ArticlePDF Available

The Emergence of Smart Service Ecosystems -The Role of Socio-Technical Antecedents and Affordances

Authors:

Abstract and Figures

As physical products are increasingly augmented with digital technology, manufacturing firms have become part of the development of so-called smart products and smart services. As such, manufacturing firms are challenged by new market participants and ecosystem partners, particularly from the software development industry, and by the dynamic nature of business relationships. While the academic literature on the distinctive characteristics of ecosystems, particularly digital ecosystems, is rich, the effect of smart service ecosystems' emergence on the foundation of smart products remains uncertain. This study reports on case study research based on 47 semi-structured interviews with four companies that participate in an industrial smart service ecosystem. Taking an affordance-theoretic perspective, we uncover the antecedents of and the process of emergent smart service ecosystems. We find that smart service ecosystems have three socio-technical antecedents: a shared worldview, structural flexibility and integrity, and an architecture of participation. We explain the emergence of smart service ecosystems as the result of specialization in shared affordances and integration of idiosyncratic affordances into collective affordances. We derive seven propositions regarding the emergence of smart services, outline opportunities for further research, and present practical guidelines for manufacturing firms.
This content is subject to copyright. Terms and conditions apply.
RESEARCH ARTICLE
The emergence of smart service ecosystemsThe
role of socio-technical antecedents and
affordances
Matthias M. Herterich
1
| Christian Dremel
2,3
|
Jochen Wulf
2,4
| Jan vom Brocke
5
1
Faculty of Information Technology,
University of Jyväskylä, Jyväskylä, Finland
2
Institute for Information Management
(IWI-HSG), University of St. Gallen, St. Gallen,
Switzerland
3
Department of Computer Science,
Norwegian University of Science and
Technology, Trondheim, Norway
4
Institute of Data Analysis and Process
Design (IDP), School of Engineering, Zurich
University of Applied Sciences (ZHAW),
Zurich, Switzerland
5
Institute of Information Systems, Vaduz,
Liechtenstein
Correspondence
Matthias M. Herterich, Faculty of Information
Technology, University of Jyväskylä,
Mattilanniemi 2, Building Agora, Jyväskylä,
Finland.
Email: matthias.m.herterich@jyu.fi
Abstract
As physical products are increasingly augmented with digital
technology, manufacturing firms have become part of the
development of so-called smart products and smart ser-
vices. As such, manufacturing firms are challenged by new
market participants and ecosystem partners, particularly
from the software development industry, and by the
dynamic nature of business relationships. While the aca-
demic literature on the distinctive characteristics of ecosys-
tems, particularly digital ecosystems, is rich, the effect of
smart service ecosystems' emergence on the foundation of
smart products remains uncertain. This study reports on
case study research based on 47 semi-structured interviews
with four companies that participate in an industrial smart
service ecosystem. Taking an affordance-theoretic perspec-
tive, we uncover the antecedents of and the process of
emergent smart service ecosystems. We find that smart ser-
vice ecosystems have three socio-technical antecedents: a
shared worldview, structural flexibility and integrity, and
architecture of participation. We explain the emergence of
smart service ecosystems as the result of specialisation in
shared affordances and integration of idiosyncratic
affordances into collective affordances. We derive seven
Received: 17 July 2019 Revised: 27 August 2022 Accepted: 14 September 2022
DOI: 10.1111/isj.12412
This is an open access article under the terms of the Creative Commons Attribution-NonCommercial License, which permits use,
distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes.
© 2022 The Authors. Information Systems Journal published by John Wiley & Sons Ltd.
524 Inf Syst J. 2023;33:524566.
wileyonlinelibrary.com/journal/isj
propositions regarding the emergence of smart services,
outline opportunities for further research, and present prac-
tical guidelines for manufacturing firms.
1|INTRODUCTION
Product-focused industrial conglomerates like General Electric and SIEMENS, and multinational automakers like
Peugeot, Mercedes Benz, and Ford have been around for more than a century. Known as Original Equipment Manu-
facturers (OEMs), they have established supplier relationships in which single actors play a focal role in value creation
by dictating standards and requirements both horizontally and vertically (Barratt, 2004; Swink et al., 2007). With the
increasing digitization of physical products (Grover et al., 2018; Yoo et al., 2010) and the emergence of services that
complement them (Lightfoot et al., 2013), traditional industry boundaries have blurred, and the mechanics that deter-
mine how organisations collaborate have changed. Specifically, OEMs' central role in value creation is dissolving. This
development challenges product-centred industries to transform their value chains that centre on customersupplier
relationships into smart service ecosystems as focal value propositions that exist beyond industries' reach (Lusch &
Nambisan, 2015; Nambisan et al., 2016; Sambamurthy et al., 2003). Recent McKinsey reports suggest that, by 2025,
today's 100-plus industries and value chains will have collapsed into a low number of multi-trillion-dollar ecosystems
that account for some $60 trillion in revenue, a third of the global economy (McKinsey & Company, 2017,2020).
Consider connected in-car entertainment and navigation systems as an example: Traditionally, car-makers have
controlled the entire supply chain and used proprietary devices that are expensive and had limited connectivity and
imperfect usability. In contrast, today's car manufacturers open their proprietary systems to allow technology compa-
nies like Apple and Google to enter the manufacturers' automotive head unit. Since car manufacturers rely on tele-
communications organisations for connectivity, these organisations provide navigation, communication, and
entertainment services. While OEMs once relied on vertical integration of car components, they no longer fully gov-
ern the supplier relationships, technical capabilities, and services that technology organisations offer drivers.
The consumer example outlined above illustrates how a smart service ecosystem emerges around the automo-
tive head unit. However, the phenomenon's implications are far greater in the industrial context. In product-intense
industries like the automotive, aviation, and industrial manufacturing industries, physical products that are aug-
mented with digital technology afford novel opportunities for organisations to create value jointly in ecosystems.
The value can be packaged as smart servicesthe application of specialised competencies based on the material
properties of smart productsthat leverage both the physical and the digital properties of smart products
(Beverungen et al., 2019; Grover et al., 2018; Porter & Heppelmann, 2014,2015; Yoo et al., 2010).
To highlight the extent of change that smart products and services bring about in the industrial context, consider
as another example KAESER, a German industrial compressed-air specialist that has been seen as an industry-leading
OEM. The compressors are primarily used on industrial shop floors and at construction sites. KAESER has gradually
moved to offer smart services in collaboration with various technology and maintenance organisations. The offerings
leverage the smart product in at least two ways: by enabling accurate, usage-based billing based on how much com-
pressed air is consumed and by enabling predictive maintenance of the compressors, thus ensuring uptime. As indus-
trial customers become accustomed to this holistic and outcome-based service offering, KAESER might eventually
transform the traditional industry of individual OEMs selling physical products to multiple actors offering smart ser-
vices and collaborating in smart service ecosystems. However, providing such smart services means not only over-
coming technological challenges but also requiring organisations to navigate emerging inter-organisational
ecosystems effectively (Lütjen et al., 2019; McKinsey & Company, 2018; Reeves et al., 2017; Svahn et al., 2017).
These two examples show the importance of understanding the peculiarities of smart service ecosystem emer-
gence. In expanding on the concept of smart products, the scholarly literature has started to recognise the concept
of smart services (Beverungen et al., 2019), but as multiple organisations become involved in realising smart services,
HERTERICH ET AL.525
smart service ecosystems emerge (Barile & Polese, 2010; Beverungen et al., 2019; Huber et al., 2019; Thomas &
Ritala, 2021). Although scholars have started to theorise about what ecosystems are and what distinguishes them
from other forms of governance (Jacobides et al., 2018), most have assumed that a central hubfirm coordinates
the network's formation by governing other network participants and designing value creation (Gebauer et al., 2013;
Kohtamäki et al., 2019). Existing work also lacks empirical evidence on the socio-technical antecedents of smart ser-
vice ecosystems and the nature of smart service ecosystem's emergence through multilateral relationships among
organisations and complementary contributions (Aarikka-Stenroos & Ritala, 2017; Taillard et al., 2016; Thomas &
Erkko, 2015; Thomas & Ritala, 2021). In other words, it is not known how smart service ecosystems emerge to real-
ise (interwoven) smart services. To address this research gap, we formulate the following research question:
How do smart service ecosystems emerge around smart products in industrial manufacturing?
To address this question, we performed a revelatory case study with four organisational ecosystem participants
from a large smart service ecosystem in the German manufacturing industry: an OEM, an analytics organisation (AO),
a smart product operator (SPO), and maintenance, repair, and overhaul organisation (MRO). We conducted 47 semi-
structured interviews to investigate how the case organisations put smart products to use to offer smart services,
leading to the emergence of a smart service ecosystem.
The remainder of this paper is structured as follows. Section 2gives the research background, introducing both
the emergence of ecosystems and affordance theory. Section 3presents the case context and details our research
design, while Section 4presents the affordances related to smart products that the organisations in our case context
perceive. Section 5discusses our derived propositions, and Section 6outlines their theoretical and managerial impli-
cations. Section 7concludes this work by summarising its key findings.
2|THEORETICAL BACKGROUND
This section introduces smart service ecosystems as the phenomenon of interest as it relates totwo streams of research:
the formation of ecosystems and innovation in ecosystems. Then, we introduce affordance theory as the lens through
which we examine and further conceptualise the process related to the emergence of smart service ecosystems.
