J PROD INNOV MANAG 2009;26:139–155

r 2009 Product Development & Management Association

Can You Hear Me Now? Communication in Virtual Product

Development Teams

Mitzi M. Montoya, Anne P. Massey, Yu-Ting Caisy Hung, and C. Brad Crisp

Globalization and technological advances are driving organizations to extend the

boundaries of new product development (NPD) teams from traditional colocated
settings to dispersed or virtual settings. Virtual NPD teams have a wide array of
information and communication technologies (ICTs) at their disposal. ICTs allow
team members to communicate and collaborate as they cope with the opportunities
and challenges of cross-boundary work. The purpose of this paper is to explore ICT
use by members of virtual NPD teams. This study presents an exploratory test and
integration of two competing perspectives of media use in virtual teams: media ca-
pacity theories and social dynamic media theories. Specifically, this paper examines
the role of task type, organizational context, and ICT type as critical contingency
variables affecting ICT use. It also examines how different patterns of ICT use
relate to individual perceptions of team performance. The findings from this study of
184 members of virtual NPD teams in three global firms suggest that communica-
tion via ICTs in virtual NPD teams is contingent on a range of factors.

Introduction every team, is a group of people who interact through

interdependent tasks guided by common purpose

G
lobalization and technological advances al-
low organizations to extend the boundaries
of new product development (NPD) teams
from traditional colocated settings to dispersed or
virtual settings (Malhotra and Majchrzak, 2004;
McDonough, Kahn, and Barczak, 2001; McDono-
ugh, Kahn, and Griffin, 1999). Today, one can easily
imagine a virtual NPD teams consisting of marketers
in the United States, software developers in India,
design engineers in Germany, manufacturing special-
ists in China, and myriad other relevant knowledge
resources in other locations. A virtual NPD team, like
Address correspondence to: Mitzi M. Montoya, Business Manage-
ment Department, Box 7229, College of Management, North Carolina
State University, Raleigh, NC 27695-7229. Tel.: (919) 515-8080.
E-mail: montoya@ncsu.edu.
 We express our appreciation to our organizational contacts and to
the Center for Innovation Management Studies at North Carolina
State University who provided access to business units and teams. We
also appreciate the support of the project managers and virtual NPD
team members who participated in this study.
(Boudreau et al., 1998; Powell, Piccoli, and Ives,
2004). However, unlike conventional teams, virtual
NPD teams work across space, time, and orga-
nizational boundaries. What makes virtual teams
historically new is the array of information and com-
munication technologies (ICTs) at their disposal.
ICTs allow members to communicate and collaborate
as they cope with the opportunities and challenges
of cross-boundary work.
For organizations, virtual teams provide a mecha-
nism to leverage local expertise at the global level as
well as to develop products that meet global needs
(McDonough et al., 1999; Schmidt, Montoya-Weiss
and Massey 2001). Research on virtual teams and the
ICTs that support them is only now emerging, espe-
cially in the NPD context (McDonough et al., 2001;
Montoya-Weiss, Massey, and Song 2001; Schmidt,
Montoya-Weiss, and Massey, 2001). Recent research
has studied virtual team inputs (e.g., design, cultural
differences), socioemotional processes (e.g., trust,
140 J PROD INNOV MANAG
2009;26:139–155
cohesion), and outputs (e.g., performance, satisfac-
tion) (for a review, see Powell et al., 2004). The focus
of this study is on the communication and task pro-
cesses of virtual NPD teams and the enabling role of
ICTs.
Virtual NPD teams must communicate, coordi-
nate, and build relationships largely—and sometimes
exclusively—via ICTs. Virtual NPD teams place
BIOGRAPHICAL SKETCHES
Dr. Mitzi M. Montoya is the Zelnak Professor of Marketing and
Innovation in the Business Management Department at North Car-
olina State University. She received her Ph.D. from Michigan State
University. Her research focuses on innovation processes and strat-
egies and the role of technology as an enabler of decision making.
