Collective action for a common service platform for independent living services

i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 2 ( 2 0 1 3 ) 922–939

journal homepage: www.ijmijournal.com

Collective action for a common service platform for independent living services Fatemeh Nikayina,∗ , Mark De Reuvera , Timo Itäläb a b

Faculty of Technology, Policy and Management, Delft University of Technology, 2600 GA Delft, The Netherlands Department of Computer Science and Engineering, Aalto University, FI-00076 Aalto, Finland

a r t i c l e

i n f o

a b s t r a c t

Article history:

Objective: The paper aims to explain how and why organizations, providing assistive devices

Received 25 November 2012

and related web services for elderly independent living services, might be willing to collab-

Received in revised form

orate and to share their resources and data on a common service platform.

25 June 2013

Method: A theoretical framework from literature on collective action theory, platform

Accepted 26 June 2013

and business ecosystem concepts was developed to explain what factors influence inter-

Keywords:

in a case study of collaborative platform project for independent living services in Finland.

Independent living

Semi-structured interviews with the project managers and the decision makers of involved

organizational collective action for a common service platform. The framework was tested

Home care services

organizations were the primary source of data collection.

Information systems

Result: Strikingly, interdependency among the organizations was not found to be impor-

Collaboration

tant for collaboration in this case. Instead, we found that a central organization can play

Incentives

an important role in initiating, facilitating and encouraging collaboration among different

Leadership

parties. Moreover, we found more willingness for collaboration when the platform is aimed to be open to third-parties to complement the platform with additional services. Conclusions: Strategies of the platform leader and openness of the platform towards third parties are the most important drivers for collective action between organizations offering independent living services. Establishing common service platforms for independent living services requires explicit attention to these inter-organizational issues. © 2013 Elsevier Ireland Ltd. All rights reserved.

1.

Introduction

As the ageing population is growing across the world [1], there will be more elderly people demanding care services. Therefore, innovative healthcare solutions are required to not only provide elderly care services with less cost but also to improve the quality of life for elderly people [2]. Currently, several assistive device providers are developing products with related web services, each addressing a niche in the market, enabling elderly people to live independently and service providers to reduce costs. Examples of such assistive devices are watches,

∗

necklaces or sensors that are used for remote medical or safety monitoring as well as emergency alarming [3]. Several studies discuss how such technologies can facilitate independent living services for elderly people (with different health conditions) and support care givers in providing the services and saving costs [e.g. 4–7]. Many scholars also explore acceptance aspects of assistive technologies by elderly people and care service providers [e.g. 8,9]. The main problem though with existing technologies and devices is that they often run on different non-interoperable service platforms [10]. This makes it difficult for care service providers to share data and to bundle services and products from different device providers. The

Corresponding author. Tel.: +31 15 278 35 99. E-mail address: [email protected] (F. Nikayin). 1386-5056/$ – see front matter © 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijmedinf.2013.06.013

i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 2 ( 2 0 1 3 ) 922–939

fragmented nature of the market also increases the time and costs to develop and implement new services. The fragmentation in the independent living market could be reduced massively if providers of assistive devices and care service providers would collaborate and jointly develop common service platforms for their service offerings. Such a common service platform should give care service providers a standard interface to connect to multiple assistive devices as well as customers’ data. Service platforms have been extensively studied, especially regarding strategies of platform providers in opening up a platform or business ecosystems to third-parties [e.g. 11–15]. However, existing studies do not consider how multiple organizations can jointly set up and commercialize a common platform. Collaboration for a common service platform involves challenges, such as lack of trust among parties [16] and challenges to align business models and business processes [17]. The aim of this paper is to study what factors influence inter-organizational collaboration in the context of establishing common service platforms for independent living services. Setting up a common service platform can be considered as a collective action problem since (1) it requires several organizations to collaborate to realize a common goal [18,19], i.e. to establish a common service platform; (2) the common goal cannot be achieved individually [20], as current individual approaches cannot provide a holistic solution for elderly independent living. While collective action theory has been widely applied to study collaboration in different fields [e.g. 21–25], the principles of the theory have hardly been applied to study inter-organizational collaboration in the context of platforms [Exceptions: 26,27]. By connecting insights from collective action theory with literature on platforms and business ecosystems, we develop a theoretical framework with a set of factors that influence collective action for joint platform development. To test this framework, we conduct a case study on a project in Finland, which aims to enable offering a combination of independent living services for elderly on a common service platform. As collaboration for common independent living service platforms rarely takes place in today’s market [28], this case is intrinsically important to study. The paper proceeds as follows. Section 2 presents the theoretical framework with a set of factors influencing collective action. Section 3 provides the case study design. The results are presented in Section 4. Finally, findings are discussed and conclusions are drawn in Section 5.

2.

Theoretical framework

In this section, we provide an overview of literature on collective action theory, platforms and business ecosystems concepts to identify factors influencing collective action for common service platform development. Collective action theory, first developed by Olson [29], is used to explain how groups of individuals may collaborate for a common goal even if the incentives to do so are smaller than not collaborating. According to Olson, the dilemma of collective action is that when benefits of the common goal cannot be excluded from non-contributors, rational individuals will

923

not contribute to the common goal and they tend to free-ride on contributions of others. Concepts like critical mass, group size, network structure, motivations and selective incentives have been studied, in different contexts of economics, social and political sciences, to explain why collective action arises in one group and not in another group [For example see: 30–35]. The theory has been also applied in studying management of natural resources (referred to as common pool resources), in which several individuals need to collaborate for utilizing a common natural resource (e.g. a grazing land for cows) and preventing overexploitation (i.e. the free-rider dilemma) [For example see: 36,37]. In this research, however, we mainly borrow insights from studies on collective action in social and economics contexts. In the ICT domain, platforms are increasingly used to provide generic functionality on which a range of services may run [38]. Popular examples include smartphones, tablets, operating systems, application stores, and service portals. What all platforms have in common is that they mediate interactions between two or more groups of organizations, typically the service providers on the one side and the service consumers on the other [15,39,40]. The networks of organizations which are collaborating and competing around a technology are often referred to as business ecosystems. The concept, which was first introduced by Moore [41], is increasingly used to analyze inter-organizational networks in high-tech industries (e.g. computer and mobile industries). Business ecosystems exhibit dynamic characteristics in which companies, their roles and the structure of business ecosystems may change over the time [42]. In this paper we refer to organizations collaborating around a platform as the ‘platform ecosystem’.

2.1.

Platform openness

The first factor we consider to have an effect on the likelihood of collective action is platform openness. Platforms may be fully open towards third party service providers but often there are restrictions to who can offer services on the platform and who not. Given the diversity between rather closed strategies from platform providers like Apple and Microsoft and more open strategies from providers like Google, fierce debate has emerged as to how open a platform should be to foster collaboration. The platform openness can be seen from two perspectives. From a technical perspective, a platform is considered open if it provides the technical means for complementary providers (i.e. companies that provide alternative technology, products or services for the platform) to access the core functions of the platform, for instance through open APIs (Application Programming Interface) or SDKs (Software Developers Kits) [43]. From an organizational perspective, platform openness implies to what extent complementary providers are allowed to participate in the development, commercialization and usage of the platform [13]. Organizational openness may be reduced by platform providers through setting up rules and contracts on which players and services are allowed to be offered on the platform. How open or closed a platform is technologically and organizationally is critical for the growth and sustainability of the platform [44]. Keeping a platform closed helps platform providers to preserve

924

i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 2 ( 2 0 1 3 ) 922–939

competitive advantages of the platform and to have better margins [45]. While opening the platform may intensify competition among complementary providers which could foster innovation around the platform, severe competition increases uncertainties and thus reduces motivation of third parties to invest in the platform [46,45,13]. To strike a balance between collaboration and competition around a platform, platform providers are recommended to select organizations in a way that reduces competition and encourages organizations to collaborate [47]. The literature on platform openness mainly focuses on the case in which there is one single platform provider rather than when multiple organizations collectively develop a platform. As such, the first proposition of this study is: P1. The platform openness towards complementary providers impacts the likelihood of inter-organizational collective action for establishing a common service platform. More specifically, we study impacts of technical openness as well as organizational openness of platforms on collective action. Therefore, we break down the first proposition into two parts: P1-A. The degree to which a platform is technically open towards complementary providers impacts the likelihood of inter-organizational collective action for establishing a common service platform. P1-B. The degree to which a platform ecosystem is organizationally open towards complementary providers impacts the likelihood of inter-organizational collective action for establishing a common service platform.

2.2.

