ArticlePDF Available

Platform adoption by mobile application developers: A multimethodological approach


Abstract and Figures

This paper investigates the factors that influence the adoption of IT platforms by software developers and how those factors differ from those that influence IT adoption by end-users. We take a multi-methodological approach, beginning with an interpretive field study where we interview mobile application developers. In the initial interpretive phase, we identify a comprehensive set of influences on developers' platform adoption, comparing them with the factors that have been identified in previous studies of end-user adoption, noting key differences. In the second phase, we empirically test the factors identified in our interviews. We find several key differences between end-user adoption of IT and developer adoption of IT platforms. Most notably, we observe the importance of network externality considerations when developers make an adoption decision, a consideration that is largely absent for end-users. Our study is among the first to comment on B2B and B2C issues in the adoption phenomenon where developers adopt a platform as technology producers (a B2B consideration) in order to ultimately provide mobile applications to end-users who are technology consumers (a B2C consideration).
Content may be subject to copyright.
Platform adoption by mobile application developers: A multimethodological approach
Jaeki Song
,Jeff Baker
,Ying Wang
,Hyoung Yong Choi
,Anol Bhattacherjee
Texas Tech University, United States
American University of Sharjah, United Arab Emirates
Northern Illinois University, United States
Georgia State University, United States
University of South Florida, United States
abstractarticle info
Article history:
Received 27 July 2017
Received in revised form 13 November 2017
Accepted 26 December 2017
Available online xxxx
This paper investigates the factors that inuence the adoption of IT platforms by software developers and how
those factors differ from those that inuence IT adoption by end-users. We take a multi-methodological ap-
proach, beginning with an interpretive eldstudy where we interview mobile application developers. In the ini-
tial interpretive phase, we identify a comprehensive set of inuences on developers' platform adoption,
comparing them with the factors that have been identied in previous studies of end-user adoption, noting
key differences. In the second phase, we empirically test the factors identied in our interviews. We nd several
key differences between end-user adoption of IT and developer adoption of IT platforms. Most notably, we
observe the importance of network externality considerations when developers make an adoption decision, a
consideration that is largely absent for end-users. Our studyis among the rst to comment on B2B and B2C issues
in the adoption phenomenon where developers adopt a platform as technology producers (a B2B consideration)
in order to ultimately provide mobile applications to end-users who are technology consumers (a B2C
Published by Elsevier B.V.
IT platform adoption
Mobile applications
Quantitative research
Network externalities
1. Introduction
The rising importance of developmentplatforms for application soft-
ware (software commonly referred to as appsor add-on extensions)
represents a major shift in software development. Rather than hard-
ware or software rms overseeing the majority of software develop-
ment, third-party developers now play a substantial role. Since the
launch of the Apple iPhone in 2007, Apple alone has paid out more
than $16 billion in royalties to app developers, and has created or sup-
ported over 1.5 million jobs in the US [1]. In Europe, the app economy
has created over 1.6 million jobs and continues to grow [2].BothApple
and Android platforms have achieved considerable scale, with each hav-
ing over 100 billion downloads of over 2 million available applications
[3]. Given the number of applications, th e amount of revenue generated,
and the number of new developers,there is clearly a need to understand
why developers choose particular platforms to develop and distribute
their applications.
In this research project, we examine two key research questions. The
rst is, What factors inuence developers' adoption of IT platforms?IT
adoption by end-users has been thoroughly studied by researchers, and
numerous theoretical explanations have been developed. These theo-
ries include the technology acceptance model [4], the unied theory of
technology acceptance and usage [5], diffusion of innovations theory
[6], and social cognitive theory [7,8]. One common theme in this extant
research is that each of these explanations focuses on how the charac-
teristics of the IT inuence its adoption. Additional factors, however,
may also be important when considering the adoption of platforms by
developers. Unlike IT that is adopted by end-users for its own sake, plat-
forms are chosen by developers for the purpose of building applications
that may be monetized for prot. That is, technologies are adopted by
technology consumers while platforms are adopted by technology
producers. The value of an IT platform to a developer depends largely
on the expected user base of that platform. Value accrues in the future
and may not be certain or apparent in the present. Platform adoption
may also differ from IT adoption because utilizing an IT platform for de-
velopment often requires advanced technological knowledge in pro-
gramming languages, software development, and various operating
systems. Thus, the adoption of IT platformsby developers may be inu-
enced more by network externality and developers' related knowledge
than by the technology characteristics identied in research that studies
IT adoption by end-users. Therefore, our second research question is,
How do the factors that inuence developers' adoption of IT platforms
differ from those that inuence end-users' adoption of IT?
Decision Support Systems xxx (2018) xxxxxx
Corresponding author.
E-mail addresses: (J. Song), (J. Baker), (Y. Wang), (H.Y. Choi),
(A. Bhattacherjee).
DECSUP-12914; No of Pages 14
0167-9236/Published by Elsevier B.V.
Contents lists available at ScienceDirect
Decision Support Systems
journal homepage:
Please cite this article as: J. Song, et al., Platform adoption by mobile application developers: A multimethodological approach, Decision Support
Systems (2018),
Identifying and testing the factors that inuence platform adoption,
then describing any differences when compared to the factors that in-
uence IT adoption, and subsequently developing a theoretically-
based explanation of these inuences is the primary intended contribu-
tion of this research. Toward that end, we employ a multi-methodology
approach. In the rst phase of our study, we interview mobile applica-
tion developers, noting their reasons for choosing to develop for a par-
ticular platform. We also compare these developers' reasons with
those that have been identied in prior research on IT adoption by
end-users. After identifying a comprehensive set of factors related to
platform adoption, we test these factors in a second phase by surveying
mobile application developers. We then discuss the differences in the
factors that we have found to inuence platform adoption and those
that inuence IT adoption. The demonstration of this multimethodology
approach is a secondary contribution.
The paper proceeds as follows. In our Literature review section, we
dene and describe IT platforms, then describe foregoing research on
IT adoption by end-users. In our Research methodologies section, we
explain the initial interview phase of our research and the subsequent
survey phase. Findings from our interviews and statistical results from
our surveys are presented as well. We then discuss the theoretical and
practical implications of our ndings, highlighting key differences be-
tween users' IT adoption and developers' platform adoption. We note
future research opportunities before concluding.
2. Literature review
2.1. IT platform adoption
An IT platform is the extensible codebase of a software-based sys-
tem that provides core functionality shared by the modules that inter-
operate with it and the interfaces through which they interoperate
([9,10], p. 676). It is a general purpose technology that enables the cre-
ation of derivative products such as application software and relevant
business opportunities [11,12]. Apple's iOS and the Android operating
system are examples of platforms for the development of modules for
smartphones and tablets; Mozilla Firefox is an example of a platform
for the development of modules for internet browsers. A module is an
add-on software subsystem that connects to the platform to add func-
tionality to the platform([10], p. 676). Smartphone apps, tablet apps,
and browser extensions are examples of modules. The platform, the
modules specic to that platform, and thehardware on which the mod-
ules run is known as the platform's ecosystem [10,13]. Platform-centric
ecosystems are a reality of competition in many industries [14]. This is
particularly evident in software development and distribution [10].
Adoption is the decision to make full use of an innovation as the
best course of action available.[6]. IT adoption refers to a decision
about adopting a technological innovation [15],adenition widely ap-
plied in IS, including in the investigation of IT usage [16], IS adoption
in small businesses [15], diffusion of IS outsourcing [17],andtheadop-
tion of IS appliances [18]. Platform adoption is thus a decision about
adopting a software-based system that enables developers to produce
a stream of derivative products.
The adoption of IT platforms to date has been considered largely an
organizational decision, because software development has historically
taken place within organizations to enhance organizational capabilities
and business value [11,19]. Indeed, researchers have suggested that IT
platform adoption is an organizational decision that can be explained
from the perspectives of technology strategy, organizational learning,
technology bandwagon, and adaptation [11]. However, since many mo-
bile applications are built by individual developers, sometimes outside
organizational constraints, individual-level factors may also be critical.
Indeed, individuals are one of the driving forces behind the app econ-
Given that thedevelopment of software applications and other types
of software modules is increasingly being driven by individuals, it is
important to examine how individuals choose a development platform.
Platforms must actively compete for developer adoption [10] and since
IT products have little or no value in isolation, factors such as popularity
and compatibility intensity signicantly inuence developers' adoption
decisions [14]. Individual characteristics such as knowledge, personal
benet, and enjoyment inuence end-users adoption decisions [22]
and may also inuence developers' adoption decisions. In IT adoption
literature, research has consistently shown that interactions in society
inuence adoption decisions of end-users [4,23,24].Furthermore,plat-
form adoption choice is market-driven because developers adopt plat-
forms to create IT products for prots [11]. Researchers have shown
that end-users tend to adopt more marketable software [25].Toattract
end-users, developers are likely to adopt platforms that have great mar-
ket potential and provide advanced technical support [25,26].Insum,IT
platform adoption by developers might differ from IT adoption by end-
users and be inuenced by factors such as platform characteristics, indi-
vidual characteristics, social interaction, and network externalities.
Since decisions about adopting development platforms are made by in-
dividuals who have some of the same considerations as end-user
adopters of IT, we next review theories of individual adoption.
2.2. IT adoption
The factors that inuence individuals' IT adoption decisions have
been studied extensively. Theories of adoption include the Theory of
Reasoned Action (TRA), Theory of Planned Behavior (TPB), Technology
Acceptance Model (TAM), and its extensions TAM2, TAM3, and the Uni-
ed Theory of Acceptance and Use of Technology (UTAUT). This family
of theories has identied perceived usefulness, perceived ease of use,
subjective norm, image, job relevance, computer self-efcacy, facilitat-
ing conditions, computer anxiety, and perceived enjoyment, as the key
drivers of individuals' intention to use IT and, as drivers of individuals'
actual IT adoption behavior [5,2731]. Voluntariness and experience
are noted as the key moderating factors in several models.
Other researchers have developed Innovation Diffusion Theory
(IDT), which describes relative advantage, complexity, image, compati-
bility, and results demonstrability as predictors of individuals' IT adop-
tion [6,32,33]. In social cognitive theory (SCT), outcome expectations,
self-efcacy, affect, and anxiety inuence adoption [7,8]. Each of these
theories of IT adoption may be relevant when considering IT platform
When considering individual-level theories of adoption, an impor-
tant consideration to note is that end-users adopt technologies for the
sake of the specic technology. In contrast, developers adopt platforms
in order to build end-user applications. In the present study, adopters
are not consumers choosing among various platforms [34]; here
adopters are individual software developers who must decide which
platform to adopt. Developers are technology producers, not technology
consumers. Thus, while factors from individual-level theories of end-
users' IT adoption may be relevant, the developer adopter of a platform
is nevertheless distinct from the end-user adopter of an IT. This is one
reason why the factors inuencing IT platform adoption may differ
from those that inuence IT adoption.
We therefore surmise that the adoption of IT platforms is inuenced
not only by the specic technological characteristics of those platforms,
such as the characteristics identied in TRA, TPB, TAM, UTAUT, IDT, and
SCT, but also by the platform's popularity, network externalities, and
potential future value [14,35,36]. Indeed, software developers' adoption
decisions are fraught with uncertainty about the future benets of the
platform, potentialcosts of making the wrongdecision, and the irrevers-
ibility of the decision [11].
Additionally, a developer must have a certain level of technical
knowledge and programming skills in order to utilize a platform prop-
erly. Technical skills such as these are largely irrelevant for an individual
end-user. Thus, again, developers have different considerations when
2J. Song et al. / Decision Support Systems xxx (2018) xxxxxx
Please cite this article as: J. Song, et al., Platform adoption by mobile application developers: A multimethodological approach, Decision Support
Systems (2018),
adopting an IT platform than end-users do when adopting an IT. Thus,
we now investigate the factors inuencing IT platform adoption.
3. Research methodologies
The overall goal of this research project is to develop a theoretical
model of IT platform adoption, elaborating the key factors inuencing
individual developers' adoption behavior. Toward that end, we conduct
two separate studies with differing methodologies. First, we take an in-
terpretive approach in order to gain a comprehensive understanding of
the inuences on developers' platform adoption decisions. From this in-
formation, we develop an IT platform developer model using survey
data, described in the Study 2 quantitative study section. We test the
proposed model using a mixed methods approach across the two stud-
ies, enabling us to evaluate the robustness and generalizability of our
proposed theoretical model [37,38]. We focus on individual developers'
adoption behaviors, whose decisions are determined by factors
inuencing individual perception of the platforms. Developers are free
to choose platforms when developing applications; those who might
be forced to adopt certain platforms by organizational decisions are be-
yond the scope of our research.