2.1 |Smart service ecosystems in the industrial manufacturing industry
The industrial manufacturing industry has traditionally been geared towards fabricating from raw materials products that
are intended for industrial use. Modern industrial products that are augmented with digital technology are referred to as
smart products (Klein et al., 2018; Porter & Heppelmann, 2015). Their material properties, which include connectivity,
sensors, actuators, interfaces, computing technologies, and the ability to allow for localization and identification, enable
the delivery of smart services (Beverungen et al., 2019). The value of smart products goes beyond their physical (and dig-
ital) materiality (Yoo, 2013; Yoo et al., 2012) to allow for the delivery of smart services. A smart service builds on the
extended features of a smart product to provide value through such extended data-based functionalities (Beverungen
et al., 2019). This convergence of information systems and physical products may be generative for combinatorial and
recombinant innovation which can be referred to as smart services (Beverungen et al., 2019; Yoo et al., 2012).
Smart services spark a transformation of traditional industries into service-based industries (Barile et al., 2016;
Beverungen et al., 2019; Wünderlich et al., 2013). Providing smart services requires capabilities that transcend the
boundaries of single organisations or even industries, so, for example, data analytics companies collaborate with tra-
ditional industrial organisations (Lightfoot et al., 2013). Against the background of increasing industry dynamics
and service orientation, ecosystems are the new organising logic for offering smart services (Barile et al., 2016).
Smart services ecosystems are constituted around smart products that allow actors to co-create value in the form of
526 HERTERICH ET AL.
smart services that are the ecosystem's focal value proposition (Adner, 2017; Barile et al., 2016; Beverungen
et al., 2019; Jacobides et al., 2018; Lusch & Nambisan, 2015; Nambisan, 2013). A firm is considered an ecosystem
actor if it must either change (e.g., increase its capacity) or substantially adapt its business activities (e.g., establish
new work routines or extend capabilities) to contribute to an ecosystem's value propositions (Adner, 2017).
Ecosystems are self-adjusting and evolving organising structures that comprise an assemblage of loosely coupled
multilateral economic actors. Digital technology enables an ecosystem's focal value propositions to materialise by
uncovering where its participants complement each other (Yoo et al., 2010), and the resulting ecosystem of interacting
organisations is enabled by modularity, as opposed to being hierarchically managed (Jacobides et al., 2018). Ecosystems
are bound together by the inability of their collective investment to be redeployed and are defined by the alignment of
the multilateral collection of partners that is required for a focal value proposition to materialise (Adner, 2017;Lusch&
Nambisan, 2015). The actors contribute to value propositions (Camagni, 1993; Jacobides et al., 2018) through multilat-
eral relationships by leveraging their complementary resources and capabilities (Jacobides et al., 2018;Lusch&
Nambisan, 2015). Value propositions are the core of an ecosystem and determine its (fuzzy) boundaries (Adner, 2017).
In summary, an ecosystem may be perceived as a living organism that emerges around the focal value propositions that
emerge from the contributions of complementary ecosystem actors (Adner, 2017).
Our study draws on two streams of research on ecosystems that view ecosystems from different angles: the for-
mation of ecosystems and innovation in ecosystems (Table 1).
The first stream of research on which our study draws is that on the formation of ecosystems, which focuses on gov-
ernance executed by a hub, an orchestrator (Lingens et al., 2020), or a keystone firm(Iansiti & Levien, 2004) as an instru-
ment with which to regulate, design, or orchestrate ecosystem formation by recruiting, motivating, and retaining
participants (Jacobides et al., 2018; Thomas & Erkko, 2015). For example, such a firm establishes contractual relationships
with suppliers and outsourcing providers, distributes decision rights by guarding end customer access, and coordinates the
provision of service to customers (Gebauer et al., 2013). This firm may play an orchestrating role in shaping or designing
the alignment of the ecosystem, which underscores that the formation of an ecosystem is dependent on the design by an
orchestrator or powerful actor in the ecosystem (Lingens et al., 2020), Although studies tend to agree that ecosystems are
not fully hierarchically managed, these studies predominantly delineate the role of these central actors (Jacobides
et al., 2018) and have started to describe the formation of ecosystems as having a subtle, multi-faceted, and emergent
nature (Aarikka-Stenroos & Ritala, 2017;Barile&Polese,2010; Taillard et al., 2016). However, these extant studies do not
provide empirical evidence for how complementing ecosystem actors link themselves and their activities multilaterally dur-
ing the emergence of a smart service ecosystem (Lyytinen et al., 2016; Taillard et al., 2016; Thomas & Erkko, 2015;
Thomas & Ritala, 2021). The locus of value creation has increasingly shifted from the relationship level (i.e., the relationship
between a single provider and its customers) to the network level, where multiple independent actors contribute incremen-
tal value to an overall smart service offering (Barile & Polese, 2010; Huber et al., 2019). Particularly in traditional industries
like manufacturing, realising complex smart services requires new market entrants that have not yet participated in value
co-creation (Barile et al., 2016; Beverungen et al., 2019; Lingens et al., 2020). Therefore, how organisations initiate multilat-
eral relationships as an antecedent to the emergence of a smart service ecosystem and their socio-technical requirements
are of particular interest to both practice and academia (Aarikka-Stenroos & Ritala, 2017;Thomas&Erkko,2015).
The second stream of research, on which our study draws is that which addresses innovation in ecosystems or
networks (cf. Autio & Thomas, 2014; Lyytinen et al., 2016). This stream of research provides useful insights into
interactions between ecosystem actors (Beverungen et al., 2019) and the role of digital technologies as important
drivers of collaboration and collective action in ecosystems (Autio et al., 2018). Innovation results from a diverse set
of actors who contribute combinations of resources that may be assembled as or allowed by material artefacts
(Lyytinen et al., 2016). Thus, because of non-linear dynamic systems and the combining role of digital technologies,
product innovation can be considered to be emergent, distributed, and socio-technical in nature (Lyytinen
et al., 2016; Van de Ven, 2017). In contrast to intra-organisational networks, inter-organisational innovation
networksecosystems or notwill likely be dynamic, uncertain and equivocal(Lyytinen et al., 2016, p. 51) when
they build on digital technologies. Innovation is no longer bound exclusively to the physical materiality of products
but requires the collective, recombinant engagement and action of economic actors (Barrett et al., 2015; Beverungen
HERTERICH ET AL.527
et al., 2019; Nambisan et al., 2016; Yoo et al., 2012). Lyytinen et al. (2016) call for clarification of the degree to which
digital technologies (particularly smart products) and their generative capacity may influence and shape inter-
organisational innovation networks, but how smart products may create, capture and distribute new value remains
unclear (Lyytinen et al., 2016; Yoo et al., 2010). Consequently, many authors call for research to extend this view to
an ecosystem perspective (while acknowledging the emergence of ecosystems) to explain how ecosystems' partici-
pants co-create value (propositions) or innovation in ecosystems, particularly smart services (Autio et al., 2018; Barile
et al., 2016; Lütjen et al., 2019; Thomas & Erkko, 2015; Vargo & Akaka, 2012).
Combining the insights from this outline of how the concept of smart service ecosystems is discussed in the literature
and how we seek to contribute, Table 2provides a summary of the foundational concepts on which our research draws.
2.2 |An affordance-theoretic perspective on the emergence of smart service
ecosystems
Although digital technology is the foundation for many innovations (Nambisan, 2013; Zittrain, 2006), an assessment
of the potential value it offers (i.e., through smart services) must take into account its use context (Autio et al., 2018;
TABLE 1 Two streams of research that are relevant to our study
Dimension
Stream of research
Formation of ecosystems Innovation in ecosystems
Focus and key
concepts
A focal firm that coordinates service
provisioning to customers is at the
centre of investigations.
Innovation is depicted as a recombinant process
of a diverse set of actors with specific roles.
The combination of knowledge and resources,
including digital technologies, may result in the
embodiment of digital innovation in material
artefacts von Briel et al., (2018).
Relevance to this
study
This stream of research defines the
boundaries of an ecosystem, which
helps to define the scope of an
ecosystem. It also delineates the
relevance of complementarity to
ecosystem formation Jacobides et al.,
(2018).
In the context of smart service ecosystems, smart
services are the raison d'etre of smart service
ecosystems. We build on the extant
understanding of (digital) innovation, as the
resulting smart services are considered to be
materialised innovations.
Operationalization This stream of research allows us to
conceptualise the scope of an
ecosystem by defining its boundaries.
Our data analysis pays particular attention to the
combined role of digital technology in the
emergence of the ecosystem. Inspired by this
angle, we propose the integration and
specialisation of affordances as foundational
aspects of an ecosystem's emergence.
Limitations for
explaining the
phenomenon of
interest
Existing work lacks empirical evidence
that delineates how complementary
ecosystem actors link themselves and
their activities multilaterally during
the emergence of a smart service
ecosystem Lyytinen et al., (2016),
Taillard et al., (2016), Thomas and
Erkko, (2015), Thomas and Ritala,
(2021). We argue that a focal firm is
not necessarily required for a smart
service ecosystem to emerge.
Existing research on innovation in ecosystems
acknowledges the emergent, distributed, and
socio-technical nature of innovation; most
work focuses exclusively on internal
organisational processes in investigating the
role of digital technologies in innovation. The
extant research lacks an understanding of how
the generative capacity of smart products may
influence and shape value creation within
inter-organisational innovation networks.