Her publications have appeared in Management Science, Marketing
Science, Journal of Marketing Research, Academy of Management
Journal, and Journal of the Academy of Marketing Science, among
others. Dr. Montoya is a member of the American Marketing As-
sociation and Product Development and Management Association.
Dr. Anne P. Massey is the Dean’s Research Professor and professor
of information systems in the Kelley School of Business at Indiana
University. She received her Ph.D. in decision sciences and engi-
neering systems from Rensselaer Polytechnic Institute. Dr. Massey’s
research focuses on how information systems and technology can be
used to support individual, group, and organizational performance.
Current areas of research interest include the usability of online
service interfaces, knowledge-intensive processes, and collaborative
work. Her work has been published in leading journals including
MIS Quarterly, Academy of Management Journal, Decision Sci-
ences, European Journal of Information Systems, IEEE Transactions
on Engineering Management, and Journal of Management Informa-
tion Systems. She is a member of the Association for Information
Systems and IEEE.
Dr. Yu-Ting Caisy Hung is a research associate in the E. Philip
Saunders of College at the Rochester Institute of Technology. Pre-
viously, she was an assistant professor in the Department of Infor-
mation Systems at the National University of Singapore. She
received her B.BA. in management information systems from
Chun-Yuan Christian University in Taiwan, her M.Sc. in informa-
tion systems from the University of Maryland, and M.B. and Ph.D.
in information systems from Indiana University. Her primary re-
search focuses on virtual team interaction through various infor-
mation and communication technologies. Her specific research
areas include intercultural communication and collaboration in
global virtual teams, temporal coordination in virtual teams, and
trust in virtual teams. She has presented her work in various inter-
national conferences and has published in Journal of Management
Information Systems, Communications of the ACM, and Small
Group Research.
Dr. C. Brad Crisp is assistant professor of information systems in
the College of Business Administration at Abilene Christian Uni-
versity. Dr. Crisp earned a Ph.D. from the University of Texas at
Austin and previously worked at Indiana University. His research
examines the use and impact of information technology in educa-
tional and workplace settings with an emphasis on social processes
such as the formation of control and trust among peers in virtual
teams. His publications have appeared in the Academy of Manage-
ment Journal and the Encyclopedia of Information Systems.
M. M. MONTOYA ET AL.
increased demands on an organization’s existing
communication and collaboration systems, and in
some cases they compel further investment in ICTs
specially dedicated to supporting virtual NPD. A
recent industry report by Gartner Group (http://
www.gartner.com) indicates that the growing demand
for ICTs will drive the team software collaboration
market to nearly $700 million in 2005, with the market
expected to reach $1.1 billion by 2008. As the use of
virtual NPD teams increases, we need to understand
how team members use the plethora of ICTs available
to them so organizations can effectively support the
work of virtual NPD teams.
The purpose of this paper is to explore ICT use by
members of virtual NPD teams. The present study
presents a test and integration of two competing per-
spectives of media use in virtual teams: media capacity
theories and social dynamic media theories. Specifi-
cally, this study explores the role of task type, orga-
nizational context, and media type as critical
contingency variables affecting ICT use. Past research
has not specified the nature of the relationships
among these contingency variables. Exploratory re-
sults are presented regarding how different patterns of
ICT use relate to individual perceptions of team per-
formance. The next section discusses the theoretical
framework for understanding communication and
task processes in virtual NPD teams and presents
the research hypothesis. Then a discussion is provided
regarding the research design and method for the
study of 184 members of virtual NPD teams in three
global firms. Finally, the paper concludes with a dis-
cussion of the study’s results and implications for
practice and research.
Conceptual Background
Virtual Team Interaction Processes
Group communication research suggests that teams
perform various simultaneous functions as they work
toward goals: (1) production (work performance); (2)
team well-being (relationships among team members);
and (3) member support (relationships with others)
(Fjermestad and Hiltz, 1998; McGrath, 1991).
