7). The other dimension, the proximity of the theoretical lenses which are combined, refers to “the conceptual distance that exists between the phenomena that the lenses address in their original conception” (ibid., p. 7). Based on the degree of compatibility and proximity Okhuysen and Bonardi (2011) establish four different types of multiple‐lens perspectives, illustrated in Figure 1.
24 Figure 1: Different types of multiple‐lens perspective
Multiple‐lens perspective builds upon the argument that the use of any single research paradigm produces too narrow a view to reflect the multifaceted nature of organizational reality (Burrell &
Morgan, 1979; Gioia & Pitre, 1990). This perspective helps organization theory to step away from mere borrowing (Oswick et al., 2011) to highlight areas of overlap or complementarity for example (Okhuysen & Bonardi, 2011). In the identified conceptions, the boundary conception of ecosystem architecture itself has elements of this theory‐building approach. When building transaction cost economics, Williamson (1985) combined the lenses of economic changes (Hayek, 1945), governance (Barnard, 1938), bounded rationality (Simon, 1957), and institutional perspective (Commons, 1932) thus establishing an interdisciplinary research project in which law, economics, and organization theory are joined. This, in terms of Whetten et al. (2009), represents horizontal, cross‐context borrowing. Given the recent advancements in the theoretical underpinnings of this study, vertical, cross‐level borrowing is also witnessed in both the boundary concept and design concept more from firm‐level analysis to ecosystem or meta‐organization level.
Feyerabend (1980, p. 47) pointed out that “a scientist who is interested in maximal empirical content, and who wants to understand as many aspects of [his or her] theory as possible, will accordingly adopt a pluralistic methodology, [he or she] will compare theories with other theories . . . and [he or she] will try to improve rather than discard the views that appear to be lost in the competition”. As researchers from different disciplines generally use different methods and have different interests toward their object of study, approaches such as the multiple‐lens perspective allow them to see things that might not be recognized or might appear inconsequential to an insider (Carlile & Christensen, 2004). Further examples of such efforts include Porter’s (1980) work on strategy by combining business policy and industrial organization economics and Baldwin and
25
Clark’s (2006) work on modularity by combining options theory from finance and studies on product development.
In order to enhance the understanding of ecosystem architecture design, the suggested explanation builds upon organization design, sociologically‐rooted network theory, and economics, varying significantly in their proximity. However, the underlying assumptions of the lenses are compatible, implying a fit between the selected lenses to combine. Thus this integrative attempt to provide a new theoretical structure represents the type I multiple‐lens perspective. Focusing explicitly on these three conceptions may entail the exclusion of some potentially applicable theories. For example, from the economics side, literature on industrial organization has recently paid much attention to near‐by phenomenon, and two‐ or multi‐sided markets (Evans, 2003; Evans &
Schmalensee, 2007). This stream of industrial organization literature was utilized in publications I, II, V and VI. However, as the abductive interplay between theory and empirical observations progressed, it was noted that the excessive focus on the nature of the context and pricing as the key to encourage access and adoption of particular ecosystem (Gawer & Cusumano, 2012) and only vague explanations of the structural characteristics of an ecosystem led to the exclusion of the industrial organization perspective from the final framework. Similar arguments can be made about a second supplement of transaction cost economics, the property rights theory and incomplete contracting. Thus, the transaction‐cost‐economics‐(and evolutionary economics)‐based ecosystem architecture explanation was chosen.
In addition, perspectives such as organizational ecology (Okoli, 2007), value networks (Allee, 2000;
Kathandaraman & Wilson, 2001), and the applications of financial portfolio theory on the examination of inter‐organizational arrangements (Chandra & Shadel, 2007; Srivastava & Gnyawali, 2011) were excluded. Exclusion of some perspectives and justification of the selection is based on the purpose of the study to understand the actor‐structure duality, the structural properties of the architecture and the processes to manage and design the properties. The individual conceptions and their contributions are discussed in the sections 3.1 – 3.3. Further, section 3.4 presents a synthesis of the conceptions and discusses the implications of the conception for value creation and appropriation.