Heterogeneity and dependency in the ecosystem

One of the challenges in platform collaboration is that complementary providers often avoid collaboration with platform providers when they have comparable resources to each other, e.g. technical resources or knowledge. The reason is that complementary providers perceive platform providers as a threat that can enter into their businesses [44]. While the literature on platform mainly discusses the competition between a platform provider and complementary providers, competition can also happen among a group of organizations developing a common platform if they have comparable resources. This implies that having comparable organizations, with respect to resources and markets, in a business ecosystem increases competition rather than collaboration [47]. From a collective action perspective, many scholars hold positive views about the impacts of heterogeneity of resources on collective action. The heterogeneity of resources can be seen as the degree to which organizations participate in collective action vary in the types of resources (e.g. information, expertise, technology, market relationships) as well as amount of time and money available to them. Several authors have described heterogeneity of resources in a group of collaborating actors as an advantageous condition that can speed up the start of collective action, because it is probable that the

most resourceful individuals contribute to collective action and provide the common goal for others [31,33]. Only a few studies show that heterogeneity of resources has no effect on collective action when group members do not communicate [48]. This line of arguments suggests that having dissimilar resources positively influences collaboration among organizations. Therefore, the second proposition in this study is: P2. The heterogeneity of resources in a platform ecosystem increases the likelihood of inter-organizational collective action for establishing a common service platform. The likelihood of collective action depends not only on heterogeneity of resources, but also on the interests of participants in collective action [49,18]. Although participants share a common goal, their interests may vary across different groups. For instance, telecom companies and assistive device providers may have different (sometimes conflicting) goals for a platform for independent living services. We refer to such differences as heterogeneity of interests. Studies on collective action show that heterogeneity of interests can be detrimental to collective action because of likely conflicts of interests [50,51]. The conflicts of interests can fragment a group into groups of rivals which then makes it challenging to achieve consensus among members and this is not in favour of collective action. Therefore, our third proposition is: P3. The heterogeneity of interests in a platform ecosystem reduces the likelihood of inter-organizational collective action for establishing a common service platform. As argued above, having dissimilar resources and capabilities in the platform ecosystem can be beneficial in itself since it reduces the fear for competition. However, the incentives for collaboration are even higher if organizations also require each other’s resources in order to ensure sustainable productivity and innovation within the ecosystem [52,53]. From a collective action perspective, interdependencies in a network of organizations can solve the problem of free-rider in collective action because when interdependency is in place organizations find it rational to collaborate [33]. From these arguments, we propose that: P4. The heterogeneity of resources in a platform ecosystem increases interdependencies in the platform ecosystem. P5. The interdependencies in a platform ecosystem increase the likelihood of inter-organizational collective action for establishing a common service platform.

2.3.

Leadership

The importance of leadership for collaboration is stressed in all three streams of literature. From a collective action perspective, it has been argued that sharing a common goal is not a sufficient condition for collective action [29] and the presence of a leader is a prerequisite to start and mobilize collective action [54,55]. Leaders are required to attract

i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 2 ( 2 0 1 3 ) 922–939

contributors [55] and select the most potential organizations for collective action [33]. In platform literature, the notion of ‘platform leadership’ is used to refer to strategies of platform providers to encourage complementary providers to invest and develop products and services for a platform [44,56,45,57]. Studies on business ecosystems use the terms ‘keystone’ or ‘dominant player’ to refer to a central organization that manages a platform and controls activities of other organizations in a business ecosystem. Iansiti and Levien [53] distinguish a keystone from a dominant player in a business ecosystem in a way that keystone strategies encourage collaboration and facilitate innovation while dominator strategies exploit other organizations in a business ecosystem and drain all the business ecosystem’s value. Therefore, keystone strategies are more effective in creating and maintaining a healthy collaborative business ecosystem [58]. This line of arguments suggests that leadership can facilitate collective action for realizing a common service platform. We define leadership as the extent to which a central organization ignites, leads, supports and coordinates collective action in a platform ecosystem. As such, our sixth proposition is as follows: P6. The presence of leadership in a platform ecosystem increases the likelihood of inter-organizational collective action for establishing a common service platform.

2.4.

925

suggested as an effective mechanism to promote participation in collective action and eliminate the free-rider problem [59,36,35]. Selective incentives, which can be in the control of a leader, can be seen in two forms: (1) private benefits or award for those who contribute to collective action and (2) penalties or punishments for free-riders [35]. Despite the potential benefits of selective incentives, many studies discuss the problem of second-order freerider behaviour for selective incentives [60]. The second-order free-rider issue raises the issue who will contribute to the administration of selective incentives [61]. This issue is more crucial for self-organizing collaboration networks that do not have a leader to coordinate collaboration. Studies on collective action suggest that deploying selective incentives requires a central authority with high interests in collaboration to impose rewards and punishments [62,54,29]. This is supported in platform studies when selective incentives created by a leader encourage participation of complementary providers [44,56,45,57]. This line of arguments suggests that selective incentives created by a central organization can assist collective action among organizations for setting up a common service platform. Therefore, we propose the followings:

P7. The presence of selective incentives in a platform ecosystem increases the likelihood of inter-organizational collective action for establishing a common service platform.

Selective incentives

Typically, opportunistic behaviour of actors and the free-rider dilemma reduce motivations of organizations to participate in collective action because of the fear that others may freeride or because everyone prefers to free-ride on contributions of others [29]. Accordingly, selective incentives have been

P8. The presence of leadership is required to deploy selective incentives in a platform ecosystem.

Fig. 1 summarizes the propositions into a theoretical framework.

Fig. 1 – The theoretical framework.

926

i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 2 ( 2 0 1 3 ) 922–939

3.

Methodology

3.1.

Case study design

According to Yin [63], case study method is especially appropriate for answering questions of ‘how’ and ‘why’ and understanding a contemporary phenomenon of interest. The case study method is suited in this paper as we study how and why collective action can be facilitated between organizations developing a common service platform for independent living services. To do so, we require insights into the contexts, organizational relationships, knowledge and experiences of informants and practitioners in the field. These needs match criteria for the case study method [63–66]. For several propositions derived in the previous section it is yet to be explored whether there is a positive or negative effect on collective action, which implies that a deep and rich understanding of the domain is needed. A single case design enables such in-depth understanding and allows building theories by validating (i.e. confirming or disconfirming) and exploring theoretical propositions [67,68]. A more pragmatic reason to choose a single case design is that although independent living devices and services are abound, actual collaboration between multiple providers for a common service platform is scarce. The unit of analysis in this study is the platform ecosystem (i.e. the organizations that are involved in development of the common service platform) and their technologies for the common service platform. We are especially interested in the reasoning of organizations about how certain factors influence the decision of companies to join a common platform development project. These decisions combined explain why collective action has arisen. The theoretical framework developed in the previous section provides us with presumed factors (i.e. platform openness, heterogeneity of resources, heterogeneity of interests, interdependency, leadership, selective incentives) that influence collective action. However, we need to validate the framework in the domain. The framework helps to stay focused during the case study and avoid ‘information overload’ [66].

3.2.

Background on the case

In Finland, local governments (i.e. approximately 440 municipalities) are responsible for providing care services for elderly people. In the Espoo area, Active Life Village Ltd. (ALV) is a nonprofit organization which has been founded by municipality of Espoo city, Aalto University, Laura University and Helsinki Metropolia University of Applied Sciences to promote development and commercialization of innovative ICT technology for independent living services. Active Life Home is one of the main projects that ALV is working on and is partially funded by Tekes (the Finnish Funding Agency for Technology and Innovation). ‘Active Life Home’ is a collaborative common platform development project for elderly independent living services. The project aims to (1) integrate the assistive devices and related customer data into one common service platform to provide information services for independent living and (2) to set up a marketplace where solutions of multiple companies are

presented so that customized combinations can be selected based on each customer’s individual needs. Fifteen companies participate in the project that develop assistive devices and related web-service packages (i.e. medicine reminders, alarms, notifications, activity and sleep quality trend analysis, and location tracking) or provide social or security services. Moreover, three departments of Aalto University participate in the development of the platform. Although the university is involved, the aim of the project is to commercialize the platform and thus to move beyond the stage of research and development. The Active Life Home project can be considered a collective action situation for joint platform development for four reasons: (1) multiple organizations are involved in developing the platform, (2) developing the platform is the common goal of the organizations, (3) no organization can develop the platform individually and (4) the risks of free-riding are high, since organizations may withhold their resources in the platform development, but reap benefits after the platform is developed. As such, we conceptualize an organization’s decision to join the platform development as the decision to engage in collective action.