Though team interaction processes can be conceptu-
alized in different ways, the present study focuses
on four broad categories of communication tasks
to describe team functions: (1) conveyance; (2) con-
vergence; (3) project management; and (4) social/
CAN YOU HEAR ME NOW? COMMUNICATION IN VIRTUAL PRODUCT DEVELOPMENT TEAMS
relational (Marks, Mathieu, and Zaccaro, 2001;
McGrath, 1991).
Conveyance tasks include information exchange be-
haviors undertaken by virtual NPD team members to
convey data, information, and knowledge. Convey-
ance efforts center on sharing information, perspec-
tives, and opinions by individual team members.
Convergence tasks are characterized by decision-mak-
ing behaviors that involve team members critically
examining others’ contributions with the goal of
converging to a common understanding such that a
decision can be reached or problem solved. Conver-
gence efforts typically involve criticisms, disagree-
ments, and qualifications of members’ ideas and
positions. Project management tasks include establish-
ing schedules, budgets, and operating procedures and
managing these things over time. Project management
tasks are decision-making behaviors directly related
to the production function of a virtual NPD team.
Social/relational tasks are characterized by interper-
sonal behaviors that involve relations among team
members as well as relations between individual mem-
bers and the team. The development of relational ties
is associated with team member support and team
well-being functions. Social/relational tasks often in-
volve interactions not germane to the focal perfor-
mance task (e.g., joking around, socializing, personal
or interpersonal discussions) (McGrath, 1991; War-
kentin, Sayeed, and Hightower, 1997).
For virtual NPD teams, conveyance, convergence,
project management, and social/relational tasks are
enabled (or hindered) by ICTs. Each of these tasks
requires appropriate ICT infrastructure (Malhotra
and Majchrzak, 2004; Majchrzak et al., 2000). Robey,
Boudreau, and Rose (2000) broadly characterize ICTs
used by virtual teams into two categories: technology
that allows for (1) ‘‘same time, different place’’ (syn-
chronous) interaction and for (2) ‘‘different time,
different place’’ (asynchronous) interaction. Synchro-
nous ICTs include telephone/audio conferencing, vi-
deo conferencing, instant messaging, live application
sharing, and electronic whiteboarding. Asynchronous
ICTs include e-mail, threaded discussions, and shared
document repositories.
The present paper considers two theoretical per-
spectives that provide a foundation for understanding
the enabling role of ICTs: (1) media capacity theories;
and (2) social dynamic media theories. Media capacity
theories assume that ICTs differ in objective charac-
teristics that are invariant despite user and context of
use. They allow for prescriptive theorizing about pos-
J PROD INNOV MANAG 141
2009;26:139–155
sible relationships between technology and interac-
tion-based task behaviors. Conversely, social dynamic
media theories emphasize the social/contextual as-
pects of communication. This perspective assumes
that properties of communication environments are
not necessarily invariant but rather the properties are
experience dependent.
Media Capacity Theories
Media capacity theories describe media in terms of
objective characteristics of the medium such as rich-
ness and social presence and characteristics of the
channel, such as speed or interactivity (Boland,
Tenkasi, and Te’eni, 1994; Dennis and Valacich,
1999; Zack, 1993). These design characteristics do
not define any specific ICT or features, but rather
they express the ideals to be achieved by the commu-
nication system (Boland et al., 1994). A review of the
literature reveals that media richness, social presence,
and interactivity are highly developed, well-studied
constructs that are significantly interrelated and
entangled (Fulk and Boyd, 1991; Webster and
Trevino, 1995; Zack, 1993).
Daft, Lengel, and Trevino (1987) defined media
richness as a blend of four factors: (1) the immediacy
of feedback; (2) the use of multiple cues (verbal and
nonverbal); (3) language variety; and (4) the ability to
personalize messages. Rich media would have high
levels of each factor, and lean media would have low
levels of each. This hierarchy of richness describes the
capacity of an ICT to convey cues and to facilitate
shared understanding in a timely manner. Social pres-
ence describes the degree to which the media allows
users to feel psychologically close or present (Fulk
and Boyd, 1991; Short, Williams, and Christie, 1976).