2.3 Reasoning and methods
How tempting would it to be to describe the research process in the following linear and sequential way: a problem was identified that is of interest to the research community, specific research questions or hypotheses were formulated relying on the theoretical resources of the research community, appropriate research strategies based on either or both deductive or inductive logic were elaborated, qualitative or quantitative measures were chosen and put to work, data compilation and analysis followed, and, with pluck and luck, plausible (or verifiable) inferences and conclusions resulted; end of story (Van Maanen et al., 2007). Van Maanen et al. (2007, p. 1146) point out that matters of how the actual research process proceeded are rarely discussed (at least in print) since in reality:
the flow on research is lengthy and uneven, is seen most clearly in hindsight, and perhaps most important, is contextually idiosyncratic, often chaotic, and always personal; how we arrive to conclusions is difficult to penetrate when publication norms do not favour the presentation of results in the manner in which they evolved and when the personal history of how the research process unfolded over time may be revised or forgotten as the project moves toward its final printed version.
26
The ironic description of the research process illustrates the overly simplified and idealized version of the interplay between theory and data which serves as the general form of representation of the results in academic writing (Van Maanen et al., 2007). Luckily, inch by inch the cognitive limitations of researchers are being incorporated into the discussion regarding the process of doing research (Alvesson & Kärreman, 2007; Van Maanen et al., 2007) and reasoning (Mantere & Ketokivi, 2013).
The interplay of theory and data does not usually follow this prescribed, almost magical, sequence (Van Maanen et al., 2007); it is more a sense‐making venture that evolves over time (Bailyn, 1977;
Weick, 1989; Alvesson & Kärreman, 2007).
Regarding the abductive reasoning logic of the dissertation entity, the following notions on the interplay of theoretical and empirical observations are in order. Understanding on the variation in ecosystem architecture was sought against the theoretical backgrounds of two‐sided markets, value co‐creation, and business models. For example, before turning to industry architecture literature, explanation about the phenomenon was sought from literature on industrial organization and two‐
sided markets (see publications I, II, V and VI). Further, in order to explain how and upon which template a firm organizes its transactions with suppliers, complementors and customers, explanation was sought from business model literature (see publications II and IV) before turning to literature on meta‐organization design. The theoretical explanations from two‐sided markets, value co‐creation, and business models literature were not seen as comprehensive enough to address the phenomenon of interest and thus they were excluded.
The research design of the overall dissertation and the individual publications is depicted in Figure 2, illustrating the different reasoning logics and related methodologies. In addition to this introductory part, the dissertation comprises six separate publications, each focusing on different aspects concerning the subject under scrutiny.
27 Figure 2: Research design of the dissertation
The essence of an argument is to proceed from premises to conclusions in a credible manner and to defend the claims made in these conclusions (Toulmin, 2003). In order to bridge premises with conclusions, scholars as scientist use various reasoning principles (Mantere & Ketokivi, 2013), namely deduction, induction and abduction. Mantere and Ketokivi (2013) explain the difference between deduction, induction and abduction in the following way. For deductive reasoning, the rule and the explanation are the premises and based on these, the observation is derived. Thus a conclusion about the particular is drawn based on the general. For inductive reasoning, the observation and the explanation are the premises and based on these, the rule is derived. Thus conclusion is drawn from particular to general. For abductive reasoning, the rule and the observation are the premises and based on these, the explanation is inferred if it accounts for the observation in light of the rule.
As depicted in the figure above, the overall research design can be characterized as interplay; the theoretical framework evolved simultaneously and interactively with empirical research (Dubois &
Gadde, 2002; Van Maanen et al., 2007; Dubois & Gibbert, 2010). As labelling the research deductive, inductive, or abductive does not by itself tell how the research process proceeded nor justify the methodological choices (Eriksson & Kovalainen, 2008), Table 1 further summarizes more precisely the research design adopted in complementary publications of the dissertation in terms of reasoning logics, methods and analysis, and the data used in the publications. The characteristics of the research phenomenon, the amount of prior research about it, and the nature of the research goals and objectives have guided the methodological and method choices of the dissertation along with the philosophical assumptions adopted. The individual publications have three different reasoning logics: abduction (publications I, II and VI), induction (publications III and IV) and