3.3.

Data collection

We relied on multiple sources of data for the case study. We interviewed organizations involved in the project to understand their reasons for joining the project and we mainly relied on their answers and discourses to validate the propositions. The interviews were semi-structured and the interview questions were designed to cover the propositions (see Appendix 1 for the interview questions). To deal with facades, we asked questions in different ways and asked follow-up questions to ensure disclosing of real information. We considered the following necessary conditions for selecting the interviewees: • The interviewee should be from a company collaborating in the development of the common service platform. Note that we interview relevant people from companies of which resources and capabilities are critical and irreplaceable for platform development as well as replaceable actors that are eager to participate in providing the platform. • The interviewee should be a high level decision maker or project manager who is involved in making strategic decision for the company. This enables us to ask the interviewee about how the decision on collaboration is made and what factors influence such decisions. • The interviewee should have a basic technical knowledge to discuss issues of collaboration created by technology architecture and infrastructure. A key informant from Aalto University helped identify potential interviewees. Interviews were conducted face-to-face (in January 2012) with four decision makers from assistive device provider and data integrator company, two project managers from Active Life Village (ALV – the leader of the project) and four academics with technical knowledge from Aalto University who are involved in the development of the platform. Each interview lasted one to two hours. We stopped the interviews

927

i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 2 ( 2 0 1 3 ) 922–939

Table 1 – List of interviewees (DI: Data Integrator; DP: Device Provider; ALV: Active Life Village; AU: Aalto University). Code

Organization

Role of the organization

Position of the interviewee

DI1

Playground Finland Ltd

Data Integrator

Co-founder

DP1 DP2 DP3

Beddit Arcticare Co. Vivago Company

Assistive Device Provider

Chief Technology Officer Managing Director President

ALV1 ALV2

Active Life Village Oy

Project Leader

Managing Director Project Manager

Platform Developer

Senior Researcher Project & Financial Manager Professor Professor and Initiator of the project

AU1 AU2 AU3 AU4

Aalto University

when no additional insights or information were presented. After an initial analysis and three months after the case study, we interviewed one of the project managers from ALV via telephone for a second time for additional information. Table 1 shows a detailed list of interviewees. In addition to interviews, we also consulted documents about the project, websites of involved organizations (for information about their resources and technologies), as well as scientific publications or reports of other researchers about the project. These documents were mainly used for the factual description of the case (i.e. understanding the platform architecture and the business ecosystem setting). The interviews were recorded and transcribed. Then, the interview transcripts were printed and read carefully. After that, all the transcripts were manually open coded in two rounds of coding; having the propositions in mind and paying attention to other possible explanatory factors (or issues) for collaboration (other than the ones in the propositions). The first two rounds of coding served as a basis to get an in-depth understanding of data. Afterward, all the transcripts were open coded for a third round using Atlas.Ti 6.2 software. The qualitative data analysis software facilitates data analysis by creating a structure for codes and memos. Moreover, the use of software provides a more systematic way of studying relationships in data which helps to avoid data analysing biases [66,69]. A code (i.e. label) was assigned to any relevant part of transcripts that constituted aspects related to the core concepts in the propositions (i.e. platform openness, heterogeneity of resources, heterogeneity of interests, interdependency, leadership and selective incentives). After the third round of coding, the quotations for each code were checked again to ensure that the underlying quotations actually stand for the assigned code; even if they have different choice of words or phrasing. The third round of coding resulted in 140 codes which include overlapping or redundant codes. To reduce the number of codes, all the overlapping codes were merged. In addition, ‘code families’ function in the software was used to assign code families to groups of codes that are conceptually related to each other. This function makes it easy to find a group of related codes. For instance, the code family ‘Technical Platform Openness’ were assigned to all the codes providing definitions or addressing requirements for a technically open platform (e.g. ‘Industry-standard interfaces’, ‘licence-based’, and ‘not-free’). It also happened that one code belonged to

different code families, which can be explained as different facets of the code [70]. Just assigning codes to text may demote analysis to merely a classification of concepts which is not sufficient for the purpose of interpretation [70]. As such, while doing coding memos were written to document the interpretations during the course of analysis. The memos were later used to develop a line of argumentation in discussion of data. As the aim of this study is to test causal relations in the propositions, code networks were also constructed during the third round of coding. The code networks visualize and clarify underlying relations and structures among codes that were inferred during the time of interpretation. In a number of cases, interviewees expressed different or even contrasting opinions about causal relations between two concepts in the propositions. For instance, interviewees held both positive and negative views regarding the impact of ‘platform openness’ on collective action. In such cases, not only the number of positive and negative views on a relation, but the contexts of those conversations as well as knowledge and/or experiences of interviewees were taken into account. Finally, based on the codes, code networks, the respective proposition was concluded to be supported or not supported and the direction of effects in propositions, if not defined in the proposition, were also indicated, i.e. a positive impact, a negative impact, no impact. To illustrate the findings, we provide a selection of quotes. Intermediate conclusions were validated throughout the process through constant communication and discussion with one of the board members of the project.

4.

Results

4.1.

Active Life Home service platform

Fig. 2 shows the Active Life Home service platform. The platform realizes integration on three levels: The user interface level (ALH Portal), the information level (Activity and Health Record, AHR) and the device level (Home Gateway, VALPAS). The ALH portal has the user database of different user groups, like administrator, elderly persons, their families, nurses, doctors and other caregivers. The portal manages access rights to various services and provides single-sign-on to the providers’ own systems. The user interface has been implemented on the portal so users have access to all services from one screen. ALH Portal is developed by Aalto University.

928

i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 2 ( 2 0 1 3 ) 922–939

is a communication device that links and transmits information between care devices at home and servers of the device providers. This means that elderly people with multiple devices would end up having multiple gateways in their homes. To avoid the added cost of gateways and their management, the ALH framework is providing integration at the home level, using VALPAS home gateway which is developed by Aalto University.

Fig. 2 – Active Life Home platform.

The Activity and Health Record (AHR) is a key integrator of data collected of the elderly people and their assistive devices. Each device provider has opened their server’s APIs to enable collection of data into the common database, i.e. AHR, provided by Playground Ltd. The decision on what needs to be collected on AHR is made by agreement between the device providers and ALV. AHR, also accessible from the ALH portal, shows the collection of data, the current status and recent events of a selected customer. Most of the device providers in this project use home gateways to receive information from devices at home. A gateway

4.2.

Findings and conceptual analysis

4.2.1.

Platform openness

Fig. 3 presents definitions of platform openness from a technical perspective and how it influences collective action. The nodes represent codes and for each code N:M represents the number of times repeated in the interviews (N: groundedness) and the number of links to other codes (M: density). The red double arrows show the relations between code families. Four interviewees from ALV, assistive device providers and Aalto University stated that the platform is technically open and they mainly defined openness in terms of open industry standard interface, meaning that once the platform is developed complementary devices or services can be integrated to the platform on the IT system level. According to the same interviewees, the main reason for using open standard interfaces was to gain more benefits by extending the functionalities of the platform and sharing data with third-party service providers to get more value out of the data. Despite the agreement on using open standard interfaces, two respondents from Aalto University remarked that still there is a need to identify what interfaces should be defined. Identifying interfaces needs agreements on what need to be shared on Activity and Health Record (AHR) database and what data should remain on the device providers’ servers. When we asked interviewees how data and interfaces are accessed and used by third-parties, two interviewees from device providers said that access to the platform will not be free. However, as there was not any specific revenue model for the platform at the time of interviews, interviewees expressed different opinions on how platform’s resources can be utilized by third-parties. One interviewee from the device providers believed that the platform should be licence-based meaning

Fig. 3 – Technical Platform Openness (TPO: Technical Platform Openness; CF: Code Family).

i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 2 ( 2 0 1 3 ) 922–939

that there should be a fee for every time that data is being used. Two other respondents from the device providers discussed a combined model of free and for-pay services for the platform to reduce barriers but also ensure sustained revenue for the platform. With regard to the impact of technical openness on collective action, many interviewees believed that they would gain more benefits by opening the platform to third party service providers. Moreover, all interviewees from the device providers group perceived the platform openness as a driver for collective action. Similarly, one respondent said “the companies are more willing to take part in the project when they know it is an open platform [. . .] many of the companies have done closed platform development and they have seen that this is not the smartest way for the future” (ALV1). He further elaborated that it is not clear what kind of companies or products will be emerging in a few year time, that’s why having this type of open platform gives the device providers, involved in the development of the platform, the opportunity to integrate new products or services to their whole service offering. In general, enabling an open service platform (i.e. open standard APIs, but not open source or free) positively influenced companies to work collectively. Interviewees believed that this is the only way for them to extend their current businesses, stay competitive and provide their customers with a total solution. Therefore, the first part of proposition 1 is supported in this case and the impact of the platform openness on collective action appeared positive. Fig. 4 shows the code network