3.1. Study 1 interpretive eld study
In the interpretive approach, researchers collect and analyze data to
identify concepts and categories that are evident in data. In our study,
interview data was collected and then analysis proceeded through a se-
ries of steps, including: (a) iterative data collection, (b) generation of
concepts, (c) development of categories and relating these categories
to concepts, (d) theoretical sampling, and (e) use of the coding para-
digm of conditions, contexts, strategies, and consequences to discuss
the phenomenon of interest [3943].
3.1.1. Data collection
In Study 1, data was collected from 35 developers currently involved
in application development forthe Apple iPhone, Google Android, or the
Microsoft Windows Phone platforms in South Korea. Semi-structured
interviews were used. Interviewees' ages ranged from 31 to 42, with a
mean of 36.08. All held either bachelor's or master's degrees, with
most holding bachelor's degrees. Full-time professional work experi-
ence ranged from 1 to 38 months with a mean of 18.38 months. Inter-
view subjects were asked a series of questions about their platform
adoption decision (see Appendix A for the interview protocol). They
were asked which platforms they considered; about their experience
with theirchosen platform, and about their future platform usage inten-
tions. Interviews were conducted in the Korean language and translated
into English for coding. A professional translator translated the original
interviews into English; another translator then translated them back to
Korean to ensure equivalence of translations.
3.1.2. Data analysis generation of concepts and development of categories
Interview transcripts were analyzed using open, axial, and selective
coding [44]. Coding was conducted by two researchers experienced in
qualitative data analysis. The initial phase, open coding, involves each
coder examining the transcripts lineby line to uncover conceptsthat po-
tentially explain developers' platform adoption decisions. Each concept
was discussed and labeled by consensus. After the process of open cod-
ing, concepts that were similar to one another were grouped into broad,
generalizable, higher order categories, which are akin to theoretical con-
structs. The development of these categories helped reduce the large
number of concepts to a more manageable list, from which a pattern
could be discerned of the inuences shaping platform adoption. Wher-
ever possible, we employed construct names from existing literature to
label our categories, since the goal of our research was toextend current
theories of adoption. We also did this to avoid duplication of terminol-
ogy and the confusion that could result. During open coding, we also
identied the characteristics or properties of each category to identify
patterns of covariation among these constructs. Additionally, to under-
stand the relative salience of a concept, we also tabulated the number
of respondents mentioning each concept.
Open coding identied 62 concepts that were grouped into 27 cate-
gories. To focus on the more salient predictors and ensure parsimony,
we decided to only consider those categories that were mentioned by
ve or more respondents. This resulted in the selection of 15 categories
as constructs of interest. Among these 15 categories were IT security
(susceptibility to hacking, viruses, and app replication) and platform
potential(potential for the platform to be a leader or part of future
technology trends). Because interviews revealed that IT securityand
platform potentialshare some similarities with relative advantage,
and are generally subsumed by the relative advantageconstruct,
they were not included in the survey phase. Additionally, forced
choice(where company directives mandated use of a specicplat-
form) was mentioned by some developers. Since such circumstances
do not reect a free adoption choice, it was dropped from further anal-
ysis. The remaining 12 categories are listed in Table 1.
The categories are numbered and grouped into four broad sets:
perceived platform characteristics, perceived network externalities,
individual characteristics, and social interaction. The number of respon-
dents that mentioned each predictor is noted in parentheses. An exam-
ination of the frequency counts suggested that network externality-
related factors (market potential, marketability, and developer tools)
are the dominant drivers of mobile platform adoption, followed by rel-
ative advantage and related knowledge, while other predictors played a
lesser role.
3.1.3. Data analysis - use of conditi ons, contexts, strategies, and consequences
The next pha se was axial coding,which refers to the process of devel-
oping main categories and sub-categories [44].Inaxialcoding,we
linked our categories using causal relationships to develop an initial ex-
planation of developers' platform adoption. Categories were classied
as either causes, effects, or conditions (circumstances in which the phe-
nomenon is embedded). After axial coding, selective coding was the last
step of interpretive analysis, which involves the integration of the cate-
gories that have been developed to form the initial theoretical frame-
work [44]. In selective coding, we aimed to integrate the categories
and relationships identied into a holistic theory. Designating IT plat-
form adoption as our central category, we systematically and logically
related this category to other categories representing its causes. We
employed diagramming as an integration technique, to depict relation-
ships between categories. Thisdiagram, shown in Fig. 1, represented our
interpretive theoryof IT platform adoption,and along with Table 1 is the
primary outcome of Study 1.
3.2. Study 2 quantitative study
After Study 1, we examined existing literature to see if the relation-
ships indicated in our model had been previously studied. While none of
the relationships have, to our knowledge, been studied in the context of
platform adoption by developers, we nevertheless want to be informed
by work on IT adoption by end-users. The additional detail provided in
foregoing research enabled the development of a more nuanced
model where some concepts/constructs inuence others (for instance,
some of the concepts/constructs listed as perceived platform charac-
teristicsinuence one another, while some inuence adoption di-
rectly). We now formally present our hypotheses and a rened
research model.
3.2.1. Hypothesis development
Relative advantage is one of the concepts that we identied within
the larger grouping of platform characteristics. Relative advantage,
which is dened as the extent to which an innovation is perceived as
being better than its precursor([32] p. 195) and has been considered
3J. Song et al. / Decision Support Systems xxx (2018) xxxxxx
Please cite this article as: J. Song, et al., Platform adoption by mobile application developers: A multimethodological approach, Decision Support
Systems (2018),
an important antecedent of innovation adoption [6,15,32,51,52].Re-
searchers have stated that relative advantage is similar to the notion
of perceived usefulness in technology acceptance [5,32]. Innovation
studies indicate that different types of innovation generate different de-
grees of advantage. To verify this relationship from prior research, we
H1a. Relative advantage is positively associated with platform adoption.
Researchers have argued that product innovativeness is an anteced-
ent of product advantage and is positively associated with it [53,54].
Similarly, product advantage assesses the degree to which a product of-
fering is superior to that of competing products [55], which is consistent
with the concept of relative advantage (e.g. [32], p. 195). Thus, product
innovativeness may lead to products with better features or perfor-
mance, creating relative advantage and thus increasing the perceived
usefulness of the product. For example, if a platform is highly innovative
and frequently provides developers with new functionality, features, or
benets, these innovations provide an advantage over competing plat-
forms. Thus, the relative advantage of a platform is enhanced by the plat-
form innovativeness. Therefore,
H1b. Platform innovativeness is positively associated with relative
In diffusion of innovation literature, Rogers [6] denes compatibility
as the degree to which an innovation is perceived as being consistent
with the existing values, past experiences, and needs of potential
adopters(p. 224). He asserted that compatibility is one of the major
drivers for people to adopt innovations or new technologies. Elsewhere,
compatibility is similarly dened as the perceived cognitive distance
between an innovation and precursor methods for accomplishing
tasks([56], p. 784), and prior studies have found that compatibility is
positively related to individuals' adoption behavior. That is, an innova-
tion is more likely to be adopted when it is perceived as consistent
with existing systems, procedures, and values of the potential adopters
[6,32,56,57].Technical compatibility is an innovation's compatibility
with the present system, including hardware and software [58]. There-
fore, technical compatibility produces organizational benets by
enabling individuals or organizations to easily integrate new technolo-
gies with existing technologies or systems [58,59]. In line with this, de-
ning technical compatibility as the extent to which a platform is
compatible with or could be integrated with other programming tech-
nologies, we propose that individuals are more likely to adopt the IT
platform that is technically compatible.
H1c. Technical compatibility is positively associated with platform
At the organizational level, openness refers to the extent to which an
organization exchanges information with its environment [6].The
openness of organizations, systems, and technologies has been regarded
as a key factor of adoption behaviors in various contexts such as innova-
tion diffusion [6], technology adoption in healthcare [60], IT adoption at
the rm level [61], and Internet adoption ([62]). Researchers have
stated that the openness of a technology makes it easy for individuals
to exchange the ideas and knowledge across organizational and geo-
graphical boundaries ([62]). As a result, technology openness motivates
and facilitates organizational adoption of the technology. Previous re-
search proposed that openness encourages the organization to (1) ex-
change new ideas, (2) evaluate new ideas more easily, and (3) more
rapidly adopt the innovations [6,60]. Some have proposed that open-
ness offers the technical value of beneting from the collected informa-
tion as well as the psychological value of promoting trust, which in turn
increases the chances of adoption [63,64].
Consistent with the literature on technology openness, studies in-
vestigating platforms also view platform openness as a governance-
related concept reecting the trade-off between retaining and
relinquishing control over a platform [65]. There are two types of plat-
form openness: horizontal openness refers to giving up some control
by licensing the platform to competitors or sponsors; vertical openness
refers to opening theplatform to third-party developers of complemen-
tary applications [66]. Since we focuson developers' adoption, we focus
on vertical openness. A platform is more opento the extent that it
places fewer restrictions on participation, development, or use across
its distinct roles, whether for developers or end user [66]. These restric-
tions could be technical in nature. Therefore, if platforms have less re-
strictions on required technologies, the possibility would be high for
developers to integrate theirexisting technology with those in the plat-
form. Platform openness is positively associated with technical compat-
ibility. Formally,
Table 1
Predictors identied in open coding.
Perceived platform characteristics refer to salient technological attributes that
inuenced adopters' choice of that platform. In our sample of application
developers, these characteristics included:
1. Relative advantage: the perceived strength of an IT platform relative to competing
platforms; this included IT platforms' performance, debugging capability, conve-
nience of use, and ease of development. (mentioned in 27 interviews)
2. Platform innovativeness: the perceived novelty of the platform, containing use-
ful functions, innovative image, and benets. (7)
3. Technical compatibility: the extent to which a platform was compatible with or
could be integrated with other programming technologies. (6)
4. Platform openness: the extent to which a platform used open source softwar e;
openness relates to the easing of restrictions on the use, development, and
commercialization of a technology.([45] p. 1851) (6)
Perceived network externalities refer to direct or indirect value expected from a
large network of adopters for a given IT platform. For network-hosted applications,
network externalities are presumably a key driver of adoption that is often
overlooked in technology acceptance research. Three network externality factors
emerged from our analysis as salient to IT platform adoption:
5. Market potential: the expected size or growth of the future market for the
platform. More platform users translate into greater revenue opportunities for
application developers. (21)
6. Marketability: the existence of a current market or app ecosystem based on the
IT platform where developers are willing to pay to enter in order to obtain the
desired good or service [46]. (16)
7. Developer tools: the availability of tools to support application development
and implementation. A large network size motivates vendors to supply tools
that can boost developers' presence and participation in the network. (25)
Individual characteristics are personal characteristics that explain differential
patterns of adoption across the developer population. In this category, two
motivational forces and two individual differences emerged from our analysis:
8. Personal benet: the potential of earning revenue through application
deployment on a platform (an extrinsic motivation). (15)
9. Enjoyment: personal expectations of fun or goal achievement from devel-
oping applications (an intrinsic motivation). Prior research suggests that
both extrinsic and intrinsic motivations are key drivers of IT adoption [47],
and that the same appears to hold true for IT platform adoption. (5)
10. Related knowledge: developers' prior knowledge of web services, Java, and
related tools which are needed to build and deploy applications on their
intended platform. (20)
11. Personal innovativeness: developers' propensity to experiment with and try out
a new platform. This construct is also known to inuence IT adoption [48].(6)
Social interaction is a link or complex interaction established via reciprocity
behavior between two actors who perceive themselves to be interdependent
[49,50]. From our analysis, this interaction refers to:
12. Social inuence, which can be dened as the degree to which an individual
perceives that important others believe he or she should use the new system
[5]. From our analysis, social inuence (mentioned in 9 interviews) inuences
IT platform adoption, because developers typically get new information
regarding platforms and application markets from their professional peers
and colleagues. (8)
4J. Song et al. / Decision Support Systems xxx (2018) xxxxxx
Please cite this article as: J. Song, et al., Platform adoption by mobile application developers: A multimethodological approach, Decision Support
Systems (2018),
H1d. Platform openness is positively associated with technical c ompatibility.