528 HERTERICH ET AL.
Barile et al., 2016; Vargo & Akaka, 2012). The literature focuses primarily on the technical aspects of the technology
but omits the potential of using it. Because of digital technology's malleability, smart products' functionality and uses
must not be studied only as deterministic (Yoo et al., 2012,2010), so the potential of using digital technology is stud-
ied in its context (Orlikowski, 2000). Leonardi (2011,2013a) finds that people approach technological artefacts itera-
tively, forming goals and the individual or collective capacity to use (or act on) an artefact's materiality. In assuming
their agency, people activate an artefact's material agency (i.e., it's capacity for enabling [inter-]action) to accomplish
their collective intentions, that is, the ability to form and realise collective goals through collective action. Against this
backdrop, socio-material thinking acknowledges that social and material aspects of digital technology are intertwined
and enacted in practice (Leonardi, 2011; Orlikowski, 2000). Specifically, an affordance-theoretic perspective helps us
to study the emergence of smart service ecosystems in context and to focus on both the organisational level and the
ecosystem level.
The origins of affordance theory, which first set out to describe how animals and humans interact with objects
(Leonardi, 2011), lie in the realm of perceptual psychology (Gibson, 1986). Affordances are traditionally defined as
emergent, interactive, and dynamic as well as knowledge and communication-intense activity(Miles, 2008, p. 117).
We follow the dominant perspective of Majchrzak and Markus (2013) and Strong et al. (2014) in defining an
affordance as the potential for behaviors associated with achieving an immediate concrete outcome and arising
from the relation between an artifact and a goal-oriented actor or actors(Strong et al., 2014, p. 12). The imperative
to actualize affordances as context- and actor-dependent use potentials has gained attention in interdisciplinary
research (Leonardi & Barley, 2008; Markus & Silver, 2008; Seidel et al., 2013; Strong et al., 2014; Volkoff &
Strong, 2013; Yoo, 2013). Nambisan et al. (2016) characterise affordance theory as a promising lens through which
to distinguish innovation outcomes from innovation processes in the context of a particular set of innovating actors.
Our motivation to draw on an affordance lens to study the emergence of smart service ecosystems is twofold,
as affordance theory allows us (1) to focus on the goal-oriented use potentials of smart products and go beyond their
material properties and (2) to investigate how smart products' use potentials lead to the value propositions of a smart
service ecosystem at both the organisational level and the ecosystem level.
TABLE 2 Definitions of concepts used in our research
Concept Definition Sources
Smart product Smart products embed connectivity, sensors, actuators, interfaces,
computing technologies, and the ability to provide localization
and identification. Smart products' material properties can be
rooted in the concept of digitised products and their modular,
layered architecture, which has a high generative capacity as a
foundation for innovative services.
Beverungen et al., (2019), Yoo,
(2013), Yoo et al., (2012)
Smart service Smart services are applications of specialised competencies
through deeds, processes, and performances and are based on
smart product's material properties.
Beverungen et al., (2019)
Ecosystem Ecosystems are interacting organisations, enabled by modularity,
that are not hierarchically managed and are bound together by
their collective investment's inability to be redeployed. They are
defined by the alignment structure of the multilateral set of
partners that must interact for a (focal) value proposition to
materialise.
Adner, (2017), Jacobides et al.,
(2018), Vargo and Lusch,
(2017)
Smart service
ecosystem
A smart service ecosystem is an ecosystem that is based on smart
products' material properties and constituted around smart
services as (focal)value propositions.
Adner, (2017), Barile et al.,
(2016), Beverungen et al.,
(2019), Jacobides et al.,
(2018), Lusch and
Nambisan, (2015)
HERTERICH ET AL.529
The first reason for our drawing on the affordance concept, which is relational in nature, is that it allows us to
focus on smart products' goal-oriented use potentials in a specific use context so we can look beyond their pure
technology. Affordance theory proposes that value arises from the relationship between the material properties of
an artefact (i.e., a smart product) and its use context (Majchrzak & Markus, 2013; Markus & Silver, 2008; Volkoff &
Strong, 2013). According to extant work that uses an affordance lens (Leonardi & Barley, 2008; Yoo, 2013), smart
products have a physical and digital materiality. Leveraging an affordance lens allows both property categories to be
subsumed under what the theory calls material properties. Whereas physical material properties refer to largely
unchangeable, visible, and tangible properties, such as sensors and actuators, digital material properties refer to
what the software incorporated into an artifact can do by manipulating digital representations(Yoo et al., 2012,
p. 1398). Affordances link an object's or technology's materiality (i.e., material properties) to its use context
(Fayard & Weeks, 2014; Markus & Silver, 2008; Zammuto et al., 2007). In a business context, organisations can draw
on smart products' material properties to achieve their organisational goals as well as goals that go beyond their
boundaries. We use affordance theory's focus on use potentials to investigate their interplay and explain the emer-
gence of service ecosystems.
The second reason for our drawing on the affordance concept is that it is a multi-level concept that allows us to
investigate the emergence of smart service ecosystems at both the organisational level and the ecosystem level:
Existing research already leverages an affordance lens to conduct investigations on multiple levels and to conceptual-
ise level-specific classes of affordances. For instance, Leonardi (2013b) focuses on multiple levels in investigating
how technology changes how individuals collaborate and how these individual-level use potentials change the prac-
tices of groups of individuals (i.e., departments in an organisation). The multi-level perspective allows investigations
to be conducted at both the organisational level and the ecosystem level and to study the effects that occur between
these levels (Savoli & Barki, 2016). For these reasons, this perspective allows us to explain the interplay among
actor-specific affordances, affordances that are shared between actors, and affordances that are perceived collec-
tively at a group level, which may go beyond the sum of the individual parts. By applying this multilevel theory, we
can investigate the emergence of smart service ecosystems by focusing on the relationship between smart products
and their use contexts across the boundaries of individual organisations and at both the organisational level and the
ecosystem level.
In sum, affordance theory allows us to distinguish between the material properties of smart products and the
context of their use at the organisational and ecosystem levels, to investigate the interplay of multiple actors in an
ecosystem, and, thus explain the emergence of smart service ecosystems.
3|RESEARCH DESIGN AND METHOD
As research on the emergence of smart service ecosystems is scarce (Barile et al., 2016; Jacobides et al., 2018), our
strategy is to study one case in depth. Since we investigate a novel phenomenon with heretofore undefined bound-
aries (Silverman, 2010; Yin, 2008), we sought an archetypical smart service ecosystem that could provide a unique,
rich set of empirical data on our phenomenon of interest that would help us develop an explanatory theory. We
investigate organisations and their interactions in an ecosystem with oscillating foci(Beir˜ao et al., 2017; Chandler &
Vargo, 2011; Vargo & Lusch, 2017), taking into account the multidimensional nature of ecosystems, that allows us to
reveal how affordances emerge at both the organisational level and the ecosystem level.
We follow the principles of interpretive case study research (Klein & Myers, 1999; Walsham, 1995,2006)to
explain complex real-world phenomena in their social or organisational embedded contexts (Eisenhardt &
Graebner, 2007; Orlikowski & Lacono, 2001).
Our case study is revelatory for two reasons. First, cases that study the emergence of smart service ecosystems
extensively are scarce and have been called for (Barile et al., 2016; Jacobides et al., 2018; Taillard et al., 2016). The
phenomenon of interest has been inaccessible to investigators because of their limited access to in-depth empirical
530 HERTERICH ET AL.
data related to a coherent smart service ecosystem. Second, our case study is revelatory since the manufacturing
industry, as our case context, is subject to change and significant transformation as a result of digitization and
servitization. Digitization refers to industrial products' having become increasingly equipped with digital technology
(smart products) (Beverungen et al., 2019; Grover et al., 2018; Porter & Heppelmann, 2014,2015; Yoo et al., 2010).
Servitization refers to the trend of moving from a traditional, dyadic goods-dominant logic that relies on the
exchange of industrial products for cash to a service-dominant logic where multiple organisations intensify value co-
creation in innovative service systems to meet their customers' needs (Klein et al., 2018; Lightfoot et al., 2013;
Neely, 2013).
3.1 |Case context
This section provides detailed information on our case context. First, we describe the chief characteristics of the key
players and the salient characteristics of the smart service ecosystem itself (Jacobides et al., 2018). Then, we intro-
duce the four organisations that we selected for interviews and outline the smart products and smart services they
produce.
We investigate the emergence of smart service ecosystems in the context of an archetypical smart service eco-
system in the industrial manufacturing industry. Accordingly, our unit of analysis is a smart service ecosystem con-
sisting of four typical industrial organisationswhich we name OEMCorp, MROCorp, SPOCorp, and AOCorpthat
leverage two types of smart products to offer smart services to their customers (Halinen & Törnroos, 2005; Miles &
Huberman, 1994). The first smart product in our case is elevators that are augmented with digital technology in the
form of sensors and actuators and the second is forklift trucks that are connected to a central big data analytics plat-
form via mobile connectivity. In line with prior research (Becker et al., 2013; Gebauer et al., 2013; Meyer
et al., 2011), the case organisations have typical industrial actors that include OEMs, SPOs, and MRO organisations.
As our ecosystem relates to smart products and since previous focus groups with practitioners highlight the value of
a dedicated AO as an organisational actor, we add the role of an AO.
OEMs are mainly responsible for building industrial products. Their organisational structure is product-focused,
and their traditional focus is the engineering and manufacture of industrial products. OEMs' organisational goals
prioritise a product's beginning of life by providing high-quality products at competitive prices (Gebauer
et al., 2013).
SPOs are the main beneficiaries of the smart service ecosystem. However, the purchase of industrial products
does not satisfy their need to gather resources to co-create value with their customers. SPOs' key goals are to
maximise the smart products' productivity, create transparency between industrial products and processes, and
pass this information on to customer organisations.