P1-A. The degree to which a platform is technically open towards complementary providers impacts the likelihood of inter-organizational collective action for establishing a common service platform. Supported-a positive impact

for platform openness from an organizational perspective. All interviewees generally pointed out that there are rules and agreements for new companies joining the platform. First of all, there should be an agreement among members (at least more than half of the members) to accept new companies. Moreover, new companies should be complementors not direct competitors of any existing members of the project. For instance, when the project was already halfway, ALV introduced a new company that could bring value to the project and then upon agreement of most of the current members, the new company joined the project. Four interviewees from the device providers and Aalto University also remarked that the platform is closed to free-riders in a way that the current members will not allow a new company to join the project without investing money or resources. They further explained that the current members are putting efforts and resources to develop the platform and if a company wants to join without contributing to the platform development, then there is no point for collaboration. According to two respondents from device providers, having a large number of parties increases difficulties in collaboration for the platform development. One of the reasons according to an interviewee from device providers is that the more a

929

platform is organizationally open the more the chance that there are competitors in the business ecosystem and this increases the conflicts of interests among the parties (i.e. an interaction effect between organizational platform openness and heterogeneity of interests). Therefore, attributing strict rules to the organizational openness of the platform during development is important to reduce competition and encourage collective action in the business ecosystem. Nevertheless, according to one interviewee from Aalto University once the platform is developed accepting new companies that have complementary solutions for the platform is required to be competitive otherwise the platform will die in a few years when new companies with new solutions emerge in the market. Regarding the impact of organizational openness of platform on collective action, one respondent put that the organizational openness is “a stimulating factor for collaboration [. . .] because the way we have marketed the project for the members is that we need to have an open platform or industry standard [. . .] this actually implies that the companies were interested in that and thus they joined” (ALV1). Moreover, two other interviewees remarked that organizational openness (to complementary providers) matters for collaboration because “it enables us to complement our service offering” (DP2) and to have “more credibility in the market” (ALV1) by working as a group than as an individual company. Although respondents argue that the platform is organizationally open to new members to complement the platform once the platform is developed, they still impose rules for accepting a new member. Therefore, based on our definition of organizationally open platform, the platform is not organizationally open. In fact, device providers are even very reluctant to accepting a new member if it has similar resources to the existing members. Therefore, we conclude that having a platform with no control or limited organizational rules for accepting new members would negatively influence the collaboration in this case. Based on this finding, the second part of proposition 1 is supported in this case and the impact of the organizational openness appeared negative.

P1-B. The degree to which a platform ecosystem is organizationally open towards complementary providers impacts the likelihood of inter-organizational collective action for establishing a common service platform. Supported-a negative impact

4.2.2.

Heterogeneity and interdependency in the ecosystem

To validate our second proposition, first we analyzed the documentary information of the project for resources that each group actively contributes to Active Life Home project. Table 2 shows a summary of resources for each group of participants. The table suggests a high heterogeneity of resources in the network. Apparently, each group contributes different resources (i.e. finance, assistive devices and services as well as technology and market knowledge) to the platform development. The participants are either non-profit organizations (i.e. ALV, Aalto University) or small device providers with limited financial and technological resources. Broadly speaking, each

930

i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 2 ( 2 0 1 3 ) 922–939

Fig. 4 – Organizational Platform Openness (OPO: Organizational Platform Openness; CF: Code Family).

Table 2 – Resources heterogeneity. Participants Tekes Active Life Village oy

Device Providers Playground Aalto University

Resources for Active Life Home project Finance Inter-organizational relationship with potential customers Planning and coordinating capabilities Business and market knowledge Marketing capabilities Assistive devices and web services packages Technical expertise in data integration Technical expertise in developing the platform

group of participants has parts (not all) of required resources to contribute to the platform development, which is one of the reasons for collective action. When we asked interviewees how different they perceive their organization from the other involved organizations in term of core business resources, we heard differing perceptions. Fig. 5 shows the respondents’ opinions about the resource heterogeneity in the business ecosystem. On the one hand, the interviewees from ALV and one interviewee from Aalto University believed in high heterogeneity of resources among the device providers, i.e. Dissimilar technical solutions, and the project manager from ALV put that having dissimilar solutions are important (but not necessary) for collaboration (i.e. a positive influence on collective action). He reasoned that “everyone must have different partial solutions for the final solution (i.e. common service platform), but it is not necessarily needed that the companies be totally different from each other” (ALV1). Obviously, heterogeneity of resources reduces competition in the business ecosystems as actors have different resources for the final solution. On the other hand, two respondents from device providers and two from Aalto University argued there is a low heterogeneity of resources in the business ecosystem. According to the respondents, the device providers are not only similar in terms of being small companies with limited financial resources; they also have overlapping functionalities in devices (e.g. using

same technologies like location tracking sensors). The issue is that “the overlapping functions are often easy and cheap to offer and thus all the device providers want to include such functionalities in their products” (AU2). Including overlapping functionalities in the devices then increases the price for customers, as customers have to pay the costs of same functions in different devices. As such, this “may result in rejection of the solution because of high price” (DP3). Moreover, it was argued by many respondents that the similarity of devices among companies in the business ecosystem increases competitions and thus negatively influences collective action (AU2, AU3, AU4, DI1, and DP3). The issue of competition among device providers and its impact on collaboration and sharing information is illustrated by the following quotes: “In this project, the companies are more or less the same (i.e. all start-up or small companies with limited financial resources) [. . .] also some of them are clear competitors, so they are not opening their interfaces, they are not telling what they are going to do or how they marketing and so on, because of their conflicting interests.” (AU2) “If there were no direct competitors in the business ecosystem, they would be more open for joint platform development in this context, but now they cannot disclose some of the relevant activities that they are doing.” (AU4) The findings suggest that the lack of heterogeneity of resources in the business ecosystem increases competition and thus influences organizations in opening up their interfaces (i.e. an interaction effect between technical platform openness and heterogeneity of resources: the more heterogeneity of resources, the more technical platform openness). Although majority of respondents from Aalto University believe that low heterogeneity of resources among device providers (i.e. limited financial resources and similar technological resources) hampers collective action, two respondents from the device providers argue that having partners with similar resources (even though not preferred) does not influence their decisions for collaboration, i.e. no influence

i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 2 ( 2 0 1 3 ) 922–939

931

Fig. 5 – Resources Heterogeneity (HR: Heterogeneity of Resource; CF: Code Family).

on collective action. They explained that developing the common platform is only one goal of joining the project and they joined the project also for other reasons, like marketing opportunities or visibility in the market. In general interviewees believe that having a heterogeneous network of organizations with different amount of financial resources and diverse technical solutions is more favourable for collective action for two reasons: (1) dissimilar technical solutions reduce competitions and increase collaboration in the business ecosystem; (2) different financial resources increase the chance that a resourceful company in the business ecosystem invest more than other to foster development of the platform. While involvement of actors with more financial, technical or organizational resources in the business ecosystem could facilitate the process of development and also attract customers for the platform, there is a risk that such actors use their positions to impose their technologies or decisions to other participants in the project and this is not encouraging for the start of collective action. Although we did not find a high heterogeneity of financial resources in the business ecosystem, the business ecosystem exhibits heterogeneity of technical resources (expect for a few similarities between two device providers) and organizational resources (i.e. relationships in the market). From the findings, we conclude that heterogeneity of resources among the participants has a positive influence on collective action in this case. Therefore, proposition 2 is supported in this case. To analyze the

P2. The heterogeneity of resources in a platform ecosystem increases the likelihood of inter-organizational collective action for establishing a common service platform. Supported

third proposition, we asked interviewees why they are participating in this project and they provided a variety of reasons, ranging from extrinsic to intrinsic interests (see Fig. 6). The primary interests for the device providers are to be visible in the market, have wider range of offerings, obtain a

strategic position in the market and create new business values. They expressed interests in business opportunities that the project can create. The following quotes illustrate marketing and business interests of the devices providers: “Marketing was one of the main reasons for us to be involved in this project, because the project was running some demonstration facilities and ALV was inviting guests from municipality of Espoo and other places and we could also use the demonstration room for our guests, so it was a good opportunity to use the facilities and have some marketing.” (DP2) “It might be going to be collaboration with three big municipalities in Finland, so of course we want to be involved” (DP3) Respondents from ALV mainly expressed interests in collaboration-oriented activities (i.e. networking, creating benefits for society). According to one interviewee from ALV, “the whole company was established for the purpose of establishing a collaborative service platform for elderly care services” (ALV1). Another interviewee from ALV emphasized that the main goal of the organization is to solve issues of isolated service offerings by gathering companies to complement their offerings and create value together. Similar to ALV, the respondent from Playground (i.e. the data integrator) also pointed to the need for collaboration among companies to complement their offerings and to create new value. The interest in collaboration for Playground is illustrated in the following quote: “I believe in the value proposition that we will be able to create together. We don’t have it yet, but we are working on it and I believe it’s going to be a great value proposition. We see that working together is going to provides us with a good business” (DI1). As Playground is not offering any products or services in the elderly market, collaboration is a practical way for them to get involved in this market. The interviewees from Aalto University were particularly interested in the project to access realistic use cases to apply their research and update their information, to solve technical problems, provide innovative solutions and in last but not least to publish papers.