Marketability refers to the existence of markets with the ability
and willingness to pay to obtain the assets, products or services
[46].Morespecically, marketability enables rms to decide which
technologies would be appropriate in commercializing new products
or services [6770]. Researchers have examined the effect of market-
ability when rms commercialize new technology products [68],
noting that it plays an important role in determining the develop-
ment and commercialization of new technologies [71]. Still others
have empirically tested the role of marketability of technology on -
nancial performance in small or medium-sized enterprises [70].In
addition, IT-driven marketability stimulates herding behavior [72],
and leads to the expansion of market size. Conversely, people are
likely to perceive greater risk of using a software when it loses mar-
ketability; thus, people tend to adopt and use more marketable soft-
ware [25]. Marketability is therefore a salient factor in determining
individuals' adoption of new technologies or products. In the current
study, we extend the notion of marketability into the context of IT
platform adoption. Dening marketability as the existence of a cur-
rent market or app ecosystem based on a development platform,
we therefore propose that marketability is an inuential factor in IT
platform adoption.
H2a. Marketability is positively associated with platform adoption.
The inuence of marketability has been explained by factors such as
market potential, market attractiveness, market competition, customer
needs, and legal regulation [68,70,73]. The marketability of technology
is specically inuenced by market potential, market characteristics,
and product competitiveness [70]. Based on prior literature, and on
the ndings of Study 1, we propose that, in the context of development
platforms, marketability can be inuenced by two factors: market po-
tential and the availability of developer tools.
Market potential has been studied primarily in the marketing litera-
ture. In our study, market potential indicates direct network externali-
ties. More platform users translate into greater revenue opportunities
for application developers, giving developers greater incentive to
adopt a specic IT platform. Market potential represents the attractive-
ness of a market and it inuences rms' marketing expansion strategy
when new products are diffused across products and countries [74].
Market potential is also regarded as a critical determinant of target mar-
ket selection and its relevant investment [75,76]. For instance, a rm
may be willing to take risks for marketing and nancial investments
when the market potential gives opportunities for returns [75]. In the
IS literature, researchers dene market potential as the total number
of potential users expected during the entire product lifetime [77]. Mar-
ket potential affects a new product adoption based on the framework
for innovation diffusion [26]. Market potential is represented by market
growth and future market size [68], and it has been considered a key di-
mension of marketability [67,68,70,73]. Therefore, in this study, we de-
ne market potential as the expected size or growth of the future
market for the development platform; and we propose that market po-
tential inuences platform marketability.
H2b. Market potential is positively associated with marketability.
Second, developer tools are the availability of tools to support applica-
tion development and implementation. The adoption of a development
platform is likely to be market-driven and depends on the characteristics
of available platforms [78]. Developers will adopt and use new tools and
methods with sufcient technical support and availability [25]. The avail-
ability and popularity of development platforms attract more people to
develop, thereby enabling the platform to be more marketable by offering
more numerous applications, as well as ones of higher quality and with
lower prices. This expands the pool of end-users who purchase more ap-
plications for the specic platform, which motivates more developers to
participate in the platform. This virtuous cycle indicates that developer
tools can be regarded as an indirect externality, and as a critical factor of
marketability in the context of IT platform adoption.
More developer tools do not necessarily increase adoption. Developers
must consider these tools useful and helpful, brining benets such as in-
creased revenue or efciency. Marketability enables rms to decide
which technologies would be appropriate in commercializing new prod-
ucts or services [67,69,70]. Marketability determines which developer
tools are desired and inuence adoption. Therefore, we propose that mar-
ketability plays a mediating role between developer tools and platform
adoption. Taken together, we propose that developer tools inuence mar-
ketability, which in turn, inuences platform adoption.
H2c. Developer tools are positively associated with marketability.
Social inuence refers to perceived pressure from social networks to
make or not to make a certain behavioral decision([79], p. 249). It has
been regarded as a salient factor in technology and innovation adoption
[79]. A number of research studies investigate the role of social inu-
ence in technology adoption. Social inuence is a form of subjective
norm in the TRA [16]; others examine its inuence on individuals' com-
mitment to and use of new IT [80]; still others extended the effect of dif-
ferent types of social inuences, including subjective norm, image, and
voluntariness, to individuals' pre- and post-adoption beliefs and atti-
tudes. Finally, social inuence has been embedded in theoretical per-
spectives such as TAM2 and UTAUT [5,16,31,32,79,81]. Thus, when
there is uncertainty about platform adoption, individuals are likely to
interact with their social network to help them decide. We hypothesize:
H3. Social inuence is positively associated with platform adoption.
According to cognitive evaluation theory [82], individual motiva-
tions can be classied as either extrinsic or intrinsic motivations [83].
The former drives an individual to perform a behavior to achieve spe-
cic goals such as rewards, whereas the latter relates to perceptions of
pleasure and satisfaction from performing a behavior. Prior IS research
indicates that extrinsic and intrinsic motivations play a critical role in
determining individuals' behavioral intention to adopt a technology
IT Platform
Perceived Platform Characteristics
Relative Advantage
Platform Innovativeness
Technical Compatibility
Platform Openness
Perceived Network Externalities
Market Potential
Developer Tools
Individual Characteristics
Personal Benefits
Related Knowledge
Personal Innovativeness
Social Interaction
Social Influence
Fig. 1. Preliminary model of IT platform adoption.
5J. Song et al. / Decision Support Systems xxx (2018) xxxxxx
Please cite this article as: J. Song, et al., Platform adoption by mobile application developers: A multimethodological approach, Decision Support
Systems (2018),
[31,84,85]. For both motivations, previous studies have identied per-
ceived benets [8587] and perceived enjoyment [31,84,85,88,89] as
major factors that inuence adoption.
There has been evidence in various contexts that perceived benets
have a positive inuence on individuals' adoption. Perceived benets in-
clude lower transaction costs, faster transaction speed, and better infor-
mation transparency [85,86]. Perceived benets, including benets that
arise from technological factors, inuence technology and innovation
adoption [90,91]. Inthe platformcontext, subsidies from Apple andAn-
droid (over $7 billion as of 2012) lure developers to write applications
for them [92], resulting in growth in the number of developers [93].
Thus, individuals are likely to adopt specic platforms when the plat-
form provides greater personal benet than existing platforms or com-
peting platforms.
H4a. Perceived personal benet is positively associated with platform
IT knowledge helps individuals to form a clear perception about IT
adoption; it amplies the perception of potential benets and increases
their condence [79]. In addition, since individuals' knowledge or expe-
rience using IT is related to their receptivity toward a change, their
knowledge of IT and receptiveness to information about IT may alleviate
a negative attitude toward new technologies or innovations [94].Thus,
higher IT knowledge, which provides a familiarity with current technol-
ogies, helps individuals evaluate the benets of adopting new technolo-
gies [95]. Individuals who areeducated about the benets of IT would be
more willing to adopt [15]. Therefore, individuals' related knowledge in-
uences theperceived personal benets of IT, which in turnleads to plat-
form adoption. We, therefore, hypothesize
H4b. Related knowledge is positively associated with perceived personal
Perceived enjoyment refers to the extent to which the activity of using
the computer is perceived to be enjoyable in its own right, apart from any
performance consequences that may be anticipated([47] p. 1113). Indi-
viduals' perceived enjoyment of the interaction with the technology has
been conrmed as a signicant antecedent in adopting and using technol-
ogy [84]. Individuals who feel pleasure when engaging in a particular be-
havior are likely to adopt the technology or innovation [85,96]. In hedonic
information systems, perceived enjoyment has been regarded as a salient
affective and intrinsic factor that leads to adoption [88,89]. Researchers
have found that perceived enjoyment has direct positive inuence on
the IT adoption [89,96]. Therefore, we hypothesize:
H4c. Perceived enjoyment is positively associated with platform adoption.
Personal innovativeness refers to an individual trait that reects an in-
trinsic willingness to try out new technologies [51]; it has been exten-
sively examined as an IT-specic trait triggering intrinsic motivation
[97], and is considered an antecedent or a component of perceived enjoy-
ment [98]. Innovations or new technologies are inherently risky even
though individuals recognize the potential benets from adoption [99].
Personal innovativeness enables individuals to be more willing to adopt
the innovations or technologies in the face of uncertainty about benets
[79].Morespecically, personal innovativeness is an individual's affective
trait that determines their affective states such as perceived enjoyment
and computer anxiety [100]. Furthermore, perceived (heightened) enjoy-
ment can be a consequence of personal innovativeness in the domain of IT
H4d. Personal innovativeness is positively associated with perceived
3.2.2. Scale development
We developed items for the constructs by applying the operational
denitions that are in line with the ndings from Study 1. To generate
a list of potential survey items, we reviewed existing literature. Where
constructs exist that are similar to the ones that we have identied
through our interviews, we evaluated the use of existing items.
Perceived relative advantage(RA) is measured by asking respondents
to indicate whether theplatform enables them to accomplish their task
quickly, could improve their quality of work, and can enhance their ef-
fectiveness on theirjob [32]. Perceived platform innovativeness (PI)mea-
sures whether the adopted platform is highly innovative compared to
other platforms, whether the platform is frequently updated with new
functions, whether it is supplemented with new functions, and whether
new functions add value to the platform [101].Wemeasuredtechnical
compatibility (TC) by asking respondents to indicate whether the plat-
form is compatible with other programming technologies, whether it
is compatible with other software, and whether the platform ts the
way respondents work [16]. We measured platform openness (PO)as
the degree of perceived openness by application developers [102]. All
constructs were measured with items using a seven-point Likert scale
ranging from strongly disagree to strongly agree (see Appendix B).
Perceived market potential (MP) is measured using four items [103]
by asking respondents to indicate whether applications developed on
this platform have many potential customers, whether customers
have a large number of potential needs, whether they perceive the po-
tential size of the market as being large, and whether they perceive
the market as having the potential to grow very quickly. Perceived mar-
ketability (MKT) was gauged by asking emphasizing the current rather
than the potential size of the market [104]. We operationalized per-
ceived developer tools (DT) as addressing the importance of having a
platform for development, using a platform for a low price, and using
platform when they need it [16].
Perceived personal benets (PERB) were measured by asking respon-
dents' to indicate the level of expected monetary rewards, extrinsic ben-
ets, and personal benets [105].Enjoyment (ENJ) is measured by
asking respondents about their level of enjoyment, fun, pleasantness,
and liking when developing applications using the platform [8].Finally,
related knowledge (RK) is measured by assessing respondents' level of
prior knowledge in using the platform they adopted [106].We
operationalize perceived personal innovativeness (PERI) by assessing re-
spondents' eagerness to experiment with new IT, eagerness to try out
new IT, and hesitancy to explore new information [48].
Social inuence (SI) is measuredusing fouritems asking respondents
about the inuence they received from their peers [5].
The dependent construct, platform adoption (ADOPT)isoperational-
ized using three items measuring respondents' level of favorable atti-
tude toward the platform they adopted [107].
3.2.3. Data collection
A pilot study was conducted with 10 mobile application developers
who had not participated in Study 1. Participants in the pilot study were
recruited from online communities in which professional developers
share their knowledge and experiences developing applications. After
modifying the instrument based on feedback from the pilot study, we
sampled from the mobile application development community in
South Korea. We identied 52 companies currently developing mobile
applications. We contacted each company to verify whether developers
were free to choose platforms. Companies that force developers to
choose specic platforms (6 companies) were excluded. Of the 46
companies allowing developer choice, 38 allowed us to distribute
our survey. All 38 companies were small companies with fewer
than 30 employees and developing various applications such as
games and social networking apps. We contacted 173 individuals
from those companies. After verifying that they were actual devel-
opers, we invited them to participate in our study and sent the sur-
vey instrument by email. Participants returned the survey by email
within 7 days. We collected 159 responses and used them for analy-
sis (91.9% response rate).
6J. Song et al. / Decision Support Systems xxx (2018) xxxxxx
Please cite this article as: J. Song, et al., Platform adoption by mobile application developers: A multimethodological approach, Decision Support
Systems (2018),
3.2.4. Data analysis
The average age of the participants was approximately 31. Average
job experience developing mobile applications was 21.8 months. The
average number of applications developed was 4.2. Most developers
use Java to develop applications (Java = 84, iOS = 31, Others = 28,
multiple = 16).
Reliability was assessed usinginternal consistency scores,calculated
by composite reliability scores. Internal consistencies of all variables are
considered acceptable if they exceed 0.70 (Nunally, [115]). As shown in
Table 2, the values of Cronbach's alpha for all constructs are greater than
or equal to 0.80. All items exhibit high loadings on their respective con-
structs. Thus, all constructs in the model exhibit good internal consis-
tency. We assessed discriminant validity between the constructs using
the square root of AVE. Table 2 indicates that the AVE for each construct
exceeded the correlations between that and all other constructs [108].