MRO organisations play important roles in traditional industrial contexts because of industrial products' long life
cycles. MRO organisations typically have a global footprint, although they are structured as regional entities. Their
key objectives are to ensure fault-free and safe operations and to reduce downtime in industrial products. Despite
their traditional mindset, MRO organisations recognise the benefits of leveraging digital technology to increase
operational efficiency in delivering their services to SPOs. MRO organisations build strong ties with SPOs because
they accompany industrial products throughout their lifecycles. MRO organisations' primary goals are to provide
efficient activities and to differentiate themselves from their competitors by providing innovative smart services
that address the industrial products' operations phase.
AOs are actors that have the resources to deal with operational data in the context of industrial manufacturing.
With the rise of smart products and the operational product data that results, the established roles in the indus-
trial manufacturing industry face increasing difficulty processing that voluminous data efficiently. As players that
play an emerging role in the industrial smart service ecosystem, AOs address this issue by providing dedicated
HERTERICH ET AL.531
resources for the collection, storage, and analysis of vast amounts of operational product data. AOs' competencies
include real-time data-streaming and big data analytics. Their organisational culture tends to be open-minded
towards innovation and digital technology.
The ecosystem we investigate in our study has all four types of industrial actors: OEM, SPO, MRO, and AO. It is
an ecosystem in the industrial manufacturing sector in Germany that involves around 50 networked firms and orga-
nisations. We applied purposeful sampling (Coyne, 1997; Patton, 1990) to select a subset of four organisations that
represent the four roles; the result is a typical set of actors representing a smart service ecosystem in the industrial
space. This smart service ecosystem is uniquely suited to our research objectives because each of its industrial orga-
nisations plays characteristic roles in a smart service ecosystem in the industrial manufacturing industry. Following
Adner (2017), we determined the boundaries of our ecosystem based on the changes in the firms' activities that
allowed them to take part in providing smart services. Figure 1provides an overview of our archetypical smart ser-
vice ecosystem, which consists of the four organisations in our case ecosystem: OEMCorp, MROCorp, SPOCorp,
and AOCorp.
We briefly characterise each organisation:
OEMCorp is a multinational conglomerate that focuses on industrial engineering of standardised intralogistics
products like elevators and escalators. OEMCorp also manufactures more extensive customised industrial solu-
tions, such as power plants and production facilities. The company employs around 160 000 people and gener-
ates annual revenue of more than 40 billion. In our case ecosystem, OEMCorp does not perform the
responsibilities of an orchestratoror exercise more power than other actors in the network does.
SPOCorp is a global steel-processing company in the business of global material processing and distribution, with
around 14 billion in annual revenue, 20 000 employees, and 500 sites worldwide. Their services include slitting,
shearing, laser cutting, sawing, drilling, milling, and coating of the materials they sell. To provide these services,
they operate a range of intralogistics equipment and moving assets, such as forklift trucks and stationary equip-
ment.
1
Other equipment includes conveyor belts, high-precision laser-cutting machines, and packaging machines.
MROCorp is a leading multinational intralogistics and materials-handling organisation that focuses on moving
assets in intralogistics. In 2018, MROCorp generated revenues exceeding $4 billion with a workforce of around
20 000 employees. The MRO portion of the business, which comprises more than half of the staff, is organised
regionally and is responsible for more than 45% of total revenue.
AOCorp, a concrete-manufacturing AO that focuses on Industrial Internet of Things (IIoT) use cases, offers a data
platform that supports the cloud-based prediction of anomalies using digital twins. This technology supports a
range of industrial assets, including moving equipment like forklift trucks and fixed assets like elevators. As
AOCorp has a close and stable relationship with one of the biggest industrial OEMs in Germany, it has profound
domain knowledge and expertise in the operational data of industrial products.
Our case organisations provide smart services that, collectively, materialise into (focal) value propositions.
Table 3provides examples of smart services in our case study's smart service ecosystem.
In line with Jacobides et al. (2018), our case ecosystem differs from other business constellations in that inter-
acting organizations [are] enabled by modularity, not hierarchically managed, [and] bound together by the
nondeployability of their collective investment elsewhere(Jacobides et al., 2018, p. 2255). When selecting the case
ecosystem, we ensured that its participants have the three crucial attributes of being enabled by modularity, not
hierarchically managed, and bound together by their inability to deploy their collective investment elsewhere. Our
case organisations are enabled by modularity in that they are distinct parts of the ecosystem, with each organisation
having distinct organisational goals and value propositions for its customers. Since our case organisations are both
separate organisational entities and not deeply bound by contractual agreements, we can confirm that they are not
hierarchically managed. When selecting the four case organisations, we ensured that they are independent yet
532 HERTERICH ET AL.
FIGURE 1 Overview of the four organisations in our smart service ecosystem
HERTERICH ET AL.533
TABLE 3 Examples of smart services identified in our case ecosystem
Sample smart
service Description Sample quotations
Triggering
maintenance
activities for
elevators based
on analysis of
historic data
Historic data of an elevator's door
movements can be tracked at each level.
Individual elevators' patterns can be
compared with the historic data of a
single piece of equipment and the entire
installation. Anomalies can trigger
maintenance activities.
We operate four seasons a year, and I hear from
my team that [the elevators] have a higher
failure rate in the summer when it's hot []
than in winter. Traditional service is not able to
address this, but with continuous data streams,
we are able to see these connections. []
Humidity in different continents is another
thing for the bearings and hoists.(Director of
Operations, SPOCorp)
Diagnosing and
solving elevators'
technical
problems
remotely
Using digital technology like sensors and
actuators that are integrated into the
elevator on each floor and the drive
system, as well as a capable
infrastructure, makes diagnosing the root
cause of technical problems possible.
Spare parts can be pre-ordered
accordingly, and on-site service visits can
be scheduled. Software problems can be
solved remotely by changing parameters
or updating the elevator's software.
We have the ability to reset the elevator
remotely or move it around safely. [] Such
remote activities also eliminate on-site trips.
(Vice President, Field Service, MROCorp)
We want to be able to update the software on
the products remotely.(Global Head of IT
Operations, OEMCorp)
Benchmarking
production
facilities'
performance
Operational data of machinery and
production facilities in the process
industry like paper production, water
treatment, steel processing, and chemical
production can be used to benchmark
throughput and production facilities.
We focus on instrumentation and control
technology. What might interest an operator
of a paper production facility? [] What does
he want? For sure, he wants optimal efficiency
and utilization. [] In the end, it is resource
efficiency and profitability.(Service Line
Manager, Plant Cloud Services, AOCorp)
Renting
intralogistics
equipment based
on actual usage
Instead of selling or leasing forklift trucks,
they can be offered as a service based on
their actual usage, which includes wear
and tear, as well as maintenance and
repair activities that involve the MRO
organisation.
For us, flexible pricing means leveraging
seasonal usefor instance, seasonal
complementing dynamic scaling of the farmers'
truck fleet in northern and southern Italyas
well as solving the financing problems of
forklift truck operators. Usage-based pricing is
a huge topic with many smaller sub-use cases.
(Managing Director, Innovation Management,
OEMCorp)
Managing and
optimising
material flows
and intralogistics
processes for
customers
Instead of selling or leasing forklift trucks,
this service offering assumes overall
responsibility for material flows at the
customer site and optimises intralogistics
processes for customers.
If we establish a partnership with both our
customers and [AOCorp], we can go beyond
just offering forklift truck control systems and
actually manage our customers' processes and
operations for them.(Director, After-Sales and
Customer Service, MROCorp)
Auto-levelling for
elevators
In the past, periodic on-site maintenance
activities were necessary on each floor to
adjust where the elevators stop. Now,
using both sensing and responding
capabilities, this adjustment can be done
as an automatic service.
Precision monitoring and leveling is a quick win.
It could be accomplished through continuous
measuring. The added value for the operator
and the user is the increased safety, the
prevention of accidents, and a reduced number
of liability losses.(Director, Sales Region A,
OEMCorp)
534 HERTERICH ET AL.
interdependent, such that a clearly defined value proposition for the ecosystem's customers is present. Further, in
line with Adner (2017), we selected case organisations whose business activities changed or were substantially modi-
fied to allow the defined value proposition to be generated. Hence, we can confirm that coordination is governed
without full hierarchical fiat. Finally, we consider our case ecosystem to be a smart service ecosystem, as its partici-
pating actors are tied together around a set of smart products that require domain expertise in the moving assets
and intralogistics industry and could not be deployed elsewhere.
The smart products themselves are the cornerstone of the smart service ecosystem. Following the affordance
theory, the material properties that characterise both forklift trucks and elevators as smart products in our focal eco-
system provide the foundation for smart services. To structure the individual modules' material properties, we draw
on the layered modular architecture Yoo et al. (2012) introduced, which consists of four layers: the device layer, the
network layer, the service layer, and the contents layer.
In the context of our case, industrial products like elevators and forklift trucks must be augmented with sensors
and actuators to protect against on-site fraud and manipulation on the device layer. Bi-directional, reliable, and secure
product connectivity with sufficient bandwidth is established in the network layer to transmit various kinds of sensor
data securely to a data platform in the service layer. (We identify standardised interfaces and protocols like MQTT
and OPCUA.) After transmission to a data platform, operational product data can be stored and analysed in the ser-
vice layer. To analyse the data in the service layer in such a way as to obtain insights and make sound decisions
based on the operational product data, data-analytics technology that can handle enormous amounts of data must
be in place. It can be distinguished between two modes of data analysis: prompt analysis of incoming data using com-
plex event processing engines and messaging buses so the firm can react to unforeseen events, and pattern detec-
tion, advanced statistical analysis, and machine learning to analyse the substantial amounts of historic data on the
data platform and on the smart products themselves. (Apache Spark and Hadoop are good examples.) Finally, the
content layer involves data received from industrial products. The data must be consistent and comparable through-
out the installed base and provide a comprehensive reflection of the industrial production in the field. (Practitioners
refer to this reflection as a thing shadowor a digital twin.) Table 4provides an overview of the material properties
of the digital technology attached to industrial intralogistics equipment as the nexus of our case ecosystem.
We chose the industrial manufacturing industry for our study of the emergence of smart service ecosystems for
three reasons (Barile et al., 2016; Tuunanen, 2012). First, innovation in industrial manufacturing is no longer bound
exclusively to the physical materiality of products or a single organisational actor but requires the involvement of
multiple ecosystem participants (Barrett et al., 2015; Nambisan et al., 2016). Services in the industrial manufacturing
industry are geared towards the long lifecycles of industrial products (Fain et al., 2017; Lightfoot et al., 2013), adding
stability to our case context in terms of the ecosystem's primary usage. Second, industrial manufacturers proliferate
digital technology for smart products rapidly, increasing the creation of collective smart services by participants in
ecosystems (Barile et al., 2016; Nambisan et al., 2016). Third, because of the long life cycles of and high investment
in industrial products, the industrial manufacturing industry is transitioning to smart service ecosystems that leverage
digital technology for smart products (Barile et al., 2016; Fain et al., 2017; Lightfoot et al., 2013). Hence, our case
ecosystem allows us to determine whether the prevalent perspective of a hub enterprise in the industrial
manufacturing industry should be deemed out-of-date industrial or manufacturing logic(Barile et al., 2016, p. 666).
3.2 |Data collection
The focus of our data collection was on information about how smart service ecosystems emerge. We conducted
47 interviews at a large German smart service ecosystem in the industrial space, involving all four types of industrial
actors. Using snowball sampling, we collected data from four industrial organisational perspectivesthose of the
OEM, the SPO, the MRO, and the AO (Myers & Newman, 2007). Interviews were conducted between March 2014
and March 2018 as the primary method of data collection. Two researchers conducted semi-structured interviews
HERTERICH ET AL.535
with executives from the IT department, with chief technology officers, and with managers responsible for a service
business, service innovation, and product digitization until we reached data saturation (Corbin & Strauss, 2008).
Our open-ended interview questions focused on five areas: the organisational context, including the job position
of the interviewee and the competitive situation of organisation; descriptions of smart products and use cases; key
features and participants of the ecosystem and collaboration with other organisations; process and activities related
to how the ecosystem was formed; and complementarity and value propositions for the end customer. To obtain a
TABLE 4 Material properties of smart products as shared digital technology
Material property Sample quotations
Physical products
augmented with tamper-
proof sensors and
actuators
You may want to monitor that particular piece of equipment, which is another
functionality of what [internal project name] can provide, and you need to utilize
either serial connections or sensors, which are independent of the control system.
This imposes limits on what you can do remotely.(Group Chief Technology Officer
and Head of Group R&D, OEMCorp)
Hardware security is super-important. [] We do not have any interest in a third
party accessing that raw operational product data in any manner.(Head of Product
Marketing, OEMCorp)
Bi-directional, reliable, and
secure product
connectivity with
sufficient bandwidth
We use various technologies to ensure reliable connectivity []. Wireless network
and Bluetooth are two options [], but the most widespread technology is 3G.
More than 12 000 connected products result in around 350GB of compressed
mobile traffic per year.(Head of Competence Center IoT Platform Architecture,
OEMCorp)
Standardised interfaces,
protocols, and data
structures
Technical standards like OPCUA that allow us to communicate directly with the cloud
infrastructure or MQTT [] are relevant for interoperability and the development of
data-driven services.(Service Product Manager, Out of the Box Analytics, AOCorp)
Data storage and retrieval An IoT software platform must be able to store operational product data, irrespective
of the age and configuration of the capturing device. Since we are obliged to meet
statutory storage policies, we might need to save collected individual measurement
values for 15 years. This requirement can only be addressed by a central platform.
(Vice President, Business Development and Operational Excellence, POCorp)
Incoming data-stream
processing and timely
alerts and notification
When I get an error, I can immediately tell the customer to stop the operations to
prevent damage.(Director, After-Sales Region A, MROCorp)
Descriptive data analytics
based on historic data
The machine is not just a simple transmitter of data; there is a lot of inbuilt
intelligence and computing power. For example, the product doesn't continuously
send data but systematically connects to a central platform after aggregating and
validating data.(Head of Competence Center IoT Platform Architecture, OEMCorp)
Predictive data analytics
based on historic and
current data
The goal of predictive maintenance is to predict when a component will break down
and then replace it during a regular service interval, such as at night, before a failure
happens, so we can increase the overall availability of the product.(Vice President,
Field Service, MROCorp)
Digital product twin and
integration of product
master data
Today, we can build digital twins that simulate and synchronize the characteristics
and behavior of the real machine.(Head of Managed Service Analytics, AOCorp)
Consistency of operational
product data throughout
the installed base
We need to establish a single point of truth and create an unambiguous data record
over the product's whole lifecycle.(Managing Director, Innovation Management,
OEMCorp)
[Irrespective of the final use cases], we aim at digitizing products in a highly
standardized and scalable way. This includes hardware, sensors, software
components, data management, data analysis, and the generation of insights.
(Service Line Manager, Asset Analytics Services, AOCorp)
536 HERTERICH ET AL.
holistic, unbiased view and balance the breadth and depth of the four perspectives to meet the requirements of
validity and reliability, we followed Eisenhardt and Graebner (2007) in recruiting interviewees from multiple hierar-
chical levels, organisational roles, and locations (Bryman & Bell, 2015; Easterby-Smith et al., 2012). We designed the
interviews to reveal how smart products are harnessed in both an organisational context and an ecosystem context,
so we structured them based on the theoretical concepts and frameworks discussed in Section 2. The average dura-
tion of the interviews was 72 min. All interviews were recorded and transcribed, resulting in 864 pages of text.
We sought to develop a clear chain of evidence by considering multiple data sources, so in addition to inter-
views, we performed supplemental activities like full-day innovation workshops, focus groups (Tremblay et al., 2010),
and conference calls. We also reviewed internal documentation, presentations, and other archival data from the
organisations we interviewed. Then we compiled actor-specific summaries for all activities other than the semi-
structured interviews. These activities gave us additional or broadened insights and triangulated the findings
obtained from the interviews (Yin, 2008). Appendix Apresents an overview of the data gathered and their sources.
3.3 |Data analysis
To mitigate interpretive research's risk of low generalizability, we applied Gioia et al.'s (2013) rigorous analysis
approach to concept development. This approach pays particular attention to linking the empirical case and the idio-
graphic findings to the existing body of knowledge as a way to balance the generation of novel theory and identifica-
tion of theoretical precedents.
Two researchers discussed and analysed the collected data line by line and in an iterative manner using an inter-
woven three-stage process of open,axial, and selective coding. In addition to employing Gioia et al.'s (2013) recom-
mendations, the data analysis is rooted in Corbin and Strauss's (1990) and Strauss and Corbin's (1997)
recommendations and in grounded theory (Glaser & Strauss, 2009). We sought to identify first-order concepts (infor-
mant-centric concepts and themes), second-order concepts (research-centric concepts and themes), and their distilla-
tion into aggregate dimensions (Gioia et al., 2013). Appendix Bprovides an overview of the data structure we
derived from open, axial, and selective coding. Following Strauss and Corbin (1990, p. 94), we never impose
[d] anything on the datawhile also applying their principle of theoretical sensitivity, which encourages the use of
analytic tools like the affordance theory to allow insights to emerge. We were aware that using analytical tools like
existing theory could carry the risk of forcing the data in preconceived ways, possibly leading to confirmation bias.
In the open coding stage, we identified recurring codes in the data to address the concepts that guided us when
we compiled the interview guideline (Van Maanen, 1979) while remaining open to identifying other salient concepts
from the data (Gioia et al., 2013). We compared the codes as they emerged to identify common first-order concepts
based on our set of empirical data. Appendix Blists the final 39 first-order concepts.
In the axial coding stage, we condensed first-order concepts into themes by drawing on the theory of
affordances to distinguish between smart products' material properties and the use context at the organisational and
ecosystem levels (Chandler & Vargo, 2011). By harnessing the full strength of human sense-making, we could explore
the qualitative relational nature of affordances in an archetypical smart service ecosystem (Leonardi, 2011; Pozzi
et al., 2014). Appendix Blists the 17 final second-order concepts.
Having first derived a workable set of first-order concepts and second-order themes (Gioia et al., 2013), we used
an integrated perspective in the selective coding stage to distill the second-order themes further into aggregate
dimensions.When we stabilised our coding structure at the beginning of the selective coding stage, we compiled a
coding scheme that was evaluated in two focus group workshops with practitioners who had not participated in the
interviews and an interdisciplinary panel of senior researchers, respectively (Tremblay et al., 2010). The evaluation
workshops led to simplifying the initial coding scheme to enhance clarity and allow for exploratory findings.
While interpretative research typically does not take into account assessments of intercoder agreements, we
wanted to see whether we had convergence on key aspects of our coding scheme. Complementing the methodology
HERTERICH ET AL.537
proposed by Gioia et al. (2013), similar to other studies (Anand et al., 2007; Clark et al., 2010; Gioia et al., 2010;Nag
et al., 2007), we sought to increase trustworthiness and credibility and bolster our confidence in our interpretations' plausi-
bility by asking two additional researchers who were not involved in this project to code an initial sample of 11 interviews
in two iterations. After each iteration, we assessed inter-coder reliability using Cohen's kappa, a coefficient that measures
whether the inter-rater proportion of agreement is greater than would be expected by chance (Rust & Cooil, 1994). After
the first round, Cohen's kappa was 0.51. We discussed the major inconsistencies in coding with independent coders and
revised the coding scheme for enhanced clarity and more consistent coding results. After the second round of coding using
these changes, Cohen's kappa increased to 0.77, which is substantially higher than the threshold level of 0.60, indicating
significant results (Moore & Benbasat, 1991). This step increased the rigour of our study. Coding the entire data set using
the coding scheme (Appendix B) resulted in 2611 codes. We used NVivo 11, a computer-assisted qualitative data analysis
tool, to analyse the interview transcripts and internal documents.
Finally, to scale up and integrate our emerging theory theoretically into other theories in the research field
(Urquhart et al., 2010), we paid close attention to nascent themes in the literature that are related to the emergence
of ecosystems, smart services, and digital innovation (Gioia et al., 2013) and that are either not referred to or dis-
cussed only conceptually or that allow theoretical knowledge to be incorporated into our first-order concepts. An
iterative process of relating our aggregate dimensions to the body of knowledge allowed us to distill propositions
that describe the relationships among our second-order concepts and the concepts of the extant body of knowledge
on ecosystems (Lusch & Nambisan, 2015), as Gioia et al. (2013) suggest. In doing so, we focused on exploring the
mechanisms at play in ecosystems' emergence. We refer to mechanisms as an intermediary level of analysis in-
between pure description and storytelling on the one hand, and universal social laws on the other(Hedström &
Swedberg, 1996, p. 281). The concept of mechanisms allows the processes that underlie cause-effect relationships
to be described (Gross, 2009). Mechanism-based theorising has proven valuable in qualitative research that focuses
on the emergent nature of digital phenomena (Henfridsson & Yoo, 2014; Huang et al., 2017).
4|RESULTS
This section describes the affordances related to smart products that the organisations in our case context perceive.
While we focus on the organisational and ecosystem levels, we identify three classes of affordances: shared
affordances, idiosyncratic affordances, and collective affordances. Appendix Cprovides a detailed overview of the
affordances we identified and their relationships. In what follows, we describe the affordances we identified and
their relationships.
4.1 |A shared worldview manifests in shared affordances
Shared affordances are those that all participants in a smart service ecosystem perceive. Our study suggests that
shared affordances arise through a shared worldview, as the subject matter experts we interviewed reported a shared
set of business assumptions and at least partly aligned organisational goals, which they considered an important pre-
requisite for developing smart products and services.
In the worst case, you talk to people from other organizations, and you notice that they have no clue
about how smart industrial equipment transforms our business. A shared perspective and a shared
language are mandatory. (Head of Managed Service/Analytics, AOCorp)
The shared affordances the participants identified are perceived equally by all case companies since they are tied
to the common denominator of organisations' goals for creating transparency and visibility while also controlling the
538 HERTERICH ET AL.
equipment state. Table 5provides an overview of the two shared affordances identified: Sensing equipment state
(SA #01) and Actuating equipment state(SA #02).
Subject matter experts in all four companies of our case ecosystem acknowledged the affordance of Sensing
equipment state(SA #01) so visibility and the state of operations is an underlying concern for all participants. One
consideration is that technological distance negatively affects inter-organisational innovation, and shared
affordances signal the potential for joint smart service innovations in the case ecosystem. Subject matter experts
from the four companies exhibited similar mental frameworks when they described the shared affordances of smart
products. For example, OEMCorp's Head of Logistics and Process Performance spoke of the transparency to see
machine conditions,the MROCorp's Director of Service Operations mentioned the ability to monitor assets,and
POCorp's Director of Operations talked about the potential of obtaining visibility.
Apart from a shared mental framework, shared affordances manifest a shared situational awareness of
technologies' affordances, which is the foundation for capitalising on inter-organisational synergies. In other
words, shared affordances motivate the participants in an ecosystem to mobilise and integrate their diverse
capabilities. A service manager at AOCorp explained how a shared affordance leverages the ecosystem's
synergies:
If we can observe systemic phenomena in the installed base, they can draw CAPEX/OPEX decisions
based on our system. This allows the MRO organization to work on strategic topics, the OEM to work
on engineering better products, and operators to optimize the setup of their installed base and utiliza-
tion. (Service Manager, Out of the Box Analytics, AOCorp)
We conclude that the shared worldview of the four organisations participating in the smart service ecosystem
manifests in the smart products' shared affordances that all participants perceived.
4.2 |Structural flexibility and integrity manifest in idiosyncratic affordances
Idiosyncratic affordances are actor-specific affordances of smart products that depend on an organisation's context,
unique goals, and capabilities. Data from our case ecosystem indicates that idiosyncratic affordances emerge through
TABLE 5 Overview of shared affordances
ID
Shared affordance
(code frequency) Sample quotations
SA #01 Sensing equipment
state (42)
It would be valuable for us to have [] transparency to see the condition and
capacity use of all of our machines.(Head of Logistics and Process
Performance, OEMCorp)
As we monitor assets, we can tie down a mean time between failures in
specific equipment types and specific equipment configurations that will
help us understand what requires more maintenance and what doesn't.
(Director, Service Operations, MROCorp)
Operational machine data is very interesting to us, as we can deduct
performance measures [], and then we obtain visibility on how efficiently
this equipment is actually used.(Director of Operations, POCorp)
SA #02 Actuating equipment
state (13)
Actuators could be added to the building infrastructure in large warehouses. A
specific example would be controlling rapid action doors based on the speed
of approaching forklift trucks. Connected actuators would ensure that the
doors open just in time to avoid emergency braking and loss of load.
(Director, New Business and Product Digitization, OEMCorp)
HERTERICH ET AL.539
a process of actors' drilling down or into or specialisingin shared affordances to pursue the goals that are specific
to their respective organisations. This sequential process of specialisation by relying on shared affordances to build
idiosyncratic affordances is reflected in our data:
The real challenge is that you have to see these general benefits through the eyes of your organiza-
tion. The benefits that we see as a manufacturer are different from what the insurance industry
comes up with. Compared to us, they will feed their insurance models and algorithms with this [opera-
tional visibility]. For them, this leads to fantastic risk evaluations. (Service Manager, Out of the Box
Analytics, AOCorp)
We refer to this process of elaborating idiosyncratic affordances that uniquely fit the individual organisation's
resources and objectives as specialisation. In other words, organisations leverage shared affordances to work
towards achieving their individual organisational goals. Table 6provides an overview of the idiosyncratic affordances
we identified.
For example, AOCorp leverages data to generate performance insights (AO #01).
It starts with simple benchmarking topicsthat is, comparisons between performances over time. This
extends to more sophisticated issues. (Service Line Manager, Asset Analytics Services, AOCorp)
However, OEMCorp identifies misuse of its products by leveraging insights that are based on domain knowledge
that is idiosyncratic to its business (OEM #03).
What are the environmental conditions of our products? Temperature? Dust and degree of air pollu-
tion? Because we want to earn more money from our customers in a fair way, we have to measure
the environmental factors of product operations. (Managing Director, Innovation Management,
OEMCorp)
Based on the misuse it identifies, MROCorp manages product operations so it can guarantee product uptime
(MRO #04):
We want to sell service contracts that say that we take care of the customer's machine so the cus-
tomer can focus on his core business. (Head of MRO Service, MROCorp)
Finally, SPOCorp leverages industrial equipment with guaranteed uptime by providing enhanced operational
transparency (SPO #02):
We can use the new data to provide our customers with more transparency and information about
the state of their orders. (Head of Logistics and Process Management, SPOCorp)
The formation of a broad set of idiosyncratic affordances by the participants in the smart service ecosystem
requires considerable structural flexibility because these affordances represent numerous opportunities for joint inno-
vations and flexible adaptation to environmental conditions. For example, SPOCorp's affordance of providing trans-
parency to customers complements MROCorp's affordance of managing industrial equipment operations to
guarantee product uptime (MRO #04). MROCorp's offering, which affords SPOCorp more reliable planning of opera-
tional processes, also complements to OEMCorp's affordance of identifying product misuse (OEM #03), because
guaranteeing uptime requires ensuring that industrial equipment operation complies with agreed terms of use. As a
third example, MROCorp's affordance (MRO #04) is further supported by AOCorp's affordance of performance
540 HERTERICH ET AL.
TABLE 6 Overview of idiosyncratic affordances
ID
Idiosyncratic
affordance (code
frequency) Sample quotations
Idiosyncratic Affordances of Analytics Organisation (AOCorp)
Organisational goal(s) of AOCorp: Provide scalable data-driven support and enable value co-creation in industrial smart
service ecosystems (16)
AO #01 Performance
benchmarking (14)
It starts with simple benchmarking topicsthat is, comparisons
between performances over time. This extends to more
sophisticated issues.(Service Line Manager, Asset Analytics Services,
AOCorp)
AO #02 Event triggering (36)[One of our service offerings] just triggers machine halts or switching
parameters between A and B. Customers can do this manually based
on the insights that we provide, or they can trust our system. I think
that this will start small: When introducing such a system, an
employee will have to confirm OK,’‘OK,’‘OK.When he has
pressed OKfor two years, and the decisions were good, he will
finally let our system take over.(Service Line Manager Energy
Management, AOCorp)
AO #03 Insights provisioning
(40)
We aimed at providing value services instead of hotlines or support
for problems. Therefore, we have established various service lines
that provide insights and various kinds of value to customers.
(Service Line Manager, Plant Cloud Services, AOCorp)
We need a system that allows us to address data-sharing topics in a
goal-oriented way.(Service Line Manager, Process Data Analytics,
AOCorp)
Idiosyncratic Affordances of Original Equipment Manufacturer (OEMCorp)
Organisational goal(s) of OEMCorp: Engineer safe, superior industrial products at a competitive price (25); offer
product-complementing services (48)
OEM #01 Product mix (117)We offer to optimize the product mix for the customer organization
considering the tasks that need to be accomplished with our
products.(Managing Director, Product Marketing and Communication,
OEMCorp)
OEM #02 Product usage
insights (45)
It would be valuable for us to have a platform where we have this
transparency to see the capacity use of all of our machines.(Head of
Logistics and Process Performance, OEMCorp)
If we know exactly how our products are used and where issues occur,
then we can improve future products or design upgrade kits to fix
existing products.(Managing Director, Product Marketing and
Communication, OEMCorp)
OEM #03 Identifying product
misuse (36)
What are the environmental conditions of our products?
Temperature? Dust and degree of air pollution? Because we want to
earn more money from our customers in a fair way, we have to
measure the environmental factors of product operations.(Managing
Director, Innovation Management, OEMCorp)
OEM #04 Product-
complementing
services (72)
Based on a cross-functional innovation initiative, our goal is to offer
entirely new services. [] Today, we can barely imagine the potential
of our products when they are augmented with digital technology.
(Director, Industrial Services, OEMCorp)
We need to get rid of this thinking in terms of steel and iron. We need
to sell more services instead of machines. We need to address our
customers' needs.(Director, New Business and Product Digitization,
OEMCorp)
(Continues)
HERTERICH ET AL.541
TABLE 6 (Continued)
ID
Idiosyncratic
affordance (code
frequency) Sample quotations
Idiosyncratic Affordances of Maintenance, Repair, and Overhaul Company (MROCorp)
Organisational goal(s) of MROCorp: Ensure error-free operations of industrial products in an efficient and effective
manner (98)
MRO #01 Triggering MRO
activities (31)
The goal of predictive maintenance is to predict when a component
will break down and then replace it during a regular service interval,
such as during the night before a failure happens, so we can increase
the overall availability of the product.(Vice President, Field Service,
MROCorp)
MRO #02 Empowering &
optimising field
service activities
(93)
Currently, all technicians have a mobile device [] to manage and
control their work. [] We need information on these devices from
each product instancereal-time information, historical information,
fault logs, [].(Vice President, Service Support, MROCorp)
MRO #03 Establishing remote
online diagnosis
(31)
An organizational function exists focusing on diagnosing and resolving
problems remotely to replace field visits, or if the field service agent
is not able to solve it, he or a smart algorithm finds the cause of the
problem and recommends a potential solution.(Director, Field Service
and MRO, MROCorp)
MRO #04 Managing product
operations and
guaranteeing
product uptime
(32)
The customer, at a certain point in time, says, Okay, you take care of
it. I just want my truck to run, to work properly, and to have no
downtime,and that's fine.(Director, Sales Region A, MROCorp)
We want to sell service contracts that say that we take care of the
customer's machine so that the customer can focus on his core
business.(Head of MRO Service, MROCorp)
Idiosyncratic Affordances of Smart Product Operator Company (SPOCorp)
Organisational goal(s) of SPOCorp: Integrate industrial products into own value co-creation (45); gain operational
transparency on value co-creation processes that utilize smart products (54)
SPO #01 Gaining transparency
on internal
processes to
manage work
orders based on
actual capacity (65)
It would be valuable for us to have timely machine data, which would
allow the foreman to see the status of work orders, such as the
current processing speed, estimated time of completion of the work
order, and disruptions in our production processes. We leverage
product data to derive our output and performance. Conclusions
about product use would also be great.(Director of Operations,
SPOCorp)
Besides asset data, contextual information such as What are the
workers doing?or Why are the machines not [operating] right
now?is needed.(Senior Manager, Maintenance and Facility
Engineering, POCorp)
We have an antiquated Enterprise Resource Planning system that's
more of an inventory system. [] We feed some operational data on
equipment operations into that system. Pairing some of that data
with the data from the equipment would help us see the whole
picture of what has occurred to that equipment over time.(Vice
President, Operations, Region B, POCorp)
SPO #02 Providing
transparency on
operations and
processing of
orders to
customers (48)
We can use the new data to provide our customers with more
transparency and information about the state of their orders.(Head
of Logistics and Process Management, SPOCorp)
542 HERTERICH ET AL.
benchmarking (AO #01), because operating industrial equipment as a service relies on the ability to manage opera-
tions efficiently. In summary, then, the formation of a broad set of idiosyncratic affordances leads to opportunities
for joint service innovations within the smart service ecosystem that can be exploited, depending on the specific
environmental conditions.
The formation of idiosyncratic affordances also generates and strengthens ties between organisations. Inter-
organisational ties relate to reciprocal dependencies, whose strategic role in providing smart services several inter-
viewees emphasised:
Real partnerships arise from reciprocal dependence. [] I think that we need to collaborate in such a
way [that] reciprocity is a continuous balancing act. (Managing Director, Innovation Management,
OEMCorp)
Structural integrity results from value propositions that participants in an ecosystem provide each other based on
their unique and complementary competencies. Our data indicate that the companies are aware of these comple-
ments, through which novel smart service opportunities may emerge. For example, OEMCorp relies on an AO's data
analytics competency:
[OEMCorp] owns the (proprietary) connectivity to our products. However, the balance of power in
the ecosystem looks very different when we go one level above and look at analytical capabilities.
This means that other actors provide information like Caution, this machine is most likely to go
down.(Service Product Manager, Out of the Box Analytics, AOCorp)
We conclude that a smart service ecosystem's structural flexibility and integrity manifest in idiosyncratic
affordances that are actor-specific and depend on the individual organisation's context and the organisational actor's
goals.
4.3 |The architecture of participation manifests in collective affordances
Collective affordances are those use potentials of smart products that are enacted by a group of participants in a
smart service ecosystem and result in joint provisioning of smart services. Our data indicate that collective
affordances are formed through the integration of various actors' idiosyncratic affordances. The mechanisms that
collective affordances build on top of idiosyncratic affordances are also reflected in our empirical data:
Yes, I am talking about this interconnectedness spanning organizational boundaries. Take the example
of the port of [City anonymized], with its smart and interconnected parking lots, cranes, and piers.
Individual organizations must first realize their use cases and benefits before they exchange informa-
tion, resulting in greater benefits beyond individual organizational boundaries. (Service Product Man-
ager, Out of the Box Analytics, AOCorp)
We identify two collective affordances that relate to two smart services: Managing and optimizing product
operations(CA #01) and Performance-based contracting.(CA #02). Table 7provides an overview of these two col-
lective affordances.
The affordance of Managing and optimizing product operations(CA #01) integrates a set of idiosyncratic
affordances of various organisations. Specifically, CA #01 integrates Event triggering(AO #02) from AOCorp;
Product usage insights(OEM #02), Identifying product misuse(OEM #03), and Product-complementing services
(OEM #04) from OEMCorp; Triggering MRO activities(MRO #01), Manage product operations and guarantee
HERTERICH ET AL.543
product uptime(MRO #05) from MROCorp; and Gaining transparency on internal processes to manage work
orders based on actual capacity(SPO #01) from SPOCorp. To manage and optimise product operations, events
like maintenance and repair activities must be triggered based on the industrial equipment's idiosyncratic con-
ditions. OEMCorp supports the optimization of product operations by relying on the data and identifying
misuse.
Our data suggest that these collective affordances create mechanisms through which the ecosystem partic-
ipants' contributions are coordinated, integrated, and synchronised. Collective affordances represent a shared
and agreed logic for the integration of the individual ecosystem participants' value propositions to produce
super-additive value in the form of smart services, as shown in the example of the collective affordance Man-
aging and optimizing product operations(CA #01). Such an architecture of participation is necessary because
developing smart services (i.e., collective affordances) across organisational boundaries requires more substan-
tial coordinated efforts than service innovation within a firm's internal boundaries (i.e., idiosyncratic
affordances) does:
We are constantly in search of smart, innovative services. However, in ecosystems, this is a little more
complicated than developing new products or increasing the internal efficiency of technical customer
service. (Managing Director of Product Marketing and Communication, OEMCorp)
Collective affordances require transparent rules for the exchange of resources and capabilities between actors.
For example, for the collective affordance Managing and optimizing product operations,all of the actors involved
share a common understanding of each other's contributions to the value proposition:
The customer will be able to be more effective and have to deal with less equipment because of the
products' increased uptime. Digital solutions like wearing sensors, as well as localization, will help to
increase the product's efficiency and lower the number of machines needed (Director of Industrial
Services, OEMCorp)
The ecosystem's participants also agree on how these value propositions are interconnected to provide value to
SPOCorp:
We draw on the expertise of various partners from different areas of expertise. We understand this
effort as a team effort and embrace participation in a regular innovation process. (Managing Director,
Innovation Management, OEMCorp)
TABLE 7 Overview of collective affordances
ID
Collective affordance
(code frequency) Sample quotations
Collective affordances within the entire smart service ecosystem
CA #01 Managing and
optimising product
operations (27)
If we establish a partnership with both our customers and [AOCorp], we can
go beyond just offering forklift truck control systems and actually manage
our customers' processes and finally manage their operations for them.