932

i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 2 ( 2 0 1 3 ) 922–939

Fig. 6 – Interests Heterogeneity (CF: Code Family; CI: Collaboration-oriented Interests; NBI: New Business Interests; KI: Knowledge Interests; MI: Marketing Interests; STI: Strategic Interests).

Fig. 7 – Conflicts of Interests (CF: Code Family; CoI: Conflict of Interests).

Despite the interests’ diversity among participants, when we asked respondents about any conflicts of interests in collaboration, the respondents from ALV believed that there are no conflicts among participants in the project (see Fig. 7). Apparently, each group benefits from the project in different ways. However, four interviewees (AU1, AU2, AU4, and DP3) raised the issue of conflict between research vs. market interests. According to interviewees, Aalto University is a research institute with no interests in the market. Therefore, once the platform is developed, Aalto University is not going to provide any support or maintenance for the platform. This, then, would disappoint ALV as their expectation is not fully met. The following quote illustrates this potential conflict: “There is a potential conflict which is not articulated properly. ALV is providing a platform, which is outsourced to Aalto University (responsible for developing the platform). So ALV might expect Aalto University to develop or deliver a ready to use platform. However, this is not the university role, the university’s role is to do research and deliver research results perhaps in form of prototype or even saying that this type of things do not work based on valid results. That is a kind of potential reason for conflict as a result of roles and expectations, not relationship between participants.” (AU4) According to interviewees from Aalto University, the potential conflict observed in the project and mentioned several times by respondents may impede the continuance of collective action in the subsequent phases (i.e. pilot, implementation

and commercialization); especially when the next phases involve high costs and market uncertainty. Despite considerable heterogeneity of interests among the groups of participants in this case and its likely negative influence in continuance of the project, the members started collaboration with each other. Apparently, the heterogeneity of interests among participants in this case has not influenced collective action, at least in the beginning of the project. Among the

P3. The heterogeneity of interests in a platform ecosystem reduces the likelihood of interorganizational collective action for establishing a common service platform. Not supported – no impact

participants in the business ecosystem, ALV could contribute more to the platform development in the following way: it has a strategic position in the ecosystem because of its relationship with potential customers (e.g. municipality of Espoo) and extending the network (i.e. organizational resources). According to three interviewees from Aalto University and two interviewees from device providers, such organizational resources make ALV an attractive partner to collaborate with. The following quote illustrates this: “ALV is a magnetic channel because of its direct connection with municipality of Espoo and that really makes ALV different from others, in terms of extending the network and accessing to customers.” (DP2) This finding implies that interdependency among the devices providers and ALV is a result of heterogeneity of resources in the business ecosystem. This supports the fourth proposition: Fig. 8 presents the respondents’ answers about interdependency in the business ecosystem. On the one hand, all interviewees from the device providers pointed that they need each other for developing the final

i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 2 ( 2 0 1 3 ) 922–939

933

Fig. 8 – Interdependency (CF: Code Family; D: Dependency; PG: Playground co.).

P4. The heterogeneity of resources in a platform ecosystem increases interdependencies in the platform ecosystem. Supported

solution (i.e. the common platform). Aalto University depends on the device providers to access the interfaces of devices and services and then they can independently develop the platform. Likewise, Playground needs data from the device providers to do the integration and the device providers depend on PG for the data integration. Nonetheless, a majority of respondents (seven interviewees) argue that they are loosely dependent on each other as companies are replaceable by other companies providing similar solutions (especially the device providers). On the other hand, two respondents (ALV1, DI1) argued that there was no interdependency at the beginning of the project and interdependencies appear as a side-effect of collective action for the platform. This finding is illustrated in the following quote: “there are things that create binding in this consortium over time, one is the technical platform: when companies agreed on technical approach and create a platform, then it is easier to work with that (and with partners) rather than just forget it and start working on a new platform (with new partners)” (DI1) Only one respondent from ALV pointed out that getting finance from Tekes is critical because such funding creates dependency in the project. However, we heard from three

interviewees that the funding is only important for Aalto University and not for the device providers. The findings indicate that even in the absence of strong interdependencies among companies in the beginning, collective action started. Strikingly, interdependencies did not appear to be necessary for collective action to arise. In this case other factors explain collective action in the absence of interdependency.

P5. The interdependencies in a platform ecosystem increase the likelihood of inter-organizational collective action for establishing a common service platform. Not-supported

4.2.3.

Leadership

All interviewees from device providers and Aalto University perceived ALV as the leader of the project. The respondents pointed several responsibilities of ALV in the project (see Fig. 9): to support finance, drive development, facilitate networking and marketing and coordinate collective action among participants by aligning interests of parties, managing conflicts of interests and creating incentives for collective action. The following quote illustrates how ALV lead the project: “ALV drives the project with soft values [. . .] not hard values like forcing the companies to participate and/or bringing lots of money on the table and say just do this plan [. . .] ALV looks at different partners’ interests and tries to align the interests so that we can work with parallel interests” (DI1)

Fig. 9 – Responsibilities of the leader in the project (CF: Code Families; L: Leadership).

934

i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 2 ( 2 0 1 3 ) 922–939

Fig. 10 – Selective Incentives (SI: Selective Incentives; CF: Code Families).

The responses from the device providers were aligned with the responses of two interviewees from ALV who described their organization as the coordinator, facilitator and driver of the platform development project. According to the managing director of ALV, the mission of the organization is to promote and support common interests of the participants. These findings show the importance of leadership for collective action in this case. Despite the general positive opinions of interviewees about the role of ALV in the project, one of the device providers criticized ALV for lack of knowledge about customers’ needs. The interviewee believed that to ensure the market success of the platform, beside the integration, the platform should address real needs of customers and this needs to be done by the leader of the project. There were also comments on the type of organization (i.e. ALV is a non-profit organization). Two respondents from the device providers and one from Aalto University discussed the positive influence of ALV as a non-profit organization on collaboration. According to the device providers, ALV does not have any incentive against the device providers and this reduces corporation risk for parties and thus creates trust in the business ecosystem. Moreover, ALV does not impose a specific technology and force the other companies to use that. However, according to a number of interviewees, a non-profit organization in the lead also imposes limitations to commercializing the platform as (1) it does not have financial resources to ensure that the project will evolve in the market and (2) it is not their objectives to have revenue from the project. The following quote illustrates the differences between a non-profit and a commercial company in the lead: “It (a non-profit organization in the lead) makes the project more risky [. . .] the question is that how to make sure that this project really develops and evolves in the market. Usually one large commercial player can attract the customers and make sure that the ecosystem is healthy. In this case, you don’t have guarantee for the market part” (AU3) Although ALV as a non-profit organization may not guarantee the future of the platform, most of the companies asserted that the collaboration started just because of ALV. In fact, if there were not such a central actor to promote and coordinate the project, it was unlikely that these small companies get to know each other and start collaboration for a common

platform. Therefore, we can conclude that leadership positively influenced collective action in this case. P6. The presence of leadership in a platform ecosystem increases the likelihood of inter-organizational collective action for establishing a common service platform. Supported

4.2.4.