The use of multiple, nonverbal communication chan-
nels and cues as well as continuous feedback can con-
vey social presence. Furthermore, high social presence
enables the conveyance of social influence and other
symbolic content and social context cues, whereas
those low in social presence filter out those cues.
Richness and social presence are highly related (Web-
ster and Trevino, 1995). Finally, both the richness and
social presence constructs are highly similar to the in-
teractivity construct discussed in the sociology litera-
ture (c.f. Zack, 1993, for review). The literature
identifies several key characteristics of interactive
communication that differentiate ICTs. These include
(1) the simultaneity and continuity of communication,
142 J PROD INNOV MANAG
2009;26:139–155
(2) the use of multiple, nonverbal cues, (3) the
spontaneity of involvement, (4) the ability to inter-
rupt or preempt, (5) the mutuality of constructing and
coordinating the communication process, and (6) the
patterns of turn taking.
Generally, face-to-face work (objectively charac-
terized as rich, high in interactivity, and high social
presence) sets the standard for communication and
forms the basis of comparison with ICTs. As pre-
scriptive theories, media capacity theories suggest that
proper ‘‘matching’’ of ICTs and tasks will enhance
communication (Daft et al., 1987; Dennis and Val-
acich, 1999; Russ, Daft, and Lengel, 1991). This sug-
gests that different technologies may be better suited
for different task processes. For example, asynchro-
nous technology (e.g., e-mail) may be preferred for
conveyance since not all virtual NPD team members
need to be available at the same time to share infor-
mation, nor must they agree on its meaning. Asyn-
chronous ICTs tend to be characterized as leaner and
lower on interactivity and social presence. Conversely,
since convergence requires the development of shared
meanings and agreement, synchronous ICTs and face
to face may be a better fit (Massey et al., 2001; Robey
et al., 2000) as they are relatively richer and higher in
terms of interactivity and social presence.
Although media capacity theories allow for theo-
rizing about task–ICT fit, several studies suggest that
ICTs can be perceived differently by team members in
actual use, regardless of the ‘‘objective’’ characteris-
tics of the ICT (Orlikowski and Gash, 1994). Media
capacity theories offer individual-level rational choice
explanations of use behavior and attribute effects to
objective characteristics. However, when media ca-
pacity theories are used to describe how individuals
actually use and perceive ICTs, empirical results have
been inconsistent. Media capacity theories have gen-
erally been supported when tested on traditional
media, such as face-to-face, telephone, and written
communication (e.g., Russ et al., 1991). However, in-
consistent empirical findings have resulted with newer
media, such as electronic mail and voice mail (e.g.,
Burke and Chidambaram, 1999). As a result, social
dynamic media theories have emerged to explain the
inconclusive findings.
Social Dynamic Media Theories
In contrast to media capacity theories, social dynamic
perspectives emphasize the collective character of ICT
M. M. MONTOYA ET AL.
use and the social construction of characteristics
(Fulk, Schmitz, and Steinfeld, 1990). Communication
is inherently an act that is socially and culturally sit-
uated since individuals are embedded within social
systems that influence their behaviors (Zack, 1993).
For example, organizational culture may shape indi-
vidual and team-based perceptions concerning
whether specific ICT use behaviors are supported
and expected (Schneider, 1990). Although an ICT
may be installed, a lack of training or clear use ratio-
nales may shape virtual team members’ perceptions,
consequently leading to use or nonuse. Thus, social
relationships and behavioral contexts may play a
large role in determining ICT use and perceptions re-
gardless of the apparent objective characteristics of an
ICT (Burke and Chidambaram, 1999; Carlson and
Zmud, 1999; Fulk, 1993; McGrath, 1991; Walther,
1992; Zack, 1993).
Among theories that consider the social/contextual
aspects of communication are structuration theory
(DeSanctis and Poole, 1994), social influence theory
(Fulk, Schmitz, and Steinfield, 1990), social informa-
tion processing theory (Walther, 1992), the time,
interaction, and performance (TIP) theory (McGrath,
1991), and channel expansion theory (CET) (Carlson
and Zmud, 1999). Structuration theory suggests that
communication and ICT use ‘‘shape’’ each other as
users selectively appropriate ICT features in the con-
text of evolving social structures (DeSanctis and
Poole, 1994). Social influence theory attempts to iden-
tify social psychological processes that can be used to
explain patterns of meanings and behaviors toward
ICTs (Fulk, 1993). Social information processing the-
ory asserts that some social information must be
processed in order to develop interpersonal relations
(Walther, 1992). TIP theory emphasizes the temporal
processes in team interaction and argues that teams
and the ICTs that support them cannot be disentan-
gled from the surrounding social and organizational
systems (McGrath, 1991). And finally, CET suggests
that as users gain various experiences (with messaging
topics, communication co-participants, the organiza-
tional context, and technology itself), perceptions of
ICTs evolve over time (Carlson and Zmud, 1999).
All of these perspectives are consistent in that they
emphasize the social and dynamic aspects of commu-
nication. The basic premise is that, regardless of
access to various ICTs with differing objective char-
acteristics, virtual NPD team members likely adapt
the technology to the context of use. For example,
Walther (1992) showed that through repeated use and
CAN YOU HEAR ME NOW? COMMUNICATION IN VIRTUAL PRODUCT DEVELOPMENT TEAMS
ingenuity, e-mail (characterized as lean, low in inter-
activity and social presence) could become ‘‘more
rich’’ and effective, although not necessarily as effi-
cient, in developing personal relationships. Burke and
Chidambaram (1999) similarly found that, with re-
peated use and interaction, users perceived a lean ICT
to be as effective as a richer ICT because it allowed the
users to focus on the activity at hand.
Clearly, the nature of the virtual NPD work may
well require a variety of ICTs. It is important to rec-
ognize that contemporary media capacity and social
dynamic theories all implicitly focus on only one ICT
at a time. These theories do not explicitly take into
account today’s advanced voice and data networks,
increasingly pervasive network access, integrated tech-
nologies, or integrated devices that facilitate simulta-
neous multimedia use (Massey and Montoya-Weiss,
2006). Moreover, to date, the growing body of re-
search on virtual teams has tended to focus on social
issues rather than on how virtual teams actually use
ICTs for specific tasks (i.e., conveyance, convergence,
project management, social/relational). Since organi-
zations must commit to a particular ICT infrastructure
based on some criteria, media capacity perspectives
provide prescriptive guidelines regarding the fit be-
tween ICTs and task requirements. Vendors of ICTs
often use prescriptive guidelines to market and sell
team collaboration products, simultaneously provid-
ing investment justification for organizations. How-
ever, without understanding how virtual NPD team
members actually use ICTs, relative to tasks and in the
context of the social systems they work within, these
investments may fall well short of desired returns.
This study concludes that both media capacity and
social dynamic media theories are important perspec-
tives relevant to building a deeper understanding of
ICT use in virtual NPD teams. Essentially, the two
competing theories suggest that ICT use is contingent
on different factors. Media capacity theories tend to
focus on task–ICT fit from an objective perspective,
and social dynamic theories tend to emphasize the
importance of the subjective effect of situational
factors that characterize the context of use (e.g., or-
ganizational, user, environmental factors). Thus, a
contingent relationship is expected to be found among
task type, ICT type, situational factors, and ICT use.
Distinct patterns of use of different types of ICTs are
also expected to be seen for different types of tasks,
and these patterns are expected to be related to dis-
tinct situational factors such as organizational, team,
and individual characteristics. Stated formally:
J PROD INNOV MANAG 143
2009;26:139–155
H1: ICT use by virtual NPD team members is contin-
gent on task type, ICT type, and situational fac-
tors.
H2: Distinct patterns of ICT use are related to (a)
task type and (b) situational characteristics.
Neither media capacity nor social dynamic theories
speak directly to the relationship between ICT use by
virtual teams and team performance. On the other
hand, the relationship between team communication
and project performance has been long established in
the research and development (R&D) management
and NPD literatures (e.g., Allen, Lee, and Tushman
1980; Pinto, Pinto, and Prescott, 1993; Song, Mon-
toya-Weiss, and Schmidt, 1997). Past research sug-
gests that the quality and quantity of team
communication are positively related to NPD project
success. The need for communication in NPD teams
stems from the complex interdependencies among
members of multidisciplinary teams working on de-
velopment projects (Pinto et al., 1993). The need for
communication in virtual NPD teams is exacerbated
because team members are distributed across space
and time. Stated formally:
H3: The different patterns of ICT use by virtual NPD
team members are associated with different levels
of perceived performance.
Research Design and Analysis
Sample and Survey Design
To test the hypotheses, a survey was developed to
collect data from individual virtual NPD team
members on the frequency of use of various ICTs
for specific tasks. The specific tasks included in the
survey were as follows: (1) conveyance (gathering and
exchanging work-related information); (2) conver-
gence (problem solving and decision making); (3) pro-
ject management (scheduling); and (4) social/
relational (relationship building or morale boosting).
For each task, members rated various ICTs for fre-
quency of use (five-point scale from 1 5 never to
5 5 all the time). Overall, the range of ICTs avail-
able to team members was consistent across organi-
zations. The survey included four synchronous ICTs:
telephone/teleconferencing, desktop videoconferenc-
ing, shared applications (e.g., via NetMeeting), and
instant messaging (e.g., Lotus Sametime, MSN
Messenger). It also included four asynchronous
144 J PROD INNOV MANAG
2009;26:139–155
ICTs: e-mail, voicemail, calendaring systems, and
shared document repositories. Since face-to-face com-
munication can still play a role in virtual NPD teams,
it is included in the present investigation as an addi-
tional synchronous ICT. Therefore, in total, the use of
nine ICTs were considered in this study. ‘‘Not appli-
cable’’ was included as an option should a team mem-
ber not have access to a particular ICT. The survey
was administered online and was hosted on a server at
the university of one of the authors.
Data were also collected on team characteristics
(i.e., team purpose, degree of team colocation) as well
as demographic characteristics of the respondent (i.e.,
gender, nationality, and work experience). Finally,
individual perceptions of team performance were cap-
tured from a process perspective using a six-item
Likert scale developed from past research on team
functioning and effectiveness (Cohen and Bailey,
1997; Kirkman et al., 2004). The perceived perfor-
mance scale consisted of the following items (Cron-
bach’s alpha 5 0.92):
 My team produced high-quality decisions.
 My team worked together very efficiently.
 My team worked together very effectively.
 I was satisfied with my team’s performance.
 I was satisfied with the process by which my team
made its decisions.
 I was satisfied with my team’s discussions.
Respondents scored each item on a five-point scale
(1 5 strongly disagree to 5 5 strongly agree).
The survey was administered to a sample of virtual
NPD team members from three global Fortune 500
firms known to use virtual NPD teams. Using pro-
fessional relationships, managers were contacted at
the three organizations to identify prospective busi-
ness units and teams. The first organization (Equip-
ment) is a developer and manufacturer of agriculture
and construction equipment, conducting business in
160 countries with more than 46,000 employees
worldwide. The second organization (Computing)
develops and manufactures advanced information
technologies (e.g., computer systems, software, stor-
age systems) and sells solutions and services via its
worldwide consulting business. It employs more than
329,000 people in 75 countries. The third organization
(Materials) designs and manufactures residential and
commercial flooring, ceilings, and cabinets. It has fa-
cilities in 12 countries and employs approximately
15,000 people worldwide.
M. M. MONTOYