Journal of Business Research 67 (2014) 1360–1367

Contents lists available at ScienceDirect

Journal of Business Research

Is the more always better? A comparative study of internal and external

integration practices in new product and new service development
Christian Homburg a,b,⁎, Christina Kuehnl a
a
Department of Marketing at the University of Mannheim, Germany
b
Department of Management and Marketing, University of Melbourne, Australia

a r t i c l e i n f o a b s t r a c t

Article history: This study investigates the relationship between internal and external integration practices and innovation
Accepted 15 August 2013 success of new products and new services. Building on the idea that key success drivers in new product and
Available online 11 September 2013 new service development may have implementing costs besides their obvious benefits, this article examines
the possibility that a nonlinear relationship in the shape of an inverted U exists between innovation success
Keywords:
and the antecedents examined in this research. The present study also addresses scholars' call for research to
New product development
New service development
investigate differences in the drivers of new product and new service success. The findings suggest that differ-
Nonlinear effects ences exist in the nature of the relationship—that is, linear versus nonlinear—between cross-functional integra-
Integration practice tion, customer integration, and interfirm collaboration and innovation success in a new product versus new
service setting.
© 2013 Elsevier Inc. All rights reserved.

1. Introduction
Practitioners and researchers consider integration practices as
critical to innovation success (Hillebrand & Biemans, 2003, 2004;
Johnsen, Phillips, Caldwell, & Lewis, 2006; Tessarolo, 2007). Integra-
tion practices refer to the cooperation and communication between
different stakeholders in the innovation process of a new service or
new product (Millson & Wilemon, 2002). Commonly, researchers
distinguish between cross-functional integration as internal integra-
tion practice and customer integration as well as interfirm collabora-
tion as external integration practices (Hillebrand & Biemans, 2003,
2004; Tessarolo, 2007).
While research provides ample evidence for the positive effects
of these integration practices on innovation success (Ernst, Hoyer,
& Rübsaamen, 2010), managers encounter major obstacles when
implementing them in business practice (BusinessWeek, 2008). For
example, customer integration may increase the risk of developing
an innovation that only serves a niche market instead of addressing
the entire customer base. Accordingly, the assumption of “the more,
the better”—that is, a positive and linear relationship between cross-
functional integration, customer integration, interfirm collaboration
and innovation success—might be questioned.
Likewise, prior literature primarily focuses on integration practices in
new product development (NPD), while leaving the effects in new
service development (NSD) rather unexplored (Hauser, Tellis, & Griffin,
⁎ Corresponding author at: Marketing Department I, University of Mannheim, Schloss,
D-68131 Mannheim, Germany. Tel.: +49 621 181 1555 (Voice); fax: +49 621 181 1556.
E-mail address: homburg@bwl.uni-mannheim.de (C. Homburg).
2006). This largely contrasts economic reality where service companies
account for 70% of the world's advanced economies' gross domestic
product (Ostrom et al., 2010) and many companies offer a portfolio of
both services and products. As such, research provides only little guid-
ance for these companies on how to implement integration practices in
a new service context (Droege, Hildebrand, & Forcada, 2009).
Against this background, the purposes of this study are as follows.
First, it questions the readily accepted linear nature of relationships
between three integration practices—cross-functional integration, cus-
tomer integration, and interfirm collaboration—and innovation success
and provides evidence for nonlinear effects. Second, by deriving an
integrative framework comprising new services and products, it aims
at extending the ongoing debate over whether success drivers of
product innovations also pertain to service innovations (Kindström,
Kowalkowski, & Sandberg, 2013). Thus, it responds to calls for research
on investigating differences in the drivers of innovation in these two
product categories and contributes to explaining inconsistencies in
results of prior research on integration practices (Droege et al., 2009;
Hauser et al., 2006). Interestingly, the results show a significant increase
in the explained variance of innovation success when analyzing the new
product and new service samples separately instead of using a joint
dataset. Furthermore, the key results of this study show substantial dif-
ferences in the nature of relationship of integration practices between
service and product innovations and hence, highlight the need to
distinguish between these two categories. Third, by examining pre-
dictors of new service success this research strives to contribute to
a better understanding of service innovations, which are “among
the least studied and understood topics in the service management
literature” (Menor, Tatikonda, & Sampson, 2002, p. 136; O'Cass,
Song, & Yuan, 2012).

0148-2963/$ – see front matter © 2013 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jbusres.2013.08.017
 C. Homburg, C. Kuehnl / Journal of Business Research 67 (2014) 1360–1367 1361

2. Conceptual background and literature review
Integration practices provide obvious benefits for innovations such
as a reduced cycle times (Bstieler & Hemmert, 2010). However, they
also vary considerably in their effect sizes and even carry alternate
signs with respect to innovation success (Evanschitzky, Eisend,
Calantone, & Jiang, 2012; Henard & Szymanski, 2001; Montoya-Weiss
& Calantone, 1994). Hence, some scholars acknowledge their negative
side effects, among which are for example the monetary and non-
monetary costs of solving conflicts in cross-functional teams (Cuijpers,
Guenter, & Hussinger, 2011). Table 1 summarizes major benefits and
costs of the three integration practices.
While the majority of prior literature focuses on the benefits of
implementing integration practices in NPD and NSD, the simultaneous
assessment of costs and benefits of integration practices constitutes an
under-researched issue in innovation management (Cuijpers, Guenter
and Hussinger, 2011). However, reviewing the theoretical and empiri-
cal arguments in favor for and against implementing integration prac-
tices as illustrated in Table 1 seems to suggest that the relationships
between determinants and innovation success may be neither purely
positive nor negative. Instead, this study integrates both—positive and
negative—positions into one overarching conceptual framework and
examines the possibility that a nonlinear relationship, and not a linear
one as most previous research suggests, exists between innovation
success and cross-functional integration, customer integration, and
interfirm collaboration.
A nonlinear relationship originating from the prevalence of costs or
benefits at some point could provide a strong explanation for inconsis-
tencies in prior results (Cohen, Cohen, West, & Aiken, 2003). More pre-
cisely, this research supposes that weak versus strong effect sizes may
arise from previous literature exclusively considering a predictor's line-
ar relationship instead of investigating its potentially nonlinear nature.
This reasoning is based on the idea that the law of diminishing returns
applies to the benefits of integration practices with regard to innovation
success, whereas monetary and non-monetary costs of their implemen-
tation continuously increase (Duysters & Lokshin, 2011). For example, a
firm may benefit from the advantages of an increasing level of cross-
functional integration until a certain inflection point, after which
marginal costs of solving conflicts between functional departments are
higher than the expected benefits from this increased integration prac-
tice, implying the existence of an optimum level (Cuijpers et al., 2011;
Ernst, Hoyer and Rübsaamen, 2010; Schleimer & Shulman, 2011).
Some researchers additionally argue that the inconsistent findings of
antecedents to innovation success might derive from product category
(i.e., product versus service) and call for the investigation of differences
in the drivers of product versus service innovations (Droege et al., 2009;
Hauser et al., 2006; Henard & Szymanski, 2001). An answer to the fun-
damental question of whether service and product innovations are sim-
ilar to or distinct from each other, might allow researchers and
practitioners to meaningfully leverage “the knowledge from new prod-
uct development and to apply it to new service process development”
(Karniouchina, Victorino, & Verma, 2006, p. 277). However, the few
existing comparative studies between NPD and NSD provide several
contradictory or inconsistent findings with respect to integration prac-
tices, and thus call for further research (for an overview, see: Droege
et al., 2009).
The present study combines both research streams and suggests that
one answer to the inconsistencies in results of prior research on integra-
tion practices could lie in differences in the nature of the relationship—
linear versus nonlinear—in a service versus goods context. Hence, the
extent of costs or benefits associated with an integration practice may
depend on the product category that is, new service or new product. Ac-
cordingly, the central premise of the study's model refers to differences
in the cost–benefit ratio when implementing the three integration prac-
tices in NSD and NPD. Extending the assumption that particularities of
services such as intangibility and inseparability may provoke differ-
ences in NSD compared to NPD (Nijssen, Hillebrand, Vermeulen, &
Kemp, 2006), this article argues that these distinctive differences
might result in the prevalence of either benefits or costs that coincide
with an integration practice.
3. Hypotheses development
3.1. Effect of cross-functional integration on innovation success
Cross-functional integration constitutes an organizational approach
for collecting, sharing, and processing information among all members
involved in NPD or NSD (Griffin, 1997). Mixed results exist on the

Table 1
Major benefits and costs1 of implementing integration practices.

Integration Major benefits of …. Major costs of ….

practice

…. cross- ▪ information exchange beyond functional boundaries
functional in- ▪ critical reassessment of each functional perspective on the NPD/NSD process
tegration ▪ reduced cycle times and development costs
▪ increased flexibility of the workforce
▪ increased functional performance of innovation
selected references: Cuijpers et al. (2011), Ernst et al. (2010), Lievens and
Moenaert (2000), and Troy, Hirunyawipada, and Paswan (2008).
…. customer ▪ provision of novel ideas
integration ▪ cost reductions
▪ faster delivery of improved product quality and superior product advantage
▪ validation of the product/service design
▪ increased willingness-to-pay and market acceptance
selected references: Alam (2002), Fang (2008), Kristensson, Matthing, and
Johansson (2008), and Magnusson, Matthing, and Kristensson (2003).
.... interfirm ▪ provision of novel ideas
collaboration ▪ cost reductions and shared responsibilities
▪ access to new knowledge, skills, and technologies
▪ efficient resource allocation
▪ improved cycle times
selected references: Emden, Calantone, and Droge (2006), Knudsen (2007),
Rindfleisch and Moorman (2001), and Schleimer and Shulman (2011).
1
In line with Cuijpers et al. (2011, p. 565), costs are not restricted to monetary expenses but also to non-monetary expenses such as effort and time.
▪ coordination of the workflow of NPD team members
▪ decision making between employees with different underlying goals
▪ conflicts over resources and technical issues
▪ budget overrun and project failure
selected references: Cuijpers et al. (2011), Ernst et al. (2010), and Maltz and
Kohli (2000).
▪ identifying and incentivizing appropriate customers
▪ capturing and converting customers' future needs into innovations
▪ higher coordination efforts and increased workload of NPD team members
▪ concerns of secrecy and ownership of intellectual property
▪ risk of serving a niche market
selected references: Alam (2002), Chan, Yim, and Lam (2010), Ernst et al.
(2010), Fang (2008).
▪ identifying appropriate collaborating firms
▪ concerns of secrecy, knowledge spillover, and ownership of intellectual
property
▪ higher coordination and monitoring efforts
budget overrun and time delays
▪ risk of opportunistic behavior by one partner
selected references: Duysters and Lokshin (2011)’ Lui, Ngo, and Hon (2006), Luo,
Rindfleisch, and Tse (2007) and Rindfleisch and Moorman (2003).
1362 C. Homburg, C. Kuehnl / Journal of Business Research 67 (2014) 1360–1367
impact of this integration practice on innovation success in a
manufacturing versus service context. Whereas Atuahene-Gima (1996)
and Henard and Szymanski (2001) found no significant difference, the
results of the meta-analysis by Troy, Hirunyawipada and Paswan
(2008) indicate a stronger, linear effect of cross-functional integration
on new service than on new product success. Against this background,
it comes as no surprise that Droege et al. (2009) call the role of cross-
functional integration in NPD versus NSD a topic for future research.
Services are classically unique in that the front office (i.e., the part of
service provision usually visible to customers) is separate from the back
office (i.e., the part of service provision usually invisible to customers;
Fitzsimmons & Fitzsimmons, 1994). Yet, since decoupling the marketing
and production activities is impossible for most service deliveries, the
two offices must function together as an integrated whole to provide a
composite service and ensure a smooth service delivery. In simple
terms, “without cross-functional teams NSD may suffer from functional-
ly departmentalized structures that impede NSD” which may harm new
service success (De Jong & Vermeulen, 2003, p. 851). This constitutes
“an issue not raised in the NPD literature, and leads to a unique chal-
lenge faced in NSD” (Menor, Tatikonda and Sampson, 2002, p. 145).
Further, because of the inseparability of service production and con-
sumption customers may easily notice a lack of cooperation among dif-
ferent functional units, leading to lower satisfaction levels. Conversely,
customers buying tangible offerings typically only experience the out-
come of manufacturing—the final product itself—and are less directly af-
fected by a lack of collaboration among functional units during NPD
(Troy et al., 2008).
In contrast to these particularities of the service delivery process,
this research suggests that in a manufacturing setting, the overall effect
of cross-functional integration in NPD is mostly a combination of prev-
alent costs and benefits. Thus, it argues that an optimum level exists be-
yond which incremental costs are too high and incremental benefits
decrease (e.g., Kratzer, Leenders, & Van Engelen, 2004). As such, benefi-
cial effects like the linkage of functional viewpoints and the embrace-
ment of information exchange beyond functional boundaries during
NPD may occur first (Lievens & Moenaert, 2000). Beyond a certain de-
gree, however, the implementing costs of cross-functional integration
may then outweigh benefits and may eventually result in a decrease
of new product success. This is because the sequential nature of new
product production and consumption may provoke a cooperation
among functional units that usually only pertains to one specific innova-
tion project and that may be less close than a service context generally
requires. Therefore, negative side effects resulting from the organiza-
tional costs, time, and efforts for team meetings, coordinating the
workflow of team members from various functional units, solving
inter-functional conflicts, and the potential information overload of
NPD members' processing capabilities (De Luca & Atuahene-Gima,
2007) may have a stronger negative effect in a manufacturing than in
a service context where employees are used to and almost obliged to
collaborate continuously.
H1a. The effect of cross-functional integration on new product success
is nonlinear in the shape of an inverted U for product innovations.
H1b. The effect of cross-functional integration on new service success is
positive and linear for service innovations.
3.2. Effect of customer integration on innovation success
Since “future NSD research could focus on how customer involve-
ment in the NSD process should be different from the type of customer
involvement in a typical NPD process” (Menor et al., 2002, p. 145), this
study examines whether the effect of customer integration on innova-
tion success exhibits significant differences in NSD versus NPD. Customer
integration refers to the extent to which a company actively involves
customers in the innovation process (Alam, 2002).
Literature in NPD strongly emphasizes the benefits that flow from
listening to the customer (Griffin & Hauser, 1993), which expectedly in-
crease new product success (Gruner & Homburg, 2000). More precisely,
customer integration implies that companies invite customers to act as
designers, to suggest ideas for innovative products, and that during the
execution of an innovation project they ask customers for continuous
feedback as to whether their needs are being met (Kristensson,
Matthing and Johansson, 2008). Hence, by this integration practice,
companies can increase the alignment of an innovation with cus-
tomers' preferences, needs, and desires. Accordingly, this research
assumes that a positive and linear effect exists between customer in-
tegration and new product success.
In contrast, while the above mentioned benefits of customer integra-
tion also generally apply to NSD, the simultaneous production and con-
sumption of services often results in greater customization in service
firms than in manufacturing companies (Anderson, Fornell, & Rust,
1997). As a result, customer integration in NSD may engender the risk
that new services are too customized and fail to target a firm's entire
customer base. Over-customization may then decrease sales of the
new service and affect innovation success negatively (Ernst et al.,
2010). Moreover, customers integrated into NSD may have difficulty
in evaluating a new service owing to its intangibility and the lack of a
prototype of a service innovation (Avlonitis, Papastathopoulou, &
Gounaris, 2001). Thus, intangibility requires more mental effort to get
an idea of a service innovation, which can make customer integration
in NSD either extremely costly or rather worthless. Finally, Chan, Yim
and Lam (2010), p. 48 found that “customers' increased involvement
in the service process (….) shift more power from service employees
to customers and thereby increase employee workloads”. Hence, this
research assumes that intense customer integration in NSD may lead
to a decrease in NSD teams' quality of new service provision that nega-
tively affects new service success. In sum, given the challenges and po-
tential problems of customer integration in NSD, this study proposes
that beyond a certain point, negative side effects override its benefits,
implying the existence of an optimum level of customer integration in
NSD.
H2a. The effect of customer integration on new product success is
positive and linear for product innovations.
H2b. The effect of customer integration on new service success is
nonlinear in the shape of an inverted U for service innovations.
3.3. Effect of interfirm collaboration on innovation success
Studies evaluating the impact of interfirm collaboration on new
product performance “present inconsistent results: some show that
these relationships are negative or insignificant, while others find they
are positive” (Tsai, 2009, p. 765) with leaving this success driver largely
unexplored in NSD (Eisingerich, Rubera, & Seifert, 2009). Hence, this
research aims to investigate whether the “fundamental tension between
the dynamics of innovation and the logic of partnering” (Bidault &
Cummings, 1994, p. 33)—owing to a divergent level of flexibility and nor-
mative contracting—exists in both NPD and NSD, or whether there are
differences in the effects between both innovation contexts. Thereby,
interfirm collaboration captures arrangements among two or more orga-
nizations to jointly acquire and utilize information and know-how relat-
ed to R&D, manufacturing or new service provision, and market launch
(Rindfleisch & Moorman, 2001).
Given the above controversy, this study agrees with the findings by
Luo, Rindfleisch and Tse (2007) that a nonlinear relationship in the
shape of an inverted U exists between interfirm collaboration and new
product success. This is because interfirm collaboration in NPD can gen-
erate, on the one hand, advantages such as efficient resource allocation
and pooling of capabilities, providing access to new skills or technolo-
gies, and sharing of R&D costs and risks leading to higher new product
 C. Homburg, C. Kuehnl / Journal of Business Research 67 (2014) 1360–1367 1363

performance (Emden, Calantone and Droge, 2006; Walter, 2003). On
the other hand, disadvantages and costs like opportunistic behavior by
one, knowledge spillover, distrust and costs of monitoring interfirm
partners as well as communication problems owing to different organi-
zational cultures (Emden et al., 2006; Lui, Ngo and Hon, 2006) can
accrue from interfirm collaboration. These disadvantages may increase
when interfirm collaboration becomes too intensive in NPD. In sum, a
moderate level of interfirm collaboration in NPD may be most profitable
taking in balance potential benefits and negative side effects derived
from collaborating with external business partners.
In contrast, this research suggests that service firms may initially
benefit from interfirm collaboration by improving innovating compe-
tences and by increasingly thinking out of the box. Thus, this variable
may exert a positive impact on new service success in the beginning.
However, due to the intangibility of service innovations their missing
patent protection provokes new challenges for interfirm collaboration
in NSD (Song, Di Benedetto, & Zhao, 1999). This missing patent protec-
tion may increase the costs for monitoring partners and retaining im-
portant information from them. Therefore, the intangibility of service
innovations may reduce the initially positive effects of interfirm collab-
oration on new service success.
Against this background, building trust between partners is an even
more critical issue in NSD than in NPD (Bstieler, 2006). This is because
trust gradually eliminates suspicions and conflicts of interests (Lui
et al., 2006) and alleviates concerns of non-patentability causing an in-
tense cooperation between partners. Hence, having built an essential
level of trust, service companies can “learn and focus on the develop-
ment of new service solutions” (Eisingerich et al., 2009, p. 347) when
collaborating intensively with business partners. Furthermore, due to
the simultaneity of service production and consumption, an efficient
interfirm collaboration may require a close and intense cooperation be-
tween business partners because—similar to cross-functional integra-
tion in NSD—customers might immediately recognize a low level of
cooperation between partners in a service setting. This may then result
in low levels of customer satisfaction and hence, in low new service
performance.
Taken together, this research proposes that the relationship between
interfirm collaboration and new service performance is U-shaped, with
new service performance declining due to the missing patent protection,
reaching its low point at a moderate level, and increasing with higher in-
tensity owing to a higher level of trust and the need for close cooperation
between business partners to guarantee a smooth new service provision.
H3a. The effect of interfirm collaboration on new product success is
nonlinear in the shape of an inverted U for product innovations.
H3b. The effect of interfirm collaboration on new service success is
nonlinear in the shape of a U for service innovations.
4. Methodology
related to service innovations and 125 to product innovations. Neither
hybrid innovations comprising service and product innovations nor
pure process innovations were taken into account for this study.
4.2. Measure development and assessment
Twenty-five participants in an executive marketing class pretested
the questionnaire. Measurement of innovation success was adapted
from De Luca and Atuahene-Gima (2007). With respect to the indepen-
dent variables, the article by Atuahene-Gima (2005) provided the mea-
surement for cross-functional integration. This variable comprises three
items concerning the existence and cooperation intensity of cross-
functional teams. The measure customer integration refers to the inten-
sity of customer integration in general, the intensity of discussions
about the innovation, and the regularity of meetings with customers. Fi-
nally, the scale for interfirm collaboration indicates the intensity of
interfirm collaboration for R&D, production, or market launch with ex-
ternal partners (Rindfleisch & Moorman, 2001).
This research used traditional psychometric procedures to assess reli-
ability and validity of the measures. The calculated coefficient alpha
values for all reflectively measured constructs exceed the recommended
threshold of .70 (Bagozzi & Yi, 1988). Subsequently, a confirmatory mea-
surement model containing all constructs in the framework using Mplus
4.1 (Muthén & Muthén, 2006) shows an adequate fit of the measurement
model to the data (χ2 = 233.1; df = 111; χ2/df = 2.1; CFI = .92;
RMSEA = .06; SRMR = .05; Hair, Black, Babin, & Anderson, 2010).
All composite reliabilities exceed the recommended threshold of .70
(Bagozzi & Yi, 1988), and with one exception, item reliabilities are larger
than .40 as recommended by Bagozzi (1994). Finally, all constructs show
discriminant validity using the procedure proposed by Fornell and
Larcker (1981). Table 2 shows descriptive statistics and correlations.
Appendix A lists the coefficient alphas, composite reliabilities, average
variance extracted, and item reliabilities.
4.3. Control variables
This research also included four control variables in the analyses that
prior literature has shown to affect innovation success. Specifically, it
takes into account that the predevelopment proficiency, that is the
quality of idea generation, idea selection, concept creation, and business
case execution, as well as launch proficiency can vary considerably and
thus, can have a powerful impact on innovation success (Henard &
Szymanski, 2001). To rule out that results may depend on the degree
of innovativeness, this study controls for its effect (Szymanski, Kroff, &
Troy, 2007). Finally, innovation success may vary with different degrees
of environmental turbulence which relate to technological and market
turbulence (Im & Nakata, 2008; Maltz & Kohli, 1996).

4.1. Data collection and sample

To obtain the necessary data, this study relied on a large-scale survey Table 2
of companies in manufacturing and service environments. A commer- Descriptive sample statistics and correlations.

cial provider delivered a random sample (n = 3301). These firms Variable Mean SD 1 2 3 4
were contacted by telephone to identify members with primary respon- 1 Innovation success 5.21 1.13 1.00 .22*** .12* .02
sibility for the latest innovation project in the company. This procedure 5.10 1.26
yielded a total of 1178 managers (e.g., Head of Innovation Management, 2 Cross-functional integration 5.78 .99 .12* 1.00 .31**** −.07
Head of Marketing, and Head of Services). Subsequently, managers re- 5.47 1.19
3 Customer integration 4.87 1.68 .11* .18** 1.00 .08
ceived a questionnaire to indicate whether their answers referred to a
4.39 1.86
service innovation or to a product innovation. With respect to general 4 Interfirm collaboration 2.96 1.76 −.28*** .08 .12* 1.00
company information, managers either referred to their SBU or, if 2.95 1.85
their company had only one SBU, to their company. In total, 285 respon- Notes: *p b .10; **p b .05; ***p b .01. The first (second) row indicates mean and standard
dents participated in the survey. The final sample consists of 257 re- deviation (SD) for product (service) innovations. Below (above) the diagonal indicated by
spondents, or 21.8%. These respondents indicated that 132 answers 1.00, correlations for service (product) innovations are provided.
1364 C. Homburg, C. Kuehnl / Journal of Business Research 67 (2014) 1360–1367

5. Analysis and results
To test for nonlinear relationships between the integration practices
and innovation success, this research followed the procedure by Lukas
and Menon (2004). After mean-centering all independent variables,
we established squared terms for the integration practices and ensured
that multicollinearity is not a problem in the regression models
(Echambadi & Hess, 2007). Two separate hierarchical analyses for either
new products or new services allowed checking whether the inclusion
of the squared terms significantly increases the explained variance of in-
novation success. Second, the investigation of the significance of each
squared term took place. More details on these two steps follow.
Table 3 presents the results.
First, the inclusion of squared terms of the three integration prac-
tices for new products and new services (see model 2 of each product
category) yields a significantly better specified model compared to
model 1 and thus provides initial support for the hypothesized
nonlinear relationships. A comparison of the variance explained be-
tween model 1 and model 2 demonstrates a larger R2 for model 2 of
new products and new services. An F test of the change in R2 indicates
that the difference is significant for each setting (p b .05). The inclusion
of cubic terms did not show a significant increase in R2, thus ruling out a
cubic function of the predictors (Cohen, Cohen, West and Aiken, 2003).
Conducting the second step of the hypothesis testing, this
research finds support for hypothesis H1a, suggesting an inverted
U-shaped relationship between cross-functional integration and in-
novation success in a new product context since the sign of the
squared term of cross-functional integration is negative and signifi-
cant (ßproduct, squared = − .17, p b .05). In contrast, H1b proposes a
positive and linear relationship between cross-functional integra-
tion on innovation success in a new service setting. The analysis
confirms this hypothesis as the result for the linear term is significant
(ßservice, linear = .19, p b .05) and the squared term is insignificant
(see Table 3, model 2: ßservice, squared = .06, p b .10). Findings do
not support H2a (ßproduct, linear = −.03, p b .10). Interestingly, however,
customer integration is U-shaped related to new product success (H2b:
ßproduct, squared = .14, p b .05). The study finds an inverted U-shaped
relationship between customer integration and new service success
as proposed by H2b (ßservice, squared = −.27, p b .01). Finally, H3a
proposes a nonlinear, inverted U-shaped relationship between
interfirm collaboration and new product success and a U-shaped
effect on new service success. Results confirm both hypotheses (H3a:
ßproduct, squared = − .15, p b .05; H3b: ßservice, squared = .19, p b .05).
6. Discussion and implications
6.1. Research issues
This study extends academic marketing research in several ways.
First, this investigation provides empirical evidence of nonlinear rela-
tionships between key success drivers and innovation success. These in-
sights propose that researchers should critically reassess previous
findings regarding the relationship between a success factor and perfor-
mance, which suggest that the more of a success driver, the better it is
for innovation success. In particular, scholars should account for costs
related to the implementation of a specific success factor because
these costs may cause a nonlinear relationship between the respective
independent and dependent variables. Therefore, future research
should explicitly rule out the existence of nonlinear effects.
Second, the finding of (inverted) U-shaped effects in NPD and/or
NSD adds a fresh perspective on the research issue of whether the man-
agement of service innovations differs from that of product innovations.
This is because results deliver a negative answer to the question of
whether knowledge gained in NPD readily applies to NSD (Hauser
et al., 2006; Karniouchina, Victorino and Verma, 2006).
Nonlinear effects in the shape of an (inverted) U exist for cross-
functional integration in NPD, and customer integration as well as
interfirm collaboration in both NSD and NPD. Specifically, the finding
of a positive and linear effect of cross-functional teams in NSD partially
replicates the results of Troy et al. (2008). However, this study shows
that in NPD, beyond a specific point, the costs of this activity outweigh
the benefits and result in an inverted U-shaped effect. Results also con-
firm the prediction of an inverted U-shaped relationship of customer in-
tegration and new service success. This finding implies that a moderate
level of customer integration in NSD is most successful. In contrast, this
study detected a U-shaped relationship between customer integration
and new product success. Therefore, incremental benefits overwhelm
incremental costs in both a low and a high degree of customer integra-
tion in NPD. Kirca, Jayachandran, and Bearden (2005) found that due to
the close interaction of customers, customer input is less common in
day-to-day business for manufacturers than for service firms and there-
fore service providers may have a better knowledge of customers'

Table 3
Results of the hierarchical regression model (dependent variable: innovation success).

Product innovation Service innovation

Model 1 Model 2 Model 1 Model 2v

Hypothesis ß hypothesis ß Hypothesis ß Hypothesis ß

Control variables
Launch proficiency .39 *** .42 *** .28 *** .30 ***
Predevelopment proficiency .20 ** .20 ** .05 .02
Environmental dynamism .02 .02 −.09 −.11 *
Innovativeness .03 −.07 .08 .06

Integration practices
1. Linear effects
Cross-functional integration .05 .00 H1b .19 ** .24 **
Customer integration −.10 H2a −.03 .09 .15 *
Interfirm integration −.36 *** −.26 *** .00 −.07
2. Nonlinear effects (squared terms)
Cross-functional integration squared H1a −.17 ** .06
Customer integration squared .14 ** H2b −.27 ***
Interfirm integration squared H3a −.15 ** H3b .19 **
R2 .35 .40 .16 .23
Δ R2 .05 .07
ΔF 8.89*** 3.1** 3.4** 3.5**

Notes:
*p b .10; **p b .05; ***p b .01.
Model 1 represents the baseline model; model 2 shows the results when simultaneously including the squared terms of the predictors of interest.
 C. Homburg, C. Kuehnl / Journal of Business Research 67 (2014) 1360–1367
Table 4
Percentage of companies meeting optimum range of integration practice.
% of companies with respect to …. …. Below optimum range
…. cross-functional integration in NPDa 25%
…. customer integration in NPDb 72%
…. interfirm collaboration in NPDa 45%
…. cross-functional integration in NSDc – –
…. customer integration in NSDa 33%
…. interfirm collaboration in NSDb 57%
Notes:
The first derivation of the each regression equation with respect to each relevant independent variable provides information on the optimum level. Accounting for the optimum level
corrected by standard deviation of each independent variable indicates the percentage of companies that are within, below, or beyond the optimum range.
a
Results indicate an inverted U-shaped relationship.
b
Results indicate a U-shaped relationship.
c
Results indicate a linear relationship. Percentage of companies meeting optimum range of integration practice cannot be computed.
needs. Consequently, manufacturing companies can benefit strongly
from integrating customers to a limited degree by getting a basic idea
of customer needs and preferences. However, to obtain genuine and
deep customer insights and capture customers' latent needs, firms
have to develop new products jointly with customers and involve cus-
tomers intensively in NPD (Moeller, 2008). In this case, high new prod-
uct success might benefit from the integration of lead users in the
development process (Von Hippel, 1986). However, as the U-shaped re-
lationship between customer integration and new product performance
suggests, manufacturers may suffer from integrating customers to a me-
dium degree in NPD since the effort to select the right set of customers
and to monitor customer behavior is disproportionally large compared
to the benefits gained. Interfirm collaboration may be beneficial to ser-
vice companies when overcoming concerns of non-patentability of ser-
vice innovations. Hence, findings show that with a low and an intense
degree of interfirm collaboration, service firms can satisfactorily co-
create value in NSD. Analogous, this research replicated the results by
Luo et al. (2007) when finding an inverted U-shaped effect of interfirm
collaboration on new product success.
Third, in line with the above discussion on the effects of three im-
portant success drivers, this study contributes to closing the research
void in service innovation literature to gain a better understanding of
NSD.
6.2. Managerial implications
This study derives three major managerial implications on how to im-
prove the development process of new services and new products. First,
practitioners should pay attention to monetary and non-monetary costs
of implementing integration practices in NPD and NSD that may out-
weigh their benefits. Thereby, the ratio of costs to benefits may differ be-
tween NPD and NSD.
Second, motivated by the findings of this study, additional anal-
yses determine the optimum level of each integration practice for
both new products and new services. These analyses demonstrate
a huge enhancement potential for companies in case of a nonlinear
relationship between an integration practice and innovation suc-
cess (see Table 4). Specifically, more than one-third of the compa-
nies in the new service and new product samples expended too
much effort, implying that they exceeded the optimum level of the
respective integration practice. Conversely, another 25% to 45% of
the companies in the sample spent insufficient effort in adequately
implementing these potential success drivers. As a result, compa-
nies may benefit from adapting the level of each integration practice
where necessary.
In detail, while managers should strive for a high degree of cross-
functional integration in NSD, a moderate level in NPD keeps in balance
the costs and benefits associated with this success driver. Yet, about
60% of the companies in the new product sample spent either too
1365
…. Meeting optimum range …. Above optimum range
40% 35%
28% –
21% 34%
–
27% 40%
43% –
much or too little effort, time, and money for its implementation. With
respect to customer integration, practitioners should moderately inte-
grate customers in NSD, while NPD managers should integrate cus-
tomers to either a low or an extensive degree to raise new product
success. Thus, about 70% of the service and manufacturing companies
could still improve efficiency of customer integration. Referring to
interfirm collaboration, approximately 60% of service companies and
about 80% of manufacturing firms in the sample could increase innova-
tion success when collaborating with business partners to a higher or
lower degree.
Third, in light of these results, practitioners should continuously
assess the level of each integration practice and correct it in case of
need. The following indicators might prove useful in this respect.
To evaluate the efficiency of cross-functional integration, this re-
search suggests questioning employees as to how much this integra-
tion practice indeed contributes to the progress of the NPD process.
For a more objective evaluation, practitioners can also keep track of
the frequency and quality of NPD teams' meetings. To obtain the
right degree of customer integration, managers could conduct an em-
ployee survey or appoint an expert team to track carefully the bene-
fits customers deliver to NSD/NPD and assess the feasibility of
customers' ideas. Service and manufacturing firms collaborating
with external partners should carefully monitor the level of trusts
and seek for those allies that offer complementary capabilities. A sur-
vey among NSD employees might be useful to determine whether
sufficient trust is shown between partners and whether an adequate
exchange of information, skills, and capabilities takes place within an
interfirm collaboration.
6.3. Limitations and avenues for future research
Future research should seek to overcome some of the limitations of
this article. One limitation stems from the static design of the study.
As prior research has shown, the importance of integration practices
may vary with the stage of the innovation process (Ernst et al., 2010).
For this reason, future research may investigate differences in the rela-
tionship between integration practices and innovation success of new
products and new services depending on the stage of the NPD/NSD pro-
cess. In addition, a more fine-grained distinction of collaborating firms,
for example referring to suppliers, universities, competitors, or organi-
zations from other industries (Knudsen, 2007), as well as service types
such as consumer and industrial services (Atuahene-Gima, 1996) may
be worthwhile to study. Finally, future research should replicate find-
ings and investigate whether nonlinear effects of further success drivers
exist in an NPD versus NSD setting.
Acknowledgment
We thank Rahel Reichmann for her support during data collection.
1366 C. Homburg, C. Kuehnl / Journal of Business Research 67 (2014) 1360–1367

Appendix A. Scale items for construct measurement.

Measures α CR AVE IR
I. Innovation success (“not at all” to “totally agree”) .94 .91 .68 –
Extent to which your company's innovation project has achieved the following innovation objectives:
▪ overall commercial success – – – .51
▪ contribution to the commercial success of the company – – – .60
▪ sales expectations relative to stated objectives – – – .84
▪ sales growth expectations relative to stated objectives – – – .89
▪ profitability expectations relative to stated objectives – – – .59
Cross-functional integration (“totally disagree” to “totally agree”) .87 .88 .71 –
Functions such as R&D, marketing, and manufacturing are tightly integrated in cross-functional teams in the innovation project. – – – .57
There is a high level of cooperation and coordination among functional units working jointly on the innovation project. – – – .93
NPD members of different functional units regularly share internal and external information regarding the innovation project. – – – .64
Customer integration (“totally disagree” to “totally agree”) .85 .86 .67 –
Customers were intensively involved in the innovation project. – – – .75
We met customers regularly in meetings arranged for the innovation project. – – – .74
We had intensive discussions with our customers about the innovation project. – – – .54
Interfirm collaboration (“not at all” to “very intensive”) .77 .79 .57 –
We collaborated for R&D with external partners. – – – .47
We collaborated for production with external partners. – – – .93
We collaborated for market launch with external partners. – – – .32

Notes: Coefficient alpha (α), composite reliability (CR), average variance extracted (AVE), item reliability (IR).

References
Alam, I. (2002). An exploratory investigation of user involvement in new service develop-
ment. Journal of the Academy of Marketing Science, 30(3), 250–261.
Anderson, E. W., Fornell, C., & Rust, R. T. (1997). Customer satisfaction, productivity, and prof-
itability: Differences between goods and services. Marketing Science, 16(2), 129–145.
Atuahene-Gima, K. (1996). Differential potency of factors affecting innovation per-
formance in manufacturing and services firms in Australia. Journal of Product
Innovation Management, 13(1), 35–52.
Atuahene-Gima, K. (2005). Resolving the capability—Rigidity paradox in new product
innovation. Journal of Marketing, 69(4), 61–83.
Avlonitis, G. J., Papastathopoulou, P. G., & Gounaris, S. P. (2001). An empirically-based typol-
ogy of product innovativeness for new financial services: Success and failure scenarios.
Journal of Product Innovation Management, 18(5), 324–342.
Bagozzi, R. P. (1994). Structural equation models in marketing research: Basic principles.
In R. P. Bagozzi (Ed.), Principles of Marketing Research (pp. 317–385). Cambridge:
Blackwell.
Bagozzi, R. P., & Yi, Y. (1988). On the evaluation of structural equation models. Journal of
the Academy of Marketing Science, 16(1), 74–94.
Bidault, F., & Cummings, T. (1994). Innovating through alliances: Expectations and limita-
tions. R&D Management, 24(1), 33–45.
Bstieler, L. (2006). Trust formation in collaborative new product development. Journal of
Product Innovation Management, 23(1), 56–72.
Bstieler, L., & Hemmert, M. (2010). Increasing learning and time efficiency in
interorganizational new product development teams. Journal of Product Innovation
Management, 27(4), 485–499.
BusinessWeek (2008). Why companies lack successful innovation.
Chan, K. W., Yim, C. K.(. B.)., & Lam, S. S. K. (2010). Is customer participation in value cre-
ation a double-edged sword? Evidence from professional financial services across
cultures. Journal of Marketing, 74(3), 48–64.
Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2003). Applied multiple regression/correlation
analysis for the behavioral sciences (3rd ed.)Mahwah, NJ: Lawrence Erlbaum.
Cuijpers, M., Guenter, H., & Hussinger, K. (2011). Costs and benefits of inter-departmental
innovation collaboration. Research Policy, 40(4), 565–575.
De Jong, J. P. J., & Vermeulen, P. A.M. (2003). Organizing successful new service development
a literature review. Management Decision, 41(9), 844–858.
De Luca, L. M., & Atuahene-Gima, K. (2007). Market knowledge dimensions and
cross-functional collaboration: Examining the different routes to product innovation
performance. Journal of Marketing, 71(1), 95–112.
Droege, H., Hildebrand, D., & Forcada, M. H. (2009). Innovation in services: Present findings,
and future pathways. Journal of Service Management, 20(2), 131–155.
Duysters, G., & Lokshin, B. (2011). Determinants of alliance portfolio complexity and its effect
on innovative performance of companies. Journal of Product Innovation Management,
28(4), 570–585.
Echambadi, R., & Hess, J.D. (2007). Mean-centering does not alleviate collinearity prob-
lems in moderated multiple regression models. Marketing Science, 26(3), 438–445.
Eisingerich, A.B., Rubera, G., & Seifert, M. (2009). Managing service innovation and
interorganizational relationships for firm performance: To commit or diversify?
Journal of Service Research, 11(4), 344–356.
Emden, Z., Calantone, R. J., & Droge, C. (2006). Collaborating for new product develop-
ment: selecting the partner with maximum potential to create value. Journal of
Product Innovation Management, 23(4), 330–341.
Ernst, H., Hoyer, W. D., & Rübsaamen, C. (2010). Sales, marketing, and research-
and-development cooperation across new product development stages: Implications
for success. Journal of Marketing, 74(5), 80–92.
Evanschitzky, H., Eisend, M., Calantone, R. J., & Jiang, Y. (2012). Success factors of product
innovation: An updated meta-analysis. Journal of Product Innovation Management,
29(S1), 21–37.
Fang, E. (2008). Customer participation and the trade-off between new product innova-
tiveness and speed to market. Journal of Marketing, 72(4), 90–104.
Fitzsimmons, J. A., & Fitzsimmons, M. J. (1994). Service management—Operations, strategy
and information technology. Boston, Mass: Irwin/McGraw-Hill.
Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with
unobservable variables and measurement error. Journal of Marketing Research,
18(1), 39–50.
Griffin, A. (1997). The effect of project and process characteristics on product development
cycle time. Journal of Marketing Research, 34(1), 24–35.
Griffin, A., & Hauser, J. R. (1993). The voice of the customer. Marketing Science, 12(1),
1–27.
Gruner, K. E., & Homburg, C. (2000). Does customer interaction enhance new product
success? Journal of Business Research, 49(1), 1–14.
Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2010). Multivariate data analysis: A
global perspective (7th ed., global ed.)NJ: Pearson: Upper Saddle River.
Hauser, J., Tellis, G. J., & Griffin, A. (2006). Research on innovation: A review and agenda
for marketing science. Marketing Science, 25(6), 687–717.
Henard, D. H., & Szymanski, D.M. (2001). Why some new products are more successful
than others. Journal of Marketing Research, 38(3), 362–375.
Hillebrand, B., & Biemans, W. G. (2003). The relationship between internal and external
cooperation: Literature review and propositions. Journal of Business Research, 56(9),
735–743.
Hillebrand, B., & Biemans, W. G. (2004). Links between internal and external cooperation
in product development: An exploratory study. Journal of Product Innovation
Management, 21(2), 110–122.
Im, S., & Nakata, C. (2008). Crafting an environment to foster integration in new product
teams. International Journal of Research in Marketing, 25(3), 164–172.
Johnsen, T., Phillips, W., Caldwell, N., & Lewis, M. (2006). Centrality of customer and sup-
plier interaction in innovation. Journal of Business Research, 59(6), 671–678.
Karniouchina, E. V., Victorino, L., & Verma, R. (2006). Product and service innovation: Ideas
for future cross-disciplinary research. Journal of Product Innovation Management,
23(3), 274–280.
Kindström, D., Kowalkowski, C., & Sandberg, E. (2013). Enabling service innovation: A
dynamic capabilities approach. Journal of Business Research, 66(8), 1063–1073.
Kirca, A. H., Jayachandran, S., & Bearden, W. O. (2005). Market orientation: A meta-
analytic review and assessment of its antecedents and impact on performance.
Journal of Marketing, 69(2), 24–41.
Knudsen, M. P. (2007). The relative importance of interfirm relationships and knowledge
transfer for new product development success. Journal of Product Innovation
Management, 24(2), 117–138.
Kratzer, J., Leenders, R. T. A. J., & Van Engelen, J. M. L. (2004). A delicate managerial chal-
lenge: How cooperation and integration affect the performance of NPD teams. Team
Performance Management, 10(1), 20–25.
Kristensson, P., Matthing, J., & Johansson, N. (2008). Key strategies for the successful
involvement of customers in the co-creation of new technology-based services.
International Journal of Service Industry Management, 19(4), 474–491.
Lievens, A., & Moenaert, R. K. (2000). New service teams as information-processing
systems: Reducing innovative uncertainty. Journal of Service Research, 3(1),
46–65.
Lui, S. S., Ngo, H., & Hon, A. H. Y. (2006). Coercive strategy in interfirm cooperation:
Mediating roles of interpersonal and interorganizational trust. Journal of Business
Research, 59(4), 466–474.
 C. Homburg, C. Kuehnl / Journal of Business Research 67 (2014) 1360–1367
Lukas, B.A., & Menon, A. (2004). New product quality: Intended and unintended consequnces
of new product development speed. Journal of Business Research, 57(11), 1258–1264.
Luo, X., Rindfleisch, A., & Tse, D. K. (2007). Working with rivals: The impact of competitor
alliances on financial performance. Journal of Marketing Research, 44(1), 73–83.
Magnusson, P., Matthing, J., & Kristensson, P. (2003). Managing user involvement in ser-
vice innovation: Experiments with innovating end users. Journal of Service Research,
6(2), 111–124.
Maltz, E., & Kohli, A. K. (1996). Market intelligence dissemination across functional
boundaries. Journal of Marketing Research, 33(1), 47–61.
Maltz, E., & Kohli, A. K. (2000). Reducing marketing's conflict with other functions: The
differential effects of integrating mechanisms. Journal of the Academy of Marketing
Science, 28(4), 479–492.
Menor, L. J., Tatikonda, M. V., & Sampson, S. E. (2002). New service development: Areas for
exploitation and exploration. Journal of Operations Management, 20(2), 135–157.
Millson, M. R., & Wilemon, D. (2002). The impact of organizational integration and product
development proficiency on market success. Industrial Marketing Management, 31(1),
1–23.
Moeller, S. (2008). Customer integration—A key to an implementation perspective of
service provision. Journal of Service Research, 11(2), 197–210.
Montoya-Weiss, M. M., & Calantone, R. (1994). Determinants of new product perfor-
mance: A review and meta-analysis. Journal of Product Innovation Management,
11(5), 397–417.
Muthén, L. K., & Muthén, B. O. (2006). Mplus, statistical analysis with latent variables, user's
guide. Los Angeles: Muthén & Muthén.
Nijssen, E. J., Hillebrand, B., Vermeulen, P. A.M., & Kemp, R. G. M. (2006). Exploring prod-
uct and service innovation similarities and differences. International Journal of
Research in Marketing, 23(3), 241–251.
O'Cass, A., Song, M., & Yuan, L. (2012). Anatomy of service innovation: Introduction to the
special issue. Journal of Business Research, 66(8), 1060–1062.
1367
Ostrom, A. L., Bitner, M. J., Brown, S. W., Burkhard, K. A., Goul, M., Smith-Daniels, V., et al.
(2010). Moving forward and making a difference: Research priorities for the science
of service. Journal of Service Research, 13(1), 4–36.
Rindfleisch, A., & Moorman, C. (2001). The acquisition and utilization of information in
new product alliances: A strength-of-ties perspective. Journal of Marketing, 65(2),
1–18.
Rindfleisch, A., & Moorman, C. (2003). Interfirm cooperation and customer orientation.
Journal of Marketing Research, 40(4), 421–436.
Schleimer, S.C., & Shulman, A.D. (2011). A comparison of new service versus new product
development: Configurations of collaborative intensity as predictors of performance.
Journal of Product Innovation Management, 28(4), 521–535.
Song, X. M., Di Benedetto, C. A., & Zhao, Y. L. (1999). Pioneering advantages in
manufacturing and service industries: Empirical evidence from nine countries.
Strategic Management Journal, 20(9), 811–836.
Szymanski, D.M., Kroff, M. W., & Troy, L. C. (2007). Innovativeness and new product suc-
cess: Insights from the cumulative evidence. Journal of the Academy of Marketing
Science, 35(1), 35–52.
Tessarolo, P. (2007). Is integration enough for fast product development? An empirical in-
vestigation of the contextual effects of product vision. Journal of Product Innovation
Management, 24(1), 69–82.
Troy, L. C., Hirunyawipada, T., & Paswan, A. K. (2008). Cross-functional integration and new
product success: An empirical investigation of the findings. Journal of Marketing, 72(6),
132–146.
Tsai, K. -H. (2009). Collaborative networks and product innovation performance: Toward
a contingency perspective. Research Policy, 38(5), 765–778.
Von Hippel, E. (1986). Lead users: A source of novel product concepts. Management
Science, 32(7), 791–805.
Walter, A. (2003). Relationship-specific factors influencing supplier involvement in cus-
tomer new product development. Journal of Business Research, 56(9), 721–733.