Selective incentives

Fig. 10 shows impacts of selective incentives on collective action in the ecosystems. Although Tekes provides a partial funding (as a selective incentive) for all the participants in this project, according to respondents from device providers and ALV, the funding is small considering the large scale of the project. Even companies may need to invest money for the development. Therefore, the financial incentive is not strong enough to encourage the device providers for collaboration. However, the funding for Aalto University covers 70% of research costs, according to the financial manager of Aalto University. Not surprisingly, one interviewee from Aalto University put that: “We cannot choose what we do. There is funding available for this project, so we do this [. . .] However, what makes this project interesting is that it is also not very expensive to study; all we need is just some PCs and components to keep project up and running.” (AU2) As research costs are not fully covered by the fund, the university could only do these types of project only if there are other similar projects that they could spend money from them, as said by two interviewees from Aalto University. Even though the interviewees from ALV assumed that funding is important to encourage the device providers to participate in the project, none of the interviewees from the device providers considered funding as a really stimulating factor for collective action. In fact, we heard a lot from the device providers that the added value of the platform to their separated offerings gives them a competitive edge compared to companies with no integration into the platform. Therefore, the competitive advantage of accessing to the platform is a stronger incentive, though a non-tangible incentive, to drive collective action.

i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 2 ( 2 0 1 3 ) 922–939

935

Fig. 11 – Summary of the case study’s results.

The findings for selective incentives imply that one underlying reason for the start of collective action in this case (even in the presence of heterogeneity of interests and absence of interdependency) is the presence of selective incentives. The

P7. The presence of selective incentives in a platform ecosystem increases the likelihood of interorganizational collective action for establishing a common service platform. Supported

findings about the role of leadership showed that the majority of respondents believe that ALV is responsible to support finance and create incentives for companies to work collectively (see Fig. 9). In fact, inclining companies to believe that the platform gives them a competitive edge over other companies may not be done without a leader. This supports our final proposition: Fig. 11 summarizes the findings of this paper about

P8. The presence of leadership is required to deploy selective incentives in a platform ecosystem. Supported

factors influencing collective action. The dashed lines show new relations (i.e. interaction effects between factors) found during the course of analysis.

4.3.

Alternative explanations

In addition to the eight propositions that we explored, respondents also pointed to other issues influencing collective action in this case.

According to a majority of respondents, one main issue in this project is the lack of a value proposition that defines the value and benefits of the project for the companies. Moreover, there are not agreements on division of revenue among the participants. Defining a business case which clarifies the financial equations would have been increased the companies’ motivations for active involvement in the project, as put by an interviewee from Aalto University. Another issue raised by respondents from Aalto University is the conflict between short term and long term plans, which creates difficulties in collaboration when value of the platform will be achieved in a long term. The following quote illustrates this issue: “The companies in this ecosystem are SMEs, so they have a very short life panel and looking at very short term issues, but we are looking into longer term issue. Trying to match these perspectives is a bit difficult because what we might ask is something that is not needed for everyday actions while they are very busy with making the income to survive this month or next month and that’s make it sort of difficult, so we have to adjust our way to work so that we can fill it in their time cycle.” (AU3) Most of the respondents were also concerned about the lack of customer cases in public, private and consumer sectors. They believed that uncertainties over the adoption of the platform may threaten the continuance of collective action for commercialization. While the municipality of Espoo city (i.e. a co-founder of ALV) is a potential customer for the platform, it is uncertain if the municipality would adopt the platform for service offering, because of high costs of integrating the platform into their current systems for care services. These findings suggest that uncertainties over business models, adoption and outcomes of the platform also influence companies’ decisions for collective action and long term investment in the platform.

936

5.

i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 2 ( 2 0 1 3 ) 922–939

Discussion

We found that four types of factors are relevant in explaining why collective action may arise: (1) the technical and organizational openness of the platform; (2) the heterogeneity and interdependency of the business ecosystem; (3) the extent of leadership in the business ecosystem; and (4) the selective incentives for participating in platform development. The findings regarding openness of the platform are in line with existing studies on platforms’ openness in mobile and computing industries, which describe strategies of platform providers to balance platform openness and closedness to encourage collaboration and innovation while keeping competitive advantages over third-parties [71,45,45]. The finding regarding interdependency in the ecosystem is contrary to previous studies on inter-organizational networks that show high level of interdependency among actors in providing services [72]. Perhaps we should distinguish between interdependency for service offerings and interdependency for platform development, as the phasing of these two are different (commercialization vs. development phases). In a service offering, there are usually customers demanding a service and there is a chain or network of providers that need each other to address customers’ demands. However, during the development of a platform, there are no customers and unless a company offers a critical resource for the platform (so called tier-1 providers), the companies in the business ecosystem are replaceable (i.e. tier-3 providers) and not interdependent [73]. Consistent with earlier research on platform leadership in other industries, e.g. computing industry [44,74], platform leadership was found to be important to ignite collaboration and support growth of the platform and the business ecosystem. We also found that, contrary to previous research that advocate the importance of tangible selective incentives (e.g. financial inducements) for organizations [75], non-tangible incentives in this project (i.e. competitive edge over companies with no integration platform) appeared to be more important for the organizations to start collaboration. This implies that providing financial incentives (i.e. governmental funding or subsidies) would not be the only way to motivate actors to work collectively. Other ways could be that: (1) governmental organizations (e.g. municipalities in this case) support adoption of joint development projects; (2) governments change regulations regarding offering independent living services with the aim of increasing market competitions and encouraging collaboration; (3) governments need to also support funding for commercializing development projects.

6.

Conclusions

This paper shows that collective action for developing a common service platform for independent living services depends on a range of technical and organizational factors. Most importantly, we found that strategies of a platform leader in bringing parties together and creating a trustworthy environment are essential for facilitating collective action. Moreover, balancing

the openness of a platform is critical as a fully open or fully closed platform are both not favoured by participants in a collaborative platform project. To a lesser extent, two other types of factors played a role. Heterogeneity of resources and interests of participants, and resulting interdependency, played a minor role to facilitate collective action. Selective incentives played a role as well to achieve collective action, but mainly regarding intangible incentives such as the vision of having a highly competitive common platform. Tangible incentives such as subsidies were present in the case, but did not play a decisive role to establish collective action. Introducing independent living services has policy implications, like the need for coordination between organizations involved in this domain [76]. This paper deepens our understanding of factors that need to be taken into account by practitioners and policy makers when planning to promote platform collaboration for independent living services. The importance of providing incentives for small companies to collaborate, adopting effective leadership strategies and avoiding competitors in the ecosystem are among the crucial factors that need to be considered carefully. Moreover, instead of being concerned or scared of a central actor, policy makers should stimulate openness, both in technological and organizational terms, and both for the development and commercialization of the platform, by providing financial means, regulations and innovative procurement. The relevance of regulations and policy has been also emphasized in other studies in this domain [2,77]. The paper contributes to collective action theory by developing a theory that explains what factors influence collective action for establishing a common platform. While the theory of collective action has been previously applied to the adoption and diffusion of information systems [78,79], it has not yet been widely applied to analysing the providers of common platforms [Exceptions: 26,27]. The limitation of this study, which is the drawback of all single case studies, is that we cannot provide a generalized conclusion from our findings [63]. However, this project was essentially important to study considering the lack of platform collaboration in the independent living domain. Although, we have identified the factors that influence collective action for common service platform development, this does not guarantee collective action for actual commercialization. For instance, one can question whether a non-profit organization will be able and motivated to lead the platform commercialization or whether the heterogeneity of interests will create conflicts in the commercialization. Such difficulties of moving from development to implementation and commercialization have been also discussed by other researchers in the e-health domain [80,81] and other empirical domains [82,83]. Moreover, the type of collective action employed, and the depth and longevity of the collaboration are relevant constructs and may well be an area for further study [84].

Author contributions F. Nikayin designed the study, conducted all the interviews, analyzed the data and drafted the manuscript. M. De Reuver supervised the study and co-authored the manuscript. T. Itälä

937

i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 2 ( 2 0 1 3 ) 922–939

Summary points Already known • The ageing population as well as the costs of care services is increasing. • Telecommunication technologies and increasing number of advanced assistive devices (for people with different health conditions) can help care providers to reduce costs and enable elderly people to live independently at their home. • One of the barriers in providing independent living services is non-interoperable platforms and technologies used by different parties. Added to our knowledge • Insights in organizational perceptions about important factors that encourage inter-organizational collaboration for establishing common platforms for independent living services. • Government can play an important role in creating incentives for small/medium companies to participate in joint platform projects. • The presence of a leader (a central actor) is crucial to bring different independent companies together and encourage them to share data and services on a common platform.

assisted in the interviews, and contributed to Sections 4.1 and 5.

Competing interests

3. What were the main drivers for your company to start this collaboration? 4. Are there any interdependencies between partners in this project? How does it affect collaboration? 5. In how far were the partners in this project different? How do these differences influence collaboration? 6. Have you encountered any difficulties during the collaboration? 7. Do you provide the technical information of the platform for third-parties to develop services/applications? How and why? Does that influence collaboration? 8. Is the project open for new parties to join? Are there any rules or agreements for accepting new members? 9. How is the collaboration coordinated? How does it influence collaboration among partners? 10. Are there any subsidies or funding available for the partners? 11. How do you see the future of the service platform in terms of adoption by customers or care service providers?

Heterogeneity of Interests Interdependency

Heterogeneity of Resources

Conflicts of Interests

Technical Openness

Organizational Openness

Platform leadership

Selective Incentives

Future adoption

The authors declare no conflicts of interest.

Acknowledgements

references

The authors wish to thank Harry Bouwman for constructive comments on an earlier version of the paper, Matti Hämäläinen for his extensive support to the research and access to the case, two anonymous reviewers of this journal and the anonymous reviewers of the 25th Bled eConference, where a preliminary version of this paper was presented. The authors would also like to thank interviewees from Aalto University, Active Life Village, Playground, Beddit, Vivago and Arcticare.

Appendix 1. Interview questions

Questions 1. Please tell me about the role of your company in this project. 2. With what companies are you actually collaborating on a daily basis?

Related concepts General Background Network Structure

[1] World Health Organization, Ageing and Life Course (accessed January 2012). Available from: http://www.who.int/ageing/en/ [2] K.A. Stroetmann, Achieving the integrated and smart health and wellbeing paradigm: a call for policy research and action on governance and business models, Int. J. Med. Inform. 82 (4) (2013) e29–e37. [3] H. Oh, C. Rizo, M. Enkin, A. Jadad, What is eHealth (3): a systematic review of published definitions, J. Med. Internet Res. 7 (1) (2005) 32–40. [4] C. Guilfoyle, R. Wootton, S. Hassall, J. Offer, M. Warren, D. Smith, M. Eddie, Videoconferencing in facilities providing care for elderly people, J. Telemed. Telecare 8 (Suppl. 2) (2002) 22–24. [5] P. Boissy, H. Corriveau, F. Michaud, D. Labonté, M.P. Royer, A qualitative study of in-home robotic telepresence for home care of community-living elderly subjects, J. Telemed. Telecare 13 (2) (2007) 79–84. [6] S. Brownsell, H. Aldred, M.S. Hawley, The role of telecare in supporting the needs of elderly people, J. Telemed. Telecare 13 (6) (2007) 293–297.

938

i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 2 ( 2 0 1 3 ) 922–939

[7] P.E. Hasvold, R. Wootton, Use of telephone and SMS reminders to improve attendance at hospital appointments: a systematic review, J. Telemed. Telecare 17 (7) (2011) 358–364. [8] R. Steele, A. Lo, C. Secombe, Y.K. Wong, Elderly persons’ perception and acceptance of using wireless sensor networks to assist healthcare, Int. J. Med. Inform. 78 (12) (2009) 788–801. [9] M. Alwan, S. Dalal, D. Mack, S. Kell, B. Turner, J. Leachtenauer, R. Felder, Impact of monitoring technology in assisted living: outcome pilot, IEEE Trans. Inf. Technol. Biomed. 10 (1) (2006) 192–198. [10] V. Jones, A. Halteren, I. Widya, N. Dokovsky, G. Koprinkov, R. Bults, D. Konstantas, R. Herzog, Mobihealth: mobile health services based on body area networks, M-Health (2006) 219–236. [11] K. Boudreau, Opening the platform vs. opening the complementary good? The effect on product innovation in handheld computing, Social Science Research Network, 2008 (accessed June 2011). Available from: http://dx.doi.org/10.2139/ssrn.1251167 [12] K. Boudreau, Open platform strategies and innovation: granting access versus devolving control, Manage. Sci. 56 (10) (2010) 1849–1872. [13] T. Eisenmann, G. Parker, M. Van Alstyne, Opening platforms: how, when and why? in: Platforms, Market and Innovation, Edgar Elgar, Cheltenham, 2008, pp. 131–162. [14] A. Hagiu, Multi-sided platforms: from microfoundations to design and expansion strategies, Harvard Business School, 2006 (accessed June 2011). Available from: http://www.hbs.edu/research/pdf/07-094.pdf [15] J. Rochet, J. Tirole, Platform competition in two-sided markets J. Eur. Econ. Assoc. 1 (4) (2003) 990–1029. [16] D.M. Kahan, The logic of reciprocity: trust, collective action, and law, Mich. Law Rev. 102 (2003) 71. [17] S. Solaimani, H. Bouwman, M. De Reuver, Smart Home: aligning business models and providers processes; a case survey, in: Proceedings of ACIS, 2010. [18] P. Oliver, G. Marwell, R. Teixeira, A theory of the critical mass. I. Interdependence, group heterogeneity, and the production of collective action, Am. J. Sociol. 91 (3) (1985) 522–556. [19] A.R. Poteete, E. Ostrom, Heterogeneity, group size and collective action: the role of institutions in forest management, Dev. Change 35 (3) (2004) 435–461. [20] R.O. Keohane, Cooperation and international regimes, in: After Hegemony: Cooperation and Discord in the World Political Economy, Princeton University Press, New Jersey, 1984, pp. 491–517. [21] J. Matta, J. Alavalapati, Perceptions of collective action and its success in community based natural resource management: an empirical analysis, Forest Policy Econ. 9 (3) (2006) 274–284. [22] K. Nyikahadzoi, Challenges to collective action in the management of the Kapenta fishery in lake Kariba, Lakes Reservoirs Res. Manage. 14 (4) (2009) 337–351. [23] N. Steins, V. Edwards, Collective action in common-pool resource management: the contribution of a social constructivist perspective to existing theory, Soc. Nat. Resour. 12 (6) (1999) 539–557. [24] B. Gebremedhin, J. Pender, G. Tesfay, Collective action for grazing land management in crop–livestock mixed systems in the highlands of northern Ethiopia, Agric. Syst. 82 (3) (2004) 273–290. [25] I. Hodge, S. McNally, Wetland restoration, collective action and the role of water management institutions, Ecol. Econ. 35 (1) (2000) 107–118. [26] M. Markus, C. Steinfield, R. Wigand, G. Minton, Industry-wide information systems standardization as

[27]

[28]

[29]

[30] [31]

[32]

[33]

[34]

[35]

[36] [37]

[38] [39]

[40]

[41]

[42]

[43]

[44]

[45] [46]

collective action: the case of the US residential mortgage industry, MIS Quart. 30 (I) (2006) 439–465. S. Klein, S. Schellhammer, Developing IOIS as collective action: a cross-country comparison in the health care sector, in: Proceedings of the 44th Hawaii International Conference on System Sciences, IEEE, 2011. European Commissions, ICT Challenge 5: ICT for health, ageing well, inclusion and governance (accessed July 2012). Available from: http://cordis.europa.eu/fp7/ict/programme/challenge5 en.html M. Olson, The Logic of Collective Action: Public Goods and the Theory of Groups, Harward University Press, Massachusetts, 1971. R.V. Gould, Collective action and network structure, Am. Sociol. Rev. 58 (1993) 182–196. D.D. Heckathorn, Collective action and group heterogeneity: voluntary provision versus selective incentives, Am. Sociol. Rev. 58 (1993) 329–350. L.E. Duncan, Motivation for collective action: group consciousness as mediator of personality, life experiences, and women’s rights activism, Polit. Psychol. 20 (3) (1999) 611–635. G. Marwell, P. Oliver, R. Prahl, Social networks and collective action: a theory of the critical mass. III, Am. J. Sociol. 94 (3) (1988) 502–534. R. Sandell, C. Stern, Group size and the logic of collective action: a network analysis of a Swedish temperance movement 1896–1937, Ration. Soc. 10 (3) (1998) 327–345. P. Oliver, Rewards and punishments as selective incentives for collective action: theoretical investigations, Am. J. Sociol. 85 (6) (1980) 1356–1375. R. Hardin, Collective Action, Resources for the Future, The Johns Hopkins University Press, Baltimore, Maryland, 1982. E. Ostrom, Governing the Commons: The Evolution of Institutions for Collective Action, Cambridge University Press, Cambridge, 1990. D. Tilson, C. Sorensen, K. Lyytinen, Platform Complexity: Lessons from Mobile Wireless, ICMB, Delft, 2012. P. Ballon, Platform types and gatekeeper roles: the case of the mobile communications industry, in: Summer Conference in CBS, Copenhagen Business School, DRUID, 2009. D. Evans, A. Hagiu, R. Schmalensee, Invisible Engines: How Software Platforms Drive Innovation and Transform Industries, The MIT Press, Cambridge, Massachusetts, London, England, 2006. J. Moore, The Death of Competition: Leadership and Strategy in the Age of Business Ecosystems, Harper Business, New York, 1996. M. Peltoniemi, E. Vuori, Business ecosystem as the new approach to complex adaptive business environments, FeBR (2004) 267–281. D. Hilkert, A. Benlian, T. Hess, The openness of smartphone software platforms – a framework and preliminary empirical findings from the developers’ perspective, 2011 (accessed January 2012). Available from: uni-potsdam.de. A. Gawer, M. Cusumano, Platform Leadership: How Intel, Microsoft and Cisco Drive Industry Innovation, Harvard Business Press, Boston, 2002. J. West, How open is open enough, Res. Policy 32 (2003) 1253–1285. H. Na, Competition between business ecosystems in quasi-open platform era: a case study of the Japanese mobile industry, Graduate School of Economics and Faculty of Economics, Kyoto University, 2008 (accessed June 2011). Available from: http://www.econ.kyoto-u.ac.jp/ daikokai/research%20outcome/outcome na2.pdf

i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 2 ( 2 0 1 3 ) 922–939

[47] E. den Hartigh, M. Tol, Business Ecosystem, 2008 (accessed July 2011). Available from: http://www.igi-global.com/chapter/ encyclopedia-networked-virtual-organizations/17600 [48] T. Bergstrom, L. Blume, H. Varian, On the private provision of public goods, J. Public Econ. 29 (1) (1986) 25–49. [49] P. Kollock, Social dilemmas: the anatomy of cooperation, Annu. Rev. Sociol. 24 (1) (1998) 183–214. [50] J.-M. Baland, J.-P. Platteau, Does heterogeneity hinder collective action? Cahiers de la Facult? des Sciences ?conomiques et Sociales. 146. Namur, Belgium: Centre de Recherche en Economie du Developpment, Facultes Universitaires Notre-Dame de la Paix, 1995. [51] W. Streeck, Interest heterogeneity and organizing capacity: two class logics of collective action, in: Political Choice, Campus-Westview, Frankfurt-Bou, 1991, pp. 161–198. [52] J.F. Moore, Business ecosystems and the view from the firm, Antitrust Bull. 51 (2006) 31–75. [53] M. Iansiti, R. Levien, Keystones and dominators: framing operating and technology strategy in a business ecosystem, Harvard Business School, 2004 (accessed January 2011). Available from: http://citeseerx.ist.psu.edu/viewdoc/ summary?doi=10.1.1.88.6137 [54] W. Sandholtz, Institutions and collective action: the new telecommunications in Western Europe, World Politics 45 (2) (1993) 242–270. [55] W.T. Bianco, R.H. Bates, Cooperation by design: leadership, structure, and collective dilemmas, Am. Pol. Sci. Rev. 84 (1) (1990) 133–147. [56] S. Greenstein, Open platform development and the commercial Internet, in: Platforms, Markets and Innovation, Edgar Elgar, Cheltenham, 2010, pp. 219–248. [57] P. Huang, M. Ceccagnoli, C. Forman, D. Wu, Participation in a platform ecosystem: appropriability, competition, and access to the installed base, NET Institute, 2009 (accessed February 2011). Available from: http://dx.doi.org/10.2139/ssrn.1480900 [58] M. Iansiti, R. Levien, Keystones and dominators: framing the operational dynamics of business ecosystems, Harvard Business School, 2002 (accessed January 2011). Available from: http://www.hbs.edu/faculty/units/tom/Pages/default.aspx [59] D. Knoke, Incentives in collective action organizations, Am. Sociol. Rev. 53 (3) (1988) 311–329. [60] N. Frohlich, T. Hunt, J. Oppenheimer, R. Wagner, Individual contributions for collective goods, J. Conflict Resol. 19 (2) (1975) 310–329. [61] R. Willer, A status theory of collective action, Adv. Group Processes 26 (2009) 133–163. [62] R.H. Salisbury, An exchange theory of interest groups, Midwest J. Polit. Sci. 13 (1) (1969) 1–32. [63] R.K. Yin, Case Study Research: Design and Methods, Sage Publications, Thousand Oaks, 2009. [64] A.L.M. Cavaye, Case study research: a multi-faceted research approach for IS, Inform. Syst. J. 6 (3) (1996) 227–242. [65] I. Benbasat, D.K. Goldstein, M. Mead, The case research strategy in studies of information systems, MIS Quart. 11 (1987) 369–386. [66] M.B. Miles, A.M. Huberman, Qualitative Data Analysis: An Expanded Sourcebook, Sage Publications, Inc., California, 1994.

939

[67] M.L. Markus, Case selection in a disconfirmatory case study The Information Systems Research Challenge: Qualitative Research Methods, vol. 1, Harvard Business School, Boston, MA, 1989, pp. 20–26. [68] A.M. Pettigrew, Contextualist research and the study of organizational change processes, in: Research Methods in Information Systems, North Holland, New York, 1985, pp. 53–78. [69] C.A. Barry, Choosing qualitative data analysis software: Atlas/Ti and Nudist compared, Sociol. Res. Online 3 (3.) (1998) http://www.socresonline.org.uk/3//4.html [70] T. Muhr, Atlas/Ti – a prototype for the support of text interpretation, Qual. Sociol. 14 (4) (1991) 349–371. [71] T. Eisenmann, G. Parker, M. Van Alstyne, Opening platforms: how, when and why? Harvard Business School, 2008. [72] M. De Reuver, H. Bouwman, T. De Koning, The mobile context explored, in: Mobile Service Innovation and Business Models, Springer Verlag, Berlin Heidelberg, 2008, pp. 89–114. [73] H. Bouwman, E. Faber, T. Haaker, B. Kijl, M. De Reuver, Conceptualizing the STOF model, in: Mobile Service Innovation and Business Models, Springer Verlag, Berlin Heidelberg, 2008, pp. 31–70. [74] K. Boudreau, The boundaries of the platform: vertical integration and economic incentives in mobile computing. MIT Sloan School of Management, 2006 (accessed June 2011). Available from: http://hdl.handle.net/1721.1/30609 [75] E. Von Hippel, G. Von Krogh, Open source software and the “private-collective” innovation model: issues for organization science, Organ. Sci. 14 (9) (2003) 209–223. [76] P. Tang, T. Venables, ‘Smart’ homes and telecare for independent living, J. Telemed. Telecare 6 (1) (2000) 8–14. [77] T. Obi, D. Ishmatova, N. Iwasaki, Promoting ICT innovations for the ageing population in Japan, Int. J. Med. Inform. 82 (4) (2013) e47–e62, http://dx.doi.org/10.1016/j.ijmedinf.2012.05.004. [78] M. Markus, Toward a “critical mass” theory of interactive media, Commun. Res. 14 (5) (1987) 491–511. [79] E. Rogers, The ‘critical mass’ in the diffusion of interactive technologies in organizations The Information Systems Research Challenge: Survey Research Methods, vol. 3, Harvard Business School, Boston, 1991, pp. 245–263. [80] V. Joseph, R.M. West, D. Shickle, J. Keen, S. Clamp, Key challenges in the development and implementation of telehealth projects, J. Telemed. Telecare 17 (2) (2011) 71–77. [81] T. Postema, J. Peeters, R. Friele, Key factors influencing the implementation success of a home telecare application, Int. J. Med. Inform. 81 (6) (2012) 415–423. [82] H. Bouwman, M. De Reuver, F. Hampe, C. Carlsson, P. Walden, Mobile R&D prototypes: what is hampering market implementation? in: 9th International Conference on Mobile Business and 9th Global Mobility Roundtable, IEEE, Athens, Greece, 2010. [83] M.T. Heikkinen, J. Still, Benefits and challenges of new mobile service development in R&D network Pers. Ubiquit. Comput. 12 (1) (2008) 85–94. [84] D. D’Amour, L. Goulet, R. Pineault, J.-F. Labadie, M. Remondin, Comparative study of interorganizational collaboration in four healthcare regions and its effects: the case of perinatal services. University of Montreal, 2004. Available from: http://www. ferasi.umontreal.ca/fra/07 info/Rapport%20ANG.pdf