Comparing inter-construct correlations with AVEs in Table 2 reveals
that all constructs share more variance with their indicators than with
other constructs.
We used SmartPLS for data analysis, conducting our analysis in two
stages. First, we tested the measurement model to ensure that the con-
structs had sufcient psychometric validity, then we assessed the struc-
tural modelin which the hypotheseswere tested. To ensure thestability
of the model developed to test the research hypotheses, we used the PLS
bootstrap resampling procedure with an iteration of 1000 sub-samples
drawn with replacement from the initial sample to generate percentile
bootstrap p values. The summary of factor analysis for testing the valid-
ity of our constructs results is shown in Table 3.
Conrmatory factor analysis (CFA) was performed to test reliability
and construct validity for all items. As shown in Table 3, all factor load-
ings are signicant and exceeded 0.70. Composite reliabilities for each
construct exceeded 0.80, and average variance extracted for each con-
struct exceeds 0.50, meeting commonly established reliability thresh-
olds [108].
There is a potential for common method bias when using self-
reported data [109]. Common method bias is variance that is attribut-
able to the measurement method rather than to the constructs the mea-
sures represent([109] p. 879). In order to test for this bias, we
performed a Harman's one-factor test in which all items were entered
into an un-rotated exploratory factor analysis. This analysis determines
whether a single factor emerges or accounts for the majority of the var-
iance. The results showed 38 factors emerged, the largest accounted for
31% of the variance. In addition, we followed the unmeasured latent
method construct (ULMC) approach [110] shown in Appendix C.In
this approach, if the method factor loadings are insignicant and the
indicators' substantive variances are substantially greater than their
method variances, we can conclude that common method bias is un-
likely to be a serious concern([110] p. 87). Appendix C demonstrates
the results of common method bias analysis including constructs, the
substantive factor loadings, and the method factor loadings. The results
reveal that only 24 (out of 38) of themethod factor loadings were statis-
tically signicant (p b0.01). The results also demonstrate that the vari-
ances of the indicators (average of 0.818) are substantially greater than
their method-based variances (average of 0.003). Given the results, we
assert that common method bias is not a concern for this study.
In a PLS structural model, paths are interpreted as standardized beta
weights in a regression analysis. The path coefcients present the esti-
mates obtained from PLS analysis (Fig. 2). For further analysis, a boot-
strapping procedure generating 1000 random samples of size of 159
was used to estimate the signicance of the coefcients and the weights
of the dimensions of constructs. All hypotheses were supported with
the exception of H4a and H4c. The model explains 20.2% of the variance
in relative advantage, 15.6% in technical compatibility, 37.3% in market-
ability, 7.1% in personal benet, 11.8% in enjoyment, and 34.4% in plat-
form adoption.
4. Discussion
The results demonstrate that IT platform adoption is inuenced by
developers' perceived platform characteristics, individual characteris-
tics, network externalities, and social inuence. First, among the con-
structs related to platform characteristics, developers will adopt
platforms if they feel that the platform has high technical compatibility
and can help them improve their work quality and effectiveness (rela-
tive advantage). To improve compatibility and enhance its relative ad-
vantage, platform owners may need to keep developing new
functionality and provide developers with straightforward instructions
on how to develop applications in the platforms. Second, regarding
the constructs related to individual characteristics, application devel-
opers seem unconcerned with personal benets and perceived enjoy-
ment. These ndings are perhaps best explained by the reality that we
survey professional company-employed developers; assessment of the
motivations of freelance developers could yield different results, a
topic to which we will return in theupcomingSection 4.4 on limitations
and future research. Third, because application developers are not sim-
ply end-user consumers but rather IT producers, developers focus on
potential future prots. Therefore, factors that are positively associated
with the marketability of the platform are positively associated with
adoption, such as market potential and developer tools. Finally, devel-
opers' adoption decisions may also be impacted by social inuence. De-
velopers exchange information with peers and colleagues, which may
lead them to select a particular platform. Following from these results,
we now consider broader implications of our study.
4.1. Theoretical implications
Our review of prior literature suggests that despite much progress in
IT adoption literature, research opportunities still remain. Platform
adoption differs from IT adoption since developers adopt platforms for
Table 2
Reliability and construct correlations
Constructs Alpha 1 2 3 45678910111213
1. Adoption 0.89 0.90
2. Developer tools (DT) 0.90 0.13 0.91
3. Enjoyment (ENJ) 0.95 0.30 0.39 0.95
4. Marketability (MKT) 0.88 0.31 0.42 0.42 0.90
5. Market potential (MP) 0.88 0.30 0.42 0.39 0.58 0.90
6. Personal benets (PERB) 0.88 0.15 0.34 0.42 0.47 0.44 0.90
7. Platform openness (PO) 0.75 0.28 0.27 0.36 0.39 0.33 0.27 0.89
8. Personal innovativeness (PERI) 0.87 0.20 0.05 0.34 0.18 0.21 0.15 0.28 0.85
9. Platform innovativeness (PI) 0.89 0.52 0.27 0.36 0.40 0.39 0.30 0.40 0.17 0.86
10. Relative advantage (RA) 0.87 0.46 0.23 0.41 0.29 0.28 0.22 0.36 0.20 0.56 0.85
11. Related knowledge (RK) 0.92 0.29 0.31 0.49 0.36 0.34 0.22 0.40 0.58 0.37 0.44 0.90
12. Social inuence (SI) 0.83 0.41 0.17 0.40 0.37 0.26 0.40 0.12 0.42 0.32 0.30 0.33 0.86
13. Technical compatibility (TC) 0.91 0.43 0.32 0.47 0.30 0.31 0.26 0.40 0.24 0.52 0.62 0.48 0.28 0.87
Note. The diagonal elements (in bold) represent the square root of the AVE.
7J. Song et al. / Decision Support Systems xxx (2018) xxxxxx
Please cite this article as: J. Song, et al., Platform adoption by mobile application developers: A multimethodological approach, Decision Support
Systems (2018),
IT product development and need to meet end-users' expectations. Our
work is related to but distinct from prior research investigatingtechnol-
ogy adoption because we analyze the adoption behaviors from an IT
platform developer's perspective by noting several key factors that
differ fromprior ndings. Our ndings do, however, share some similar-
ities with foregoing research on ITadoption. Specically, our study pro-
vides evidence that perceived technology characteristics, individual
characteristics, and social inuence all play a role in the adoption of IT
Table 3
Item loadings and cross loadings.
Adoption ADOPT1 0.90 0.07 0.23 0.24 0.24 0.10 0.25 0.15 0.38 0.39 0.20 0.37 0.35
ADOPT2 0.92 0.09 0.25 0.29 0.25 0.14 0.23 0.13 0.49 0.38 0.24 0.37 0.36
ADOPT3 0.89 0.19 0.33 0.32 0.32 0.17 0.29 0.24 0.49 0.37 0.33 0.38 0.46
Developer tools (DT) DT1 0.17 0.90 0.41 0.43 0.44 0.35 0.29 0.09 0.25 0.19 0.37 0.18 0.39
DT2 0.07 0.93 0.32 0.35 0.37 0.31 0.23 0.04 0.16 0.17 0.26 0.15 0.23
DT3 0.10 0.91 0.33 0.37 0.33 0.24 0.20 0.07 0.18 0.17 0.18 0.12 0.24
Enjoyment (ENJ) EJ1 0.30 0.37 0.94 0.43 0.41 0.39 0.36 0.28 0.28 0.31 0.43 0.34 0.46
EJ2 0.28 0.37 0.96 0.39 0.34 0.40 0.33 0.32 0.29 0.31 0.40 0.38 0.40
EJ3 0.28 0.37 0.94 0.38 0.37 0.41 0.33 0.33 0.36 0.36 0.41 0.40 0.42
Marketability (MKT) MKT1 0.27 0.44 0.35 0.88 0.47 0.36 0.31 0.14 0.21 0.17 0.36 0.29 0.22
MKT2 0.30 0.37 0.38 0.91 0.52 0.45 0.35 0.12 0.33 0.15 0.38 0.27 0.23
MKT3 0.28 0.34 0.40 0.90 0.57 0.44 0.39 0.14 0.34 0.27 0.27 0.35 0.28
Market potential (MP) MP1 0.25 0.34 0.30 0.50 0.89 0.35 0.25 0.16 0.29 0.16 0.28 0.20 0.18
MP2 0.26 0.43 0.41 0.53 0.92 0.41 0.30 0.21 0.32 0.17 0.33 0.23 0.25
MP3 0.29 0.37 0.34 0.53 0.89 0.42 0.34 0.18 0.37 0.27 0.33 0.24 0.34
Personal benefits (PERB) PB1 0.13 0.27 0.34 0.50 0.43 0.90 0.22 0.17 0.23 0.14 0.28 0.30 0.24
PB2 0.17 0.35 0.40 0.42 0.44 0.94 0.31 0.18 0.26 0.16 0.25 0.37 0.26
PB3 0.10 0.28 0.39 0.34 0.31 0.86 0.20 0.12 0.31 0.10 0.14 0.34 0.17
Platform openness (PO) PO1 0.30 0.23 0.38 0.38 0.29 0.28 0.95 0.32 0.34 0.32 0.34 0.13 0.41
PO2 0.18 0.26 0.22 0.32 0.31 0.18 0.82 0.17 0.30 0.18 0.32 0.08 0.27
Personal innovativeness
PERI1 0.26 0.01 0.19 0.16 0.17 0.16 0.20 0.86 0.16 0.05 0.46 0.42 0.12
PERI2 0.17 0.04 0.38 0.15 0.21 0.15 0.29 0.90 0.21 0.22 0.52 0.42 0.34
PERI3 0.08 0.04 0.30 0.09 0.16 0.15 0.29 0.91 0.13 0.18 0.48 0.29 0.23
Platform innovativeness
PI1 0.47 0.15 0.26 0.24 0.32 0.35 0.28 0.16 0.88 0.36 0.25 0.29 0.36
PI2 0.41 0.23 0.31 0.33 0.35 0.20 0.35 0.17 0.90 0.45 0.37 0.27 0.41
PI3 0.49 0.21 0.32 0.32 0.32 0.28 0.33 0.17 0.94 0.39 0.29 0.33 0.39
Relative advantage (RA) RA1 0.38 0.26 0.33 0.18 0.22 0.14 0.22 0.16 0.44 0.89 0.35 0.20 0.59
RA2 0.42 0.12 0.30 0.20 0.20 0.13 0.29 0.15 0.42 0.91 0.32 0.21 0.49
RA3 0.31 0.13 0.28 0.21 0.16 0.12 0.30 0.14 0.34 0.87 0.31 0.27 0.48
Related knowledge (RK) RK1 0.19 0.27 0.42 0.32 0.34 0.26 0.31 0.49 0.29 0.29 0.92 0.29 0.40
RK2 0.28 0.32 0.45 0.37 0.34 0.19 0.35 0.48 0.36 0.42 0.94 0.33 0.43
RK3 0.31 0.26 0.35 0.34 0.30 0.23 0.36 0.54 0.30 0.33 0.91 0.29 0.41
Social influence (SI) SI1 0.28 0.08 0.35 0.30 0.23 0.24 0.05 0.29 0.19 0.18 0.25 0.84 0.18
SI2 0.43 0.13 0.31 0.32 0.24 0.40 0.13 0.34 0.33 0.25 0.28 0.90 0.23
SI3 0.34 0.22 0.37 0.25 0.17 0.31 0.13 0.47 0.30 0.20 0.31 0.84 0.23
Technical compatibility (TC) TC1 0.37 0.27 0.41 0.25 0.27 0.17 0.33 0.23 0.37 0.58 0.48 0.21 0.92
TC2 0.44 0.32 0.45 0.28 0.31 0.26 0.40 0.27 0.36 0.53 0.42 0.24 0.93
TC3 0.38 0.30 0.37 0.22 0.21 0.24 0.35 0.22 0.45 0.49 0.32 0.24 0.87
8J. Song et al. / Decision Support Systems xxx (2018) xxxxxx
Please cite this article as: J. Song, et al., Platform adoption by mobile application developers: A multimethodological approach, Decision Support
Systems (2018),
platforms. This conrms much foregoing work on IT adoption, and con-
rms the importance of these constructs in the study of platform adop-
tion as well.
The positive and signicant relationship from perceived network ex-
ternalities to platform adoption, however, is an important distinction
between IT adoption research and our platform adoption research.
Previous literature has called for investigation on externalities in
platforms due to the rapid growth of the mobile application market
[111]. Our paper is among the rst to highlight the importance of
network externalities when individuals make an adoption decision.
By observing a link between network externalities and adoption,
our paper identies that some of the factors that are considered rel-
evant in diffusion of innovations research apply to the adoption of
An additional difference between our work on platform adoption
and foregoing work on IT adoption is that we did not nd a signi-
cant link from enjoyment to platform adoption. Perceived e njoyment
is an intrinsic motivation construct has been noted in TAM2 [31].
Given that developers are adopting a platform with the intent of de-
veloping an application commercially, rather than an end-user
adopting a technology with the simple intent of using it, perceived
enjoyment may be less important. They may consider specictech-
nology characteristics, but we surmise that they nd there are
more utilitarian rather than hedonic motivations involved here. Plat-
form adopters are primarily prot-seeking developers. They are
making a business decision; they are not simply considering their
personal preferences. For these reasons, the relationship between
this construct and platform adoption is weakened.
In sum, this study has served to provide an initial understanding
of the factors inuencing platform adoption. While most adoption
studies (from an end-user perspective) have emphasized the roles
of ease of use and usefulness-related factors, our theoretical devel-
opment emphasizes the characteristics of the technology itself and
its network effects. In adoption studies from the end-user perspec-
tive, the role of technology characteristics such as those studied
here has not been identied. Nevertheless, such characteristics are
understandably important to developers because they shape their
perceptions about their ability to successfully utilize a particular
platform. We also found the importance of network externalities by
considering perspectives and rationales from other studies. Market
potential indicates whether developers can expect nancial, reputa-
tional, or other benets with a given platform.
4.2. Methodological implications
Mixed methods approaches help researchers understand novel and
complex phenomena [38]. This paper demonstrates the use of sequen-
tial mixed methods, allowing a comprehensive initial explanation of a
phenomenon to be compared to existing theory. In Study 1's interpre-
tive approach, not only were well-researched factors (technology
characteristics, individual characteristics, and social interactions) re-
vealed to be inuential in platform adoption decisions, but a largely
uninvestigated factor in individual adoption, perceived network exter-
nalities, was also identied. The demonstration of a mixed-methods ap-
proach is a secondary contribution of this study. By utilizing a mixed
methods approach across two studies, we are able to explore key factors
that inuence developers' adoption, even though IT adoption has been
extensively investigated. This multi-methodological approach could be
similarly applied to examine areas within other well-studied research
4.3. Practical implications
The practical implications of our study are primarily for those who
own, manage, or otherwise oversee the various platforms that devel-
opers choose to adopt. Clearly, an understanding of developers' wants
can help platform mangers attract a critical mass of developers to en-
sure the viability of their platform. One need look no further than the
fall of BlackBerry to see the importance of this. A paucity of app devel-
opers limited BlackBerry's revenue and growth, and was one of the rea-
sons why it was unable to keep pace with competitors. The needs of the
developer community were perhapsequal in importance to the creation
of innovative devices [112]. Platform owners should aggressively mar-
ket information about their user base to developers and focus intently
on creating tools to support developers' work.
Additionally, platform features that enable an online developer com-
munity could be an important channel that benets both developers
and platform owners. Online community forums enable developers to
interact with one another, enhancing their professional networks and
enabling their information gathering. Platform owners can also reach
out to developers via the community to understand their needs.
Attracting and supporting developers will enable the creation of more
and better applications, which should attract more end-users, ulti-
mately generating more revenue for developers and platform owners.
Fig. 2. Research model results.
9J. Song et al. / Decision Support Systems xxx (2018) xxxxxx
Please cite this article as: J. Song, et al., Platform adoption by mobile application developers: A multimethodological approach, Decision Support
Systems (2018),
Furthermore, we observe that developers' evaluation of platforms
resembles B2B e-commerce models in some important ways. Devel-
opers play the role of sellers and make platform adoption decisions by
considering factors that inuence the salesof their products (which
are mobile applications) to buyers(which are app stores). While the
roles of end-user/buyersand platform/sellersare straightforward
and well-understood, the roles of developer/sellersand platform/
buyersare less thoroughly explored. Understanding the key charac-
teristics considered by developer/sellers will help platform and app
store owners improve platform features and attract greater numbers
of developers.
4.4. Limitations and future research
One limitation of this study is that we only consider developers
who develop for a single platform. In future research, it would be
possible to study developers who choose to simultaneously develop
for multiple platforms, a phenomenon known as multi-homing.
While multi-homing was not seen in our initial interviews, and was
only seen to a very limited degree in our surveys, some developers
consider multiple platforms [113,114]. Multiple-platform adoption
will likely be driven by factors similar to the ones we have identied
in this study. In addition, when apps can be easily ported to another
platform, we surmise that this platform characteristic will facilitate
An additional limitation is that we did not examine differencesin the
drivers of platform adoption between professional developers, part-
time commercial developers, and hobbyists or open-source idealists
who have no expectation of remuneration. Comparisons between
these groups could yield interesting insights and reveal the boundaries
of generalizability for this study. Yet another focus of investigation in fu-
ture research would be developers who are end-users in multiple eco-
systems, such as those who use an iPhone as well as an Android
device, but choose to develop within one ecosystem, but not both.
Finally, an additional limitation is the determination that was made
to limit thecategories shown inthe model from Study 1 to those catego-
ries that were mentioned by 5 or more developers. There is no direct
statistical test to identify the correct number of categories and the deci-
sion to include only those mentioned by 5 or more developers repre-
sents an attempt to strike a balance between comprehensiveness and
parsimony. Future studies may opt to include either more or less vari-
ables to glean other insights.
5. Conclusion
This research yields insights for developers, application stores,
and consumers. For developers, a well-chosen platform provides
considerable competitive benets, while a poorly-chosen one puts
them at a disadvantage. When a critical mass of developers begins
to utilize a particular platform, certain ecosystems become domi-
nant. The rise and fall of rms like Palm, BlackBerry, Apple, and Goo-
gle, and their associated hardware manufacturers, can be better
understood by examining how many developers choose to develop
apps for each platform. And for consumers, the choices that they
have in the market for mobile technologies are determined, at least
in part, by the size of the developer group for a given platform. If re-
searchers can understand the reasons why certain platforms are
adopted (or not), they may be able to identify emerging trends in
IT at an early stage. Because of the importance of platform adoption
for developers, for consumers, and for platform creators, we con-
tinue to seek to develop a theory of platform adoption. A well-
developed theory will help the stakeholders in each ecosystem bet-
ter understand how to compete in this dynamic marketspace.
Appendix A. Study 1 interview protocol
Appendix B. Constructs and measures
Table B.1
Reliability and validity measures.
Item Mean Standard
Weight Loadings Instrument items
Relative advantage (RA)
RA1 5.25 1.15 0.40 0.89 Using the platform that I have chosen enables me to accomplish tasks more quickly
RA2 5.17 1.15 0.41 0.91 Using the platform that I have chosen improves the quality of the work I do
RA3 5.14 1.12 0.32 0.87 Using the platform that I have chosen enhances my effectiveness on the job.
Platform innovativeness (PI)
PI1 4.52 1.15 0.33 0.88 The platform that I have chosen is highly innovative compared to other platform
PI2 4.78 1.03 0.41 0.90 The platform that I have chosen is frequently updated with new functions that provide new alternatives for
PI3 4.52 1.16 0.36 0.94 The platform that I have chosen frequently incorporates new functions that are valuable to the platforms.
Technical compatibility (TC)
TC1 5.41 1.12 0.34 0.92 The platform that I have chosen is compatible with the other programing technologies that I use.
1. Do you use Apple iPhone or Google Android for application development? When did you rst start using this technology platform?
2. What reasons motivated you to choose this technology platform?
3. Was your decision inuenced by technology reviews in the popular press or the internet, the choices of your fellow developers, experts' sugges-
tions, and/or others?
4. Did you consider any alternative technology platform? If so, which ones? Why did you not choose those platforms?
5. What problems did you encounter with your choice of technology platform compared to alternative platforms?
6. To what extent did your prior IT knowhow and knowledge required to use your technology platform inuence your platform choice?
7. To what extent did your expectations of the future popularity of your technology platform and availability of future applications on this
platform inuence your platform choice?
8. Are you satised with your choice of technology platform? Why or why not?
9. How do you foresee your future use of this platform? Do you plan to try or switch to alternative platforms?
10. Tell me anything else if you have other comments.
10 J. Song et al. / Decision Support Systems xxx (2018) xxxxxx
Please cite this article as: J. Song, et al., Platform adoption by mobile application developers: A multimethodological approach, Decision Support
Systems (2018),
Table B.1 (continued)
Item Mean Standard
Weight Loadings Instrument items
TC2 5.45 1.08 0.40 0.93 For me, the platform that I have chosen is compatible with the other software that I use.
TC3 5.55 1.17 0.35 0.87 I think that using the platform that I have chosen ts well with the way that I like to work.
Platform openness (PO)
PO1 4.67 1.16 0.67 0.95 Developers are given clear requirements for using the marketplace.
PO2 4.76 1.23 0.44 0.82 The processes of uploading and downloading applications to the marketplace are straightforward.
Market potential (MP)
MP1 5.19 1.09 0.35 0.89 There are many potential customers for applications developed on this platform.
MP2 5.07 1.02 0.38 0.92 Potential customers have a great need for applications on this platform.
MP3 5.19 1.13 0.38 0.89 The potential dollar size of the market for applications developed on this platform is very large.
Marketability (MKT)
MKT1 5.48 1.16 0.36 0.88 Currently, There are many customers for applications developed on this platform.
MKT2 5.09 1.17 0.37 0.91 Currently, customers have a great need for applications on this platform.
MKT3 5.36 1.11 0.38 0.90 The current dollar size of the market for applications developed on this platform is very large.
Developer tools (DT)
DT1 5.62 1.11 0.41 0.90 For me, having a platform for development is important
DT2 5.55 1.07 0.34 0.93 For me, being able to use a platform for a low price is important
DT3 5.65 1.09 0.35 0.91 For me, being able to use a platform when I need it is important.
Personal benets (PERB)
PERB1 4.75 1.48 0.44 0.90 I will receive monetary rewards in return for application development by using the platform.
PERB2 4.89 1.23 0.43 0.94 I will receive extrinsic benet for application development by using the platform.
PERB3 5.09 1.27 0.23 0.86 The extrinsic rewards in return for application development using the platform are important.
Enjoyment (ENJ)
ENJ1 5.06 1.11 0.34 0.94 I nd that using the platform to develop applications is enjoyable .
ENJ2 5.08 1.14 0.35 0.96 The actual process of using the platform to develop applications is pleasant.
ENJ3 4.97 1.14 0.36 0.94 I have fun using the platform for developing application.
Related knowledge (RK)
RK1 5.03 1.16 0.43 0.92 I have a good technical knowledge concerning the platform for developing application.
RK2 5.08 1.05 0.29 0.94 I have a good knowledge of the materials that are required in order to use the platform.
RK3 5.21 1.13 0.36 0.91 I have a good knowledge that helps me build my intended applications.
Personal innovativeness (PERI)
PERI1 4.75 1.09 0.24 0.86 If I hear about a new information technology, I look for ways to experiment with it.
PERI2 4.81 1.10 0.49 0.90 Among my peers, I am usually the rst to try out new information technologies.
PERI3 4.86 1.06 0.38 0.91 I am hesitant to try out new information technologies. (reverse scale)
Social inuence (SI)
SI1 4.91 0.95 0.31 0.84 People who inuence my behavior think that the platform that I have chosen is better than other platform.
SI2 4.88 1.02 0.47 0.90 People who are important to me think that the platform that I have chosen is good for application development.
SI3 4.81 0.98 0.37 0.84 I think that others' inuences are important to me in selecting a platform for developing applications.
Platform adoption (ADOPT)
ADOPT1 4.74 1.25 0.36 0.90 I seldom consider switching to another platform.
ADOPT2 4.48 1.24 0.36 0.92 When I need to develop applications, this platform is my rst choice.
ADOPT3 4.87 1.22 0.39 0.89 To me, the platform that I have chosen is the best platform on which to develop applications.
Appendix C. Supplemental analysis
Table C.1
Testing for common method bias.
Constructs Items Substantive factor loading (R1) R1
Method factor loading (R2) R2
Adoption Adop1 0.902⁎⁎⁎ 0.814 0.084⁎⁎⁎ 0.007
Adop2 0.920⁎⁎⁎ 0.846 0.033 0.001
Adop3 0.884⁎⁎⁎ 0.781 0.120⁎⁎⁎ 0.014
Development tools (DT) DT1 0.881⁎⁎⁎ 0.776 0.144⁎⁎⁎ 0.021
DT2 0.936⁎⁎⁎ 0.876 0.052⁎⁎⁎ 0.003
DT3 0.922⁎⁎⁎ 0.850 0.087⁎⁎⁎ 0.008
Enjoyment (ENJ) ENJ1 0.947⁎⁎⁎ 0.897 0.015 0.000
ENJ2 0.963⁎⁎⁎ 0.927 0.048⁎⁎⁎ 0.002
ENJ3 0.940⁎⁎⁎ 0.884 0.033⁎⁎ 0.001
Marketability (MKT) MKT1 0.894⁎⁎⁎ 0.799 0.054⁎⁎⁎ 0.003
MKT2 0.913⁎⁎⁎ 0.834 0.006 0.000
MKT3 0.889⁎⁎⁎ 0.790 0.060⁎⁎⁎ 0.004
Market potential (MP) MP1 0.894⁎⁎⁎ 0.799 0.094⁎⁎⁎ 0.009
MP2 0.923⁎⁎⁎ 0.852 0.010 0.000
MP3 0.880⁎⁎⁎ 0.774 0.084⁎⁎⁎ 0.007
Personal benets (PERB) PB1 0.907⁎⁎⁎ 0.823 0.007 0.000
PB2 0.940⁎⁎⁎ 0.884 0.034⁎⁎ 0.001
(continued on next page)
11J. Song et al. / Decision Support Systems xxx (2018) xxxxxx
Please cite this article as: J. Song, et al., Platform adoption by mobile application developers: A multimethodological approach, Decision Support
Systems (2018),
Table C.1 (continued)
Constructs Items Substantive factor loading (R1) R1
Method factor loading (R2) R2
PB3 0.848⁎⁎⁎ 0.719 0.045⁎⁎ 0.002
Personal innovativeness (PERI) PERI1 0.861⁎⁎⁎ 0.741 0.030 0.001
PERI2 0.901⁎⁎⁎ 0.812 0.077⁎⁎⁎ 0.006
PERI3 0.915⁎⁎⁎ 0.837 0.049⁎⁎⁎ 0.002
Platform innovativeness (PI) PI1 0.893⁎⁎⁎ 0.797 0.044⁎⁎ 0.002
PI2 0.884⁎⁎⁎ 0.781 0.052⁎⁎ 0.003
PI3 0.941⁎⁎⁎ 0.885 0.008 0.000
Platform openness (PO) PO1 0.898⁎⁎⁎ 0.806 0.087⁎⁎⁎ 0.008
PO2 0.892⁎⁎⁎ 0.796 0.089⁎⁎⁎ 0.008
Relative advantage (RA) RA1 0.878⁎⁎⁎ 0.771 0.058⁎⁎ 0.003
RA2 0.910⁎⁎⁎ 0.828 0.020 0.000
RA3 0.880⁎⁎⁎ 0.774 0.039⁎⁎⁎ 0.002
Related knowledge (RK) RK1 0.920⁎⁎⁎ 0.846 0.039⁎⁎⁎ 0.002
RK2 0.949⁎⁎⁎ 0.901 0.037⁎⁎⁎ 0.001
RK3 0.905⁎⁎⁎ 0.819 0.001 0.000
Social inuence (SI) SI1 0.862⁎⁎⁎ 0.743 0.080⁎⁎⁎ 0.006
SI2 0.882⁎⁎⁎ 0.778 0.038 0.001
SI3 0.846⁎⁎⁎ 0.716 0.041 0.002
Technical compatibility (TC) TC1 0.917⁎⁎⁎ 0.841 0.012 0.000
TC2 0.935⁎⁎⁎ 0.874 0.035 0.001
TC3 0.897⁎⁎⁎ 0.805 0.024 0.001
Average 0.904 0.818 0.000 0.003
⁎⁎ pb0.05.
⁎⁎⁎ pb0.01.
[1] Apple, Two Million U.S. Jobs and Counting,
2017, Accessed date: 6 November 2017.
[2] M. Mandel, App Economy Jobs in Europe (Part 1), Institute, P.P., Washington, DC,
[3] Statista, Cumulative number of apps downloaded from the Apple App Store from
July 2008 to September 2016 (in billions), 2016 https://www.
(retrieved March 7, 2017).
[4] F.D. Davis, Perceived usefulness, perceived ease of use, and user acceptance of in-
formation technology, MIS Quarterly 13 (1989) 319339.
[5] V. Venkatesh, M.G. Morris, G.B. Davis, F.D. Davis, User acceptance of information
technology: toward a unied view, MIS Quarterly 27 (2003) 425478.
[6] E.M. Rogers, Diffusion of Innovations, 4th ed. The Free Press, New York, 1995.
[7] A. Bandura, Social Foundations of Th ought and Action: A Social Cogniti ve Ap-
proach, 1986.
[8] D.R. Compeau, C.A. Higgins, Computer self-efcacy: development of a measureand
initial test, MIS Quarterly (1995) 189211.
[9] C.Y. Baldwin, C.J.Woodard, The architecture of platforms:a unied view, Platforms,
Markets and Innovation 2009, pp. 1944.
[10] A. Tiwana, B. Konsynski,A.A. Bush, Research commentaryplatform evolution: co-
evolution of platform architecture, governance, and environmental dynamics, In-
formation Systems Research 21 (2010) 675687.
[11] R.G. Fichman, Real options and IT platform adoption: implications for theory and
practice, Information Systems Research 15 (2004) 132154.
[12] M.H. Meyer, A.P. Lehnerd, The Power of Product Platforms, Simon and Schuster,
[13] A. Gawer, M.A. Cusumano, Platform Leadership, Harvard Business School Press,
Boston, 2002.
[14] M.L. Katz, C. Shapiro, Systems competition and network effects, Journal of Eco-
nomic Perspectives 8 (1994) 93115.
[15] J.Y.L. Thong, An integrated model of information systems adoption in small busi-
nesses, Journal of Management Information Systems 15 (1999) 187214.
[16] S. Taylor, P.A. Todd, Understanding information technology usage: a test of com-
peting models, Information Systems Research 6 (1995) 144176.
[17] Q. Hu, C. Saunders, M. Gebelt, Diffusion of information systems outsourcing: a re-
evaluation of inuence sources, Information Systems Research 8 (1997) 288301.
[18] S.-J. Hong, K.Y. Tam, Understanding the adoption of multipurpose information ap-
pliances: the case of mobile data services, Information Systems Research 17 (2006)
[19] J. Dedrick, J. West, An exploratory study into open source platform adoption, Pro-
ceedings of the 37th Annual Hawaii International Conference on System Sciences,
2004, IEEE, 2004 (10 pp.).
[20] C.Z. Liu, Y.A. Au, H.S. Choi, Effects of freemium strategy in the mobile app market:
an empirical study of Google Play, Journal of Management Information Systems 31
(2014) 326354.
[21] D. Streitfeld, As Boom Lures App Creators, Tough Part is Making a Living, The New
York Times, New York, New York, 2012 A1.
[22] B. Kogut,A. Metiu, Open-source software development and distributed innovation,
Oxford Review of Economic Policy 17 (2001) 248264.
[23] T. Barnett, A.W. Pearson, R. Pearson, F.W.Kellermanns, Five-factor model personal-
ity traits as predictors of perceived and actual usage of technology, European Jour-
nal of Information Systems 24 (2015) 374390.
[24] I. Junglas, L. Goe l, C. Abraham, B. Ives, The social co mponent of information
systems-how sociability contributes to technology acceptance, Journal of the Asso-
ciation for Information Systems 14 (2013) 585.
[25] R. Agarwal, J. Prasad, A eld study of the adoption of software process innovations
by information systems professionals, IEEE Transactions on Engineering Manage-
ment 47 (2000) 295308.
[26] S. Kalish, A new product adoption model with price, advertising, and uncertainty,
Management Science 31 (1985) 15691586.
[27] I. Ajzen, The theory of planned behavior, Organizational Behavior and Human De-
cision Processes 50 (1 991) 179211.
[28] F.D. Davis, R.P. Bagozzi, P.R. Warshaw, User acceptance of computer technology: a
comparison of two theoretical models, Management Science 35 (1989) 9821003.
[29] M. Fishbein, I. Ajzen, Belief, Attitude, Intention, and Behavior: An Introduction to
Theory and Research, Addison-Wesley Publishing, Reading, MA, 1975.
[30] V. Venkatesh, H. Bala, Technology acceptance model 3 and a researchagenda on in-
terventions, Decision Sciences 39 (2008) 273315.
[31] V. Venkatesh, F.D. Davis, A theoret ical extension of the technology acceptance
model: four longitudinal eld studies, Management Science 46 (2000) 186204.
[32] G.C. Moore, I. Benbasat, Development of an instrument to measure the perceptions
of adopting an information technology innovation, Information Systems Research
[33] G.C. Moore, I. Benbasat, Integrating diffusion of innovationsand theory of reasoned
action models to predict utilization of information technology by end-users, Diffu-
sion and Adoption of Information Technology 1996, pp. 132146.
[34] X. Xu, V. Venkatesh, K.Y. Tam, S.-J. Hong, Model of migration and use of platforms:
role of hierarchy, current generation, and complementarities in consumer settings,
Management Science 56 (2010) 13041323.
[35] T. Eisenmann, G. Parker, M.W. Van Alstyne, Strategies for two-sided markets, Har-
vard Business Review 84 (2006) 92.
[36] J.C. Rochet, J. Tirole, Two-sided markets:a progress report,The Rand Journal of Eco-
nomics 37 (2006) 645667.
[37] A. Bhattacherjee, G. Premkumar, Understanding changes in belief and attitude to-
wards information technology usage: a theoretical model and longitudinal test,
MIS Quarterly 28 (2004) 229254.
[38] V. Venkatesh, S.A. Brown, H. Bala, Bridging the qualitative-quantitative divide:
guidelines for conducting mixed methods research in information systems, MIS
Quarterly 37 (2013).
[39] J. Corbin, A. Strauss, Grounded theory research: procedures, canons and evaluative
criteria, Zeitschrift für Soziologie 19 (1990) 418427.
[40] C.W. Crook, R.L. Kumar, Electronic data interchange: a multi-industry investigation
using grounded theory, Information Management 34 (1998) 7589.
[41] H.K. Klein, M. Myers, A set of principles for conducting and evaluating interpretive
eld studies in information systems, MIS Quarterly 23 (1999) 6794.
[42] G. Walsham, The emergence of interpretivism in IS research, Information Systems
Research 6 (1995) 376394.
[43] G. Walsham, Doing interpretive research,European Journal ofInformation Systems
15 (2006) 320330.
[44] J. Corbin, A.L. Strauss, Basics of Qualitative Research: Techniques and Procedures
for Developing Grounded Theory, 3rd ed. Sage Publications, Thousand Oaks, CA,
[45] K. Boudreau, Open platform strategies and innovation: granting access vs. devolv-
ing control, Management Science 56 (2010) 18491872.
[46] J. Mundt, F.S. Houston, Ubiquitous externalities:characteristics, climate, and impli-
cations for post-acquisition b ehaviors, Journal of Macromarke ting 30 (2010)
12 J. Song et al. / Decision Support Systems xxx (2018) xxxxxx
Please cite this article as: J. Song, et al., Platform adoption by mobile application developers: A multimethodological approach, Decision Support
Systems (2018),
[47] F.D. Davis, R.P. Bagozzi, P.R. Warshaw, Extrinsic and intrinsic motivation to use
computers in the work place, Journal of Applied Social Psycholo gy 22 (1992)
[48] R. Agarwal, J. Prasad, Are individual differences germane to the acceptance of new
information technologies? Decision Sciences 30 (1999) 361391.
[49] H. Ling, F. Zhao, Y. Wang, Impact of synergy between IT and business process on
organizational performance: a perspective of ambidexterity theory, PACIS 2009
Proceedings 2009, p. 116.
[50] L. Thompson, G.A. Fine, Socially shared cognition, affect, and behavior: a review
and integration, Personality and Social Psychology Review 3 (1999) 278302.