(Director, After-Sales and Customer Service, MROCorp)
CA #02 Performance-based
contracting (32)
Flexible pricing means for us to leverage seasonal utilizationfor instance,
seasonal complementing dynamic scaling of farmerstruck fleets in northern
and southern Italy as well as solving financing problems of forklift truck
operators. Usage-based pricing is a huge topic with many smaller sub-use
cases.(Managing Director, Innovation Management, OEMCorp)
544 HERTERICH ET AL.
Finally, the ecosystem parties must be aware of the multilateral connections required to produce super-additive
value to SPOCorp:
We consider this an ecosystem with multiple organizational actors who are all working on the same
challenge. I believe that organizational borders are blurred. (Service Product Manager, Out of the Box
Analytics, AOCorp)
Hence, we conclude that a smart service ecosystem's architecture of participation manifests in the collective
affordances of smart products that are enacted by a group of participants in an ecosystem and result in the joint pro-
visioning of smart services.
The three types of affordances we identified help us to explain how the case's smart service ecosystem
emerged. Specifically, we revealed two shared affordances that relate to the properties of smart products that allow
their operational states to be sensed and are jointly perceived by all ecosystem participants. Thirteen idiosyncratic
affordances are role-specific and depend on the goals the respective organisation associates with the material prop-
erties of smart products. Finally, two collective affordances describe the ultimate affordances of smart products that
are jointly enacted by a group of ecosystem participants and result in the joint provisioning of smart services.
5|DISCUSSION
This section describes the results we found in the data in light of the extant literature. We develop a model that
explains the emergence of smart service ecosystems, as presented in Figure 2. We explain and discuss this model in
detail and derive propositions that conceptualise the relationships shown in the model.
Based on the material properties of smart products, smart service ecosystems emerge through specialising shared
affordances, integrating idiosyncratic affordances, and jointly establishing collective affordances. Our data reveals
three socio-technical antecedents of smart service ecosystem emergence: a shared worldview,structural flexibility and
integrity, and architecture of participation. These antecedents manifest in the three classes of affordances and gener-
ate value propositions of smart service ecosystems.
The first antecedent, a shared worldview, manifests in shared affordances because, when participants in an eco-
system perceive shared affordances in a smart product, these participants exhibit low technological distance and
shared situational awareness. The literature on firm alliances uses the term technological distance,which refers to
how organisations differ in terms of their perceptions and understanding of their technology profile's role in compar-
ison to their partners in the ecosystem (Gilsing et al., 2008; Nooteboom et al., 2007; Wuyts et al., 2005). We argue
that a shared worldview reduces the technological distance among participants in an ecosystem and increases their
shared situational awareness of possibilities for the integration and exchange of resources. Sharing information in a
smart service ecosystem also increases shared situational awareness among its participants. Consequently, we for-
mulate proposition 1a as follows:
Proposition 1a. A shared worldview manifests in shared affordances.
The second antecedent, structural flexibility and integrity, leads organisations to determine what their unique
contributions to an emerging smart service ecosystem might be, recognising that they might have to alter how they
do things and collaborate with others to allow for flexible formation of configurations of actors and complementary
idiosyncratic affordances. Therefore, we argue that idiosyncratic affordances require a considerable degree of struc-
tural flexibility and integrity. The ecosystem literature also mentions that organisations need to develop strategies
that recognize and manage indirect links is one of the key distinctions between traditional strategy and ecosystem
strategy(Adner, 2017, p. 44). Through structural flexibility, an ecosystem's participants can react to external
HERTERICH ET AL.545
conditions by forming dynamic, cooperative constellations (Adner, 2017; Moore, 1997). In our case ecosystem,
MROCorp's organisational goal is to ensure error-free operations of industrial products in an efficient and effective
way, while OEMCorp's goal is to engineer safe, superior industrial products and complementary services for a com-
petitive price. Through structural integrity, strong ties are formed that hold the ecosystem participants together
(Lusch & Nambisan, 2015). For example, MROCorp's and OEMCorp's idiosyncratic affordances are complementary
to SPOCorp's idiosyncratic affordances. Therefore, we state proposition 1b:
Proposition 1b. Structural flexibility and integrity manifest in idiosyncratic affordances.
The third antecedent, architecture of participation, ensures collaboration across the ecosystem by providing
mechanisms to ensure coordination, integration, and synchronisation of the ecosystem's participants (Lusch &
Nambisan, 2015, p. 165). Collective affordances determine how to coordinate, integrate, and synchronise the contri-
butions of the ecosystem's participants to realise complex smart services, thus specifying an architecture of partici-
pation. Our case ecosystem's architecture of participation fosters collaborative development of joint value
propositions for the end customer. For example, the affordance Performance-based contracting(CA #02), allows
industrial products like forklifts to be billed by the amount of use they deliver (e.g., the number of tonnes they move
within a settlement period, including wear and tear and maintenance). According to the Managing Director, Innova-
tion Management of OEMCorp, partners from different areas of expertise [collaborate and] understand this effort
as a team effort and embrace participation in a regular innovation process.Accordingly, architecture of participation
allows an ecosystem's participants to operate in unison without hierarchical structures or control mechanisms
(Lusch & Nambisan, 2015). An architecture of participation is characterised mainly by an ecosystem's institutional
arrangements, such as rules, norms, and practices used by actors to coordinate actions (Lusch & Nambisan, 2015).
These rules, norms, and practices are a result of the non-fungibility of ecosystem actors' investments in connecting
to a smart service ecosystem (Jacobides et al., 2018), bringing us to proposition 1c.
Proposition 1c. An architecture of participation manifests in collective affordances.
Our data also shows that the emergence of a smart service ecosystem is a process of affordance specialisation
and integration that leads to joint value propositions. Organisational actors draw on shared affordances to form
FIGURE 2 Theoretical model explaining the emergence of smart service ecosystems based on material properties
of smart products and socio-technical antecedents
546 HERTERICH ET AL.
idiosyncratic affordances through specialisation, taking into account their organisations' goals and capabilities. Spe-
cialisation refers to the actors' differing perspectives on which smart products can be used as complements because
of their generative nature or capacity (Cennamo & Santal
o, 2019; Yoo et al., 2012). In our case, each ecosystem actor
builds on shared affordances to come up with their specialised idiosyncratic affordances. For example, OEMCorp
specialises in Product-complementing services(OEM #04) by drawing on both Sensing equipment state(SA #01)
and Actuating equipment state(SA #02). With the example of an elevator as a smart product, a product-
complementing service could be a continuous overweight check (drawing on Sensing the equipment state) that
would avoid damaging the elevator by preventing it from moving (drawing on Actuating equipment state) if over-
weight is detected. A maintenance organisation like MROComp might build on these two affordances in a different
way, such as by identifying misuse that might void the warranty.
Idiosyncratic affordances build on the recombinant and generative innovation potential of shared affordances'
laying the foundations for materialising joint value propositionsthat is, complex smart services in a smart service
ecosystem (Cennamo & Santal
o, 2019; Lyytinen et al., 2016; Thomas & Ritala, 2021; Yoo et al., 2012). As such, our
notion of idiosyncratic affordances supports the rationale for a set of an ecosystem's actors' multilateral relationships
that are due to these actors' complementarities (Adner, 2017; Jacobides et al., 2018; Shipilov & Gawer, 2020).
In summary, based on our data, we argue that the emergence of a smart service ecosystem depends on socio-
technical antecedents. Further, the multilateral linking of an ecosystem's participants builds on the specialisation of
the actors' activities, specifically their adoption of shared affordances (Adner, 2017). In this process of specialisation,
a participant in an ecosystem establishes its position in the flow of the ecosystem's activities that materialise focal
value propositions (Adner, 2017). This logic leads us to proposition 2a.
Proposition 2a. Idiosyncratic affordances are the result of a process in which actors specialise shared
affordances to pursue idiosyncratic organisational goals.
Collective affordances arise through the integration of idiosyncratic affordances; that is, the integration of idio-
syncratic affordances related to a smart product is a decentralised, emergent process that results in the collective
affordances that underlie ecosystems' joint value propositions. For example, the collective affordance Performance-
based contracting(CA #02) requires the actors in our case's smart service ecosystem to integrate Product usage
insights(OEM #02) and Managing product operations and guarantee product uptime(MRO #04). As the Managing
Director of Innovation Management at OEMCorp observed, this collective affordance results from integrating a set
of idiosyncratic affordances from various actors in the ecosystem:
Usage-based pricing is a huge topic with many smaller sub-use cases. (Managing Director, Innovation
Management, OEMCorp)
Demarcating three classes of affordances and the mechanisms of specialisation and integration highlights the
role of the collective affordances that underlie the realisation of joint focal value propositions as the ultima ratio or
raison d'être of a smart service ecosystem. While the process of specialisation allows a participant in the ecosystem
to position itself in the flow of activities, the process of integration culminates in multilateral ties that are bound by a
smart products' capacity for action (Adner, 2017). Consequently, we formulate proposition 2b:
Proposition 2b. Collective affordances are the result of a convergence process in which actors inte-
grate the idiosyncratic affordances that lead to an ecosystem's joint focal value propositions.
Our results indicate that the emergence of a smart service ecosystem relies on the generative nature of smart prod-
ucts' material propertiesthat is, their generative capacity for action and allowance for non-designed useand conver-
gence properties, referring to merging formerly separate entities (Beverungen et al., 2019; Cennamo & Santal
o, 2019).
HERTERICH ET AL.547