[51] R. Agarwal, J. Prasad, A conceptual and operational denition of personal innova-
tiveness in the domain of information technology, Information Systems Research
[52] V. Choudhary, E. Karahanna, The relative advantage of electronic channels: a mul-
tidimensional view, MIS Quarterly 32 (2008) 179200.
[53] H. Gatignon, J.-M. Xuereb, Strategic orientation of the rm and new product per-
formance, Journal of Marketing Research 34 (1997) 7790.
[54] D.H. Henard, D.M. Szymanski, Why some new products are more successful than
others, Journal of Marketing Research 38 (2001) 362375.
[55] R.J. Calantone, K. Chan, A.S. Cui, Decomposing product innovativeness and its ef-
fects on new product success, Journal of Pr oduct Innovatio n Management 23
(2006) 408421.
[56] E. Karahanna, R. Agarwal, C.M. Angst, Reconceptualizing compatability beliefs in
technology acceptance research, MIS Quarterly 30 (2006) 781804.
[57] G. Premkumar, K. Ramamurthy,S. Nilakanta, Implementation of electronic data in-
terchange: an innovation diffusion perspective, Journal of Management Informa-
tion Systems 11 (1994) 157186.
[58] M. Bradford, J. Florin, Examining the roleof innovation diffusion factors on the im-
plementation success of enterprise resource planning systems, International Jour-
nal of Accounting Information Systems 4 (2003) 205225.
[59] L.G. Tornatzky, K.J. Klein, Innovation characteristics and innovation ad option-
implementation: a meta-analysis of ndings, IEEE Transactions on Engineering
Management 29 (1982) 2845.
[60] I. England, D. Steward, S. Walker, Information technology adoption in healthcare:
when organisations and technology collide, Australian Health Review 23 (2000)
[61] T. Olivera, M.F. Martins, Literature review of information technology adoption
models at rm level, The Electronic Journal Information Systems Evaluation 14
(2011) 110121.
[62] T.S.H. Teo, M. Tan, W.K. Buk, A contingency model of internet adoption in
Singapore, International Journal of Electronic Commerce 2 (199798) 95118.
[63] V.S.Lai, J.L. Guynes, Anassessment of the inuence of organizationalcharacteristics
on information technolog y adoption decision: a discriminative approach, IEEE
Transactions on Engineering Management 44 (1997) 146157.
[64] S.D. Saleh, C.K. Wang, The management of innovation: strategy, structure, and or-
ganizational climate, IEEE Transactions on Engineering Management 40 (1993)
[65] A. Benlian, D. Hilkert, T. Hess, How open is this platform? The meaning and mea-
surement of platform openness from the complementors' perspective, Journal of
Information Technology 30 (2015) 209228.
[66] T.R. Eisenmann, G. Parker, M. Van Alstyne, Opening platforms : how, when and
why? in: Annabelle Gawer (Ed.), Platforms , Markets and Innovation, Edward
Elgar Publishing, 2009 (Chapter 6).
[67] M. Bastic, Success factors in transition countries, European Journal of Innovation
Management 7 (2004) 6579.
[68] J. Cho, J. Lee, Development of a new technolog y product evaluation model for
assessing commercialization opportunities using Delphi method and fuzzy Ahp
approach, Expert Systems with Applications 40 (2013) 53145330.
[69] M.E. Parry, X.M. Song, Identifying new product successes in China, Journal of Prod-
uct Innovation Management 11 (1994) 1530.
[70] S.Y. Sohn, H.S. Kim, T.H. Moon, Predicting the nancial performance index of tech-
nology fund for SME using structural equation model, Expert Systems with Appli-
cations 32 (2007) 890898.
[71] V. Kumar, P.K. Jain, Commercialization of new technologies in India: an empir-
ical study of perceptions of technology institutions, Technovation 23 (2003)
[72] M. Marinc, Banks and i nformation technology: marketability vs. relationships,
Electronic Commerce Research 13 (2013) 71101.
[73] R.G. Cooper, New product strategies: what distinguishes the top performers? Jour-
nal of Product Innovation Management 1 (1984) 151164.
[74] D. Talukdar, K. Sudhir, A. Ainslie, Investigating new product diffusion across prod-
ucts and countries, Marketing Science 21 (2002) 97114.
[75] S. Malhotra, K. Sivakumar, P. Zhu, Distance factors and target market selection: the
moderating effect of market potential, International Marketing Review 26 (2009)
[76] J. Whitelock, D. Jobber, An evaluation of external factors in the decision of Uk in-
dustrial rms to enter a new non -domestic market: an explorato ry study,
European Journal of Marketing 38 (2004) 14371455.
[77] M.F. Niculescu, S. Whang, Codiffusion of wireless voice and data services: an em-
pirical analysis of the Japanese mobile telecommunications market, Information
Systems Research 23 (2012) 260279.
[78] D. Gavalas, D. Economou, Development platforms for mobile applications: status
and trends, IEEE Software (2011) 7786.
[79] J. Lu, J.E. Yao, C.S. Yu, Personal innovativeness, social inuences and adoption of
wireless internet services via mobile technology, Journal of Strategic Information
Systems 14 (2005) 245268.
[80] Y. Malhotra, D.F. Galletta, Extending the technology acceptance model to ac-
count for social inuence: theoretical bases and empirical validation, Proceed-
ings of the 32nd Annual Hawaii International Conference on Information
Systems, 1999.
[81] E. Karahanna, D.W. Straub, Information technology adoption across time: a cross-
sectional comparison of pre-adoption and post-adoption beliefs, MIS Quarterly
23 (1999) 183213.
[82] E. Deci, Effects of externally mediated rewards on intrinsic motivation, Journal of
Personality and Social Psychology 18 (1971) 105115.
[83] E. Deci, R. Ryan, Intrinsic Motivation and Self-determination in Human Behavior,
Springer, 1985.
[84] R. Agarwal, E. Karahanna, Time ies when you're having fun: cognitive absorption
and beliefs abou t information te chnology usage, MIS Quarterly 24 (2000)
[85] H.W. Kim, H.C. Chan, S. Gupta, Value-based adoption ofmobile internet: an empir-
ical investigation, Decision Support Systems 43 (2007) 111126.
[86] M.C. Lee, Factors inuencing the adoption of internet banking: an integration of
Tam and Tpb with perceived risk and perceived benet, Electronic Commerce Re-
search and Applications 8 (2009) 130141.
[87] J. Mehrtens, P.B. Cragg, A.M. Mills, A model of internet adoption by SMEs, Informa-
tion Management 39 (2001) 165176.
[88] K.Y. Lin, H.P. Lu, Why people use social networking sites: an empricial study inte-
grating network externalities and motivation theory, Computers inHuman Behav-
ior 27 (2011).
[89] H. van derHeijden, User acceptance of hedonic information systems, MIS Quarterly
28 (2004) 695704.
[90] R.C. Beatty, J.P. Shim, M.C. Jones, Factors inuencing corporate web siteadoption: a
time-based assessment, Information Management (2001) 38.
[91] A. Sharma, A. Citurs, Radio frequency identication (RFID) adoption drivers: a rad-
ical innovation adoption perspective, Proceedings of the Americas Conference on
Information Systems, Omaha, NE, 2005.
[92] G. Lee, T.S. Raghu, Determinants of mobile apps' success: evidence from the app
store market, Journal of Management Information Systems 31 (2014) 133170.
[93] R. Garg, R. Telang, Inferring app demand from publicly available data, MIS Quar-
terly 37 (2012) 12531264.
[94] N. Jeong, Y. Yoo, T.Y. Heo, Moderating effectof personal innovativeness on mobile-
RFID services: based on Warshaw's purchase intention model, Technological Fore-
casting and Social Change 76 (2009) 154164.
[95] G. Bassellier, B.H. Reich, I. Benbasat, Information technology competence of busi-
ness managers: a denition and research model, Journal of Management Informa-
tion Systems 17 (2001) 159182.
[96] J.W. Moon, Y.G. Kim, Extending the TAM for a world-wide-web context, Informa-
tion Management 38 (2001) 217230.
[97] R. Jia, Problematic technology use: a negative outcome of computer playfulness
and personal innovativeless? Sprouts: working papers on, Information Systems 9
(2009) 127.
[98] A. Dickinger, M. Arami, D. Meyer, The role of perceived enjoymentand social norm
in the adoption of technology with network externalities, European Journal of In-
formation Systems 17 (2008) 411.
[99] R. Agarwal, J. Prasad, The role ofinnovation characteristics and perceivedvoluntar-
iness in the acceptance of information technologies, Decision Sciences 28 (1997)
[100] Y. Hwang, D.J. Kim, Customer self-service systems: the effects of perceived web
quality with service contents on enjoyment, anxiety, and E-trust, Decision Support
Systems 43 (2007) 746760.
[101] R.M. Stock, How does product program innovativeness affect customer satisfac-
tion? A comparison of goods and services, Journal of the Academy of Marketing
Science 39 (2011) 813827.
[102] D. Hilkert, A. Benlian, M. Sarstedt, T. Hess, Perceived Software Platform Openness:
The Scale and Its Impact on Developer Satisfaction, 2011.
[103] X.M.Song, M.E. Parry, A cross-national comparative study of new product develop-
ment processes: Japan and the Unit ed States, The Journal of Marketing (1997)
[104] L.T. Eby, M. Butts, A.Lockwood, Predictors of success in the era of the boundaryless
career, Journal of Organizational Behavior 24 (2003) 689708.
[105] G.-W. Bock, R.W. Zmud, Y. -G. Kim, J.-N. Lee, Behavioral intention formation in
knowledge sharing: examining the roles of extrinsic moti vators, social-
psychological forces, and organizational climate, MIS Quarterly 29 (2005) 87111.
[106] R. Faullant, E. Schwartz, L. Krajger, R.J. Breitenecker, Towards a comprehensive un-
derstanding of lead userness: the search for individual creativity, Creativity and In-
novation Management 21 (2012) 7692.
[107] S.S. Srinivasan,R. Anderson, K. Ponnavolu, Customer loyalty in e-commerce:an ex-
ploration of its antecedents and conseque nces, Journal of Retailing 78 (200 2)
[108] C. Fornell, D. Larcker, Evaluating structural equation models with unobservable
variablesand measurement error, Journalof Marketing Research 18 (1981) 3950.
[109] P.M. Podsakoff, S.B. MacKenzie, J.Y. Lee, N.P. Podsakoff, Common method biases in
behavioral research:a criticalreview of the literatureand recommended remedies,
Journal of Applied Psychology 88 (2003) 879.
[110] H. Liang, N. Saraf, Q. Hu, Y. Xue, Assimilation of enterprise systems: the effect of in-
stitutional pressures and the mediating role of top management, MIS Quarterly 31
(2007) 5987.
[111] P. Roma, F. Zambuto, G. Perrone, The role of the distribution platform in price for-
mation of paid apps, Decision Support Systems 91 (2016) 1324.
[112] K. Bent, The Decline of BlackBerry: Where RIM Went Wrong, 2011 (27
13J. Song et al. / Decision Support Systems xxx (2018) xxxxxx
Please cite this article as: J. Song, et al., Platform adoption by mobile application developers: A multimethodological approach, Decision Support
Systems (2018),
[113] T.F. Bresnahan, J.P. Davis, P.-L. Yin, Economic value creation in mobile applications,
The Changing Frontier: Rethinking Science and Innovation Policy, University of
Chicago Press 2014, pp. 233286.
[114] A. Ghose, S. Han, Estimating demand for mobile applications in the new economy,
Management Science 60 (2014) 14701488.
[115] J.C. Nunnally, Psychometric Theory, 2nd edition McGraw-Hill, New York, 1978.
Jaeki Songis the Jerry S. Rawls Endowed Professor of MIS at the Rawls College of Business
at Texas Tech University. His primary area of research is innovation in Information Sys-
tems. Specic research issues include service innovations, diffusion of information tech-
nologies, social and economic consequences of information technologies, and business
analytics. His researchndings have appearedN60 refereed journalpapers in Management
Science,Journal of Management Information Systems,theJournal of the Association for Infor-
mation Systems,Journal of Strategic Information Systems,IEEE Transactions on Professional
Communication,Decision Support Systems,Information& Management,Information Systems
Frontier, and other journals.
Jeff Baker is an As sociate Profess or in the School o f Business Administration at the
American University of Sharjah. His research focuses on IT strategy, with specicempha-
ses in e-commerce, mobile applications, and strategic alignment. His research has ap-
peared in the Jour nal of the AIS,Decision Support Systems,Communications of the AIS,
Information Systems Frontier, and other leading journals.
Ying Wang is anAssistant Professor at the Northern IllinoisUniversity. Herresearch inter-
ests primarily focuson business analytics, adoption and diffusion of smarttechnology, and
text analytics techniques. Her dissertation focuses on investigating the effects of online
reviews on product sales by considering the structure of reviews and the effect of time.
Her work has been pr esented at national and internati onal conference s, such as
Americas Conference on Information Systems, Association for Information Systems - Spe-
cial Interest Group on HumanComputer Interaction, IEEEInternationalConference on Ser-
vice Systems and Service Management, and Big XII+ MIS Symposium.
Hyoung Yong Choi has a Ph.D. student at the Georgia State University. His research has
appeared at leading IS conferences, with one paper nominated for a Best Paper award at
the AmericasConference on Information Systems2013. He is currentlyapplying to Ph.D. pro-
grams in the United States.
Anol Bhattacherjeeis a Professor of Information Systemsand Dean's Fellow at the Univer-
sity of South Florida.His research focuses on a broadrange of topics relatedto creation and
management of innovations, healthcare informatics, and social enterprises. He has pub-
lished 55 refereed journal papers that have received N6000 citations on Google Scholar.
His 2001 MIS Quarterly paper, cited N1800 times, is credited with creating a new stream
of research on technology continuance. He served on the editorial board of MIS Quarterly
for four years and is on the current editorial board of the Journal of the AIS.
14 J. Song et al. / Decision Support Systems xxx (2018) xxxxxx
Please cite this article as: J. Song, et al., Platform adoption by mobile application developers: A multimethodological approach, Decision Support
Systems (2018),
... However, the "adoption" process and users' behavioral factors that influence this process are substantially under-investigated (Harris et al., 2016). Interestingly, expanding the perspective on the literature on apps to include the adoption of more general information and communication technology (ICTs), the number of studies on their adoption increases (Palau-Saumell, Forgas-Coll, Sánchez-García, & Robres, 2019;Song, Baker, Wang, Choi, & Bhattacherjee, 2018). ...
... Studies investigating the adoption of ICTs have primarily applied different theoretical models and theories (Palau-Saumell et al., 2019), such as the theory of reasoned action (TRA), the theory of planned behavior (TPB), the technology acceptance model (TAM), the unified theory of acceptance and use of technology (UTAUT) 1 and 2, the innovation diffusion theory (IDT), and the social cognitive theory (SCT) (Song et al., 2018). The TRA has been developed to study human behavior from a social psychology perspective by considering one's feeling toward a specific behavior (attitude) and their perception regarding other people's perspective about that specific behavior (subjective norms; Kang, Ha, & Hambrick, 2015). ...
... Finally, the UTAUT-2 suggests that the intention to use a technology is influenced by performance expectancy, effort expectancy, social influence, hedonic motivation, price value, and habit; at the same time, the actual usage is mainly influenced by the intentions to use, facilitating conditions, and habit (Palau-Saumell et al., 2019). An additional theory that has been often used in ICTs literature is the IDT (Song et al., 2018). This theory includes variables such as relative advantage (of economic gains or perceived convenience), complexity (of the necessary effort), compatibility (with users' values, needs, and experience), observability (the degree to which innovation products are visible to others), and trialability (the degree to which products are experimented with before adoption) (Song et al., 2018;Min, So, & Jeong, 2019, based on Rogers, 1995. ...
This study focuses on the promotion and use of tourist-oriented built-for-disasterpurpose mobile applications. The investigation relies on two trends that are characterizing modern societies. The first trend refers to the fact that recently, rapidly, and worldwide, the number of mobile applications users increased. The second trend refers to the fact that, worldwide, international tourism demand highly increased from the beginning of the 1960s to the midyears of 2010s. These travelers are also users of mobile applications, and they use apps for several reasons, including those related to security/emergency issues. International travelers may need information such as risks at the destination, warnings, shelter locator services, emergency routes information, and traveler telephone hotline to be used in the case of earthquakes, tsunamis, infectious diseases, or other disasters. We considered the role of built-for-disasterpurpose mobile applications as a tool to provide information to tourists, increase their risk-awareness, and improve their disaster-preparedness. Previous studies suggest that these tools may have a high impact in that sense. However, the number of users informed about these applications tends to be very low. We systematically reviewed interdisciplinary academic contributions to analyze research on apps-users' intention to adopt mobile applications. Findings suggest opportunities for future tourism-oriented studies aiming at increasing the number of users of built-for-disasters mobile applications. In particular, we provide a research agenda taking into consideration the geographical spread of the studies, the used research techniques, and the adopted theories.
... Moreover, various smartphone developing organizations have been providing event clients with customized tools and programming platforms in accordance with their needs which can highly enable efficiency. Thus, the usage of the available development platforms makes it easier to solve complex problems that may arise in the process of implementing innovative mobile application ideas [9]. In addition, Wasserman's study 2010 as cited in [10] supports the idea of using robust tools and frameworks in the application development phase as it can ease the process. ...
... A key DOI enabler is technical compatibility, which is an innovation's compatibility with the current system in place, including hardware and software applications (Song et al., 2018). Therefore, technical compatibility can offer organizational benefits by enabling firms to easily integrate new technologies with existing technologies or systems. ...
The purpose of this study is to investigate the diffusion of smartness in hospitality and, furthermore, the assimilation of smart technologies in hotel business units. A qualitative exploratory investigation with an abductive research approach was adopted and data were collected through in-depth interviewing of 39 hotel managers in the highly advanced hospitality industry in the United Arab Emirates (UAE). Analysis of the data was conducted inductively to elicit themes within the technological, organizational, and environmental contexts and to mark trailheads for future research. The study yielded important insights into how barriers to the diffusion of smartness can be surpassed, and how smart technologies can be assimilated into existing hotel organizational structures and operations. The proposed aggregate dimensions for achieving assimilation of smart technologies in hotels highlight the importance of organizational sustained learning and smart networks synergism while shaping configurational system innovations.
... Next, in aiming to extract latent concepts in the data, we examined which descriptive codes hinted at broader themes and generated interpretative codes or second-order categories. This transformation allowed a manageable number of constructs to be gleaned from the original multitude of concepts (Song et al., 2018). Subsequently, inter-relationships between and across the second-order categories were organised to generate the overall theoretical explanation (Nag & Gioia, 2012). ...
We examine the co-creation of value by bottom of the pyramid consumers in the health services context. Following a qualitative approach, we use the empirical context of the National Rural Health Mission (NRHM) in India, one of the largest healthcare policy interventions in the world. Our grounded theory-based research used semi-structured interviews conducted in 28 villages across three states using a purposive sample (NRHM beneficiaries). The qualitative data were analysed utilising a discovery-oriented approach, in accordance with the value co-creation and service-dominant logic frameworks. We draw insights about the roles of individual agency, the individual’s social milieu, and supplementary forces in affecting different types of individual wellbeing. Our results help to explicate the processes of the exploitation, combination and reinforcement of resources within a co-created healthcare ecosystem, and six propositions explain value co-creation-driven health service delivery. We also discuss implications for marketers and policy implementers.
... Thus, we formed the second hypothesis of the model: In the original model of Venkatesh et al. [8] and the extended model UTAUT 2 [9], social influence measures the extent to which users are influenced by what others (friends, family) think about the use of technology. Some previous papers in the same online purchase context show that social influence is a very strong antecedent of behavioral intention [20,[34][35][36] and adoption of an IT platform [37]. Thus, we formed a new hypothesis: ...
Full-text available
The increasing use of artificial intelligence (AI) to understand purchasing behavior has led to the development of recommendation systems in e-commerce platforms used as an influential element in the purchase decision process. This paper intends to ascertain what factors affect consumers’ adoption and use of online purchases recommendation systems. In order to achieve this objective, the Unified Theory of Adoption and Use of Technology (UTAUT 2) is extended with two variables that act as an inhibiting or positive influence on intention to use: technology fear and trust. The structural model was assessed using partial least squares (PLS) with an adequate global adjustment on a sample of 448 users of online recommendation systems. Among the results, it’s highlighted the importance of the inhibiting role of technology fear and the importance that users attach to the level of perceived trust in the recommendation system are highlighted. The performance expectancy and hedonic motivations have the greatest influence on intention to use these systems. Based on the results, this work provides a relevant recommendation to companies for the design of their e-commerce platforms and the implementation of online purchase recommendation systems.
... In this respect, complementor support for a new platform in its initial launch can be interpreted by the platform owners as a sign of commitment to the relationship (J. Song, Baker, Wang, Choi, & Bhattacherjee, 2018;Srinivasan & Venkatraman, 2018). In fact, mutual trust between platform owners and complementors stimulates investment in new technologies (Kapoor & Lee, 2013) and hence, differentiated complementary products. ...
Platform ecosystems include complementary product markets that are creating unprecedented business opportunities. Competition within complementary product markets is increasingly intense and fast-changing, and this leads to increased complexity in designing strategies to gain competitive advantage. Prior literature has extensively analyzed the key drivers of platform adoption by end users. Recent research is highlighting the relevance of complementors as critical ecosystem agents that have some strategic freedom. However, current insights into complementors still provide a limited and fragmented explanation of their competitive advantages. This conceptual paper examines current research on platform ecosystems, open innovation and market-based standardization , and analyzes complementor strategic decisions aimed at building a competitive advantage. Our framework contributes to the literature by providing a complementor-centered approach and suggesting a set of propositions that may guide future research. Future studies may extend the insights into the competitive and corporate strategies of complementors in platform ecosystems.
... Song, et al. offered evidence that perceived technical features, human characteristics and social impact all play a role in the adoption of IT technologies such as mobile phone applications. The paper refined understanding of factors affecting platform adoption that emphasizes not only the functions of user-friendliness and utility-related factors but also the theoretical growth of the system itself and its network effects [3]. ...
The purpose of this study is to investigate the factors that influence the usage intention for different instant messaging application platforms. This study targeted the widespread instant messaging software LINE, for which a survey of 485 valid respondents was conducted in Taiwan. Statistical and model structure analyses were applied to validate the hypotheses proposed in this study. We examined the differences between single- and multi-platform use cases in terms of the relationship between consumer value and behavioral intention. The results show that perceived interactivity, perceived playfulness, and perceived usefulness have significant effects on the continued usage and recommendation intentions of users in a single-platform use case. However, perceived interactivity and perceived playfulness did not have significant effects in multi-platform use cases. The findings of this study provide a new perspective for investigating the intention to continue using a multiplatform application. The results will benefit subsequent studies in this area and provide guidance for the development of instant messaging and other social media applications.
Purpose Aging has detrimental effects on elders due to their physical health and financial hardship. Elders face neglect, insult and abuse in society due to causes related to physical health and financial issue from caregivers. This study aims to identify the measures of physical health and financial hardship and classifies elders under neglect, insult and abuse categories. Design/methodology/approach The propositions of existence and classifying elders under neglect, insult and abuse categories were tested by using discriminant analysis and their profiling was done by perceptual mapping technique. Findings The elder neglect category identified as a prominent category due to physical health while elder insult and abuse were caused by physical health and financial hardship both. The present study portrays the multi-dimensional facets related to elders’ ill-treatment. The elder’s ill-treatment categories were profiled to imply the measure of elevating elders’ dignity and care at a personal level and society at large. Originality/value This study classifies elders under neglect, insult and abuse categories. This classification may facilitate the medical practitioners, academicians and government and non-government social welfare agencies in understanding elder abuse with new perspectives.
Conference Paper
Complementary products and services of third-party complementors have become one of the cornerstones for the success and sustainability of digital platforms. Information Systems (IS) research has paid considerable attention to the effects of control modes on shaping platform governance and thus regulating complementors and their complements. However, there is still a lack of comprehension of the effects of a widely applied, yet underexamined control mode, namely input control (i.e., the set of mechanisms that screen and sort out complementors and their complements before entering the digital plat-form's ecosystem). We tested our hypotheses using an integrated research approach that includes quantitative data from a survey with 114 web browser extension developers and qualitative data based on semistructured interviews with 8 developers. The results demonstrate convergent findings in support of the assertion that both complementor-and complement-related perceived input control negatively impact complementors' perceived usefulness, satisfaction and continuance intentions. As such, our paper contributes to IS governance research primarily by conceptually distinguishing between complementor-related and complement-related input control and by uncovering their distinct effects on critical com-plementor attitudes and behavioral intentions. Moreover, our study offers insights that can help platform providers to effectively manage their gatekeeping and screening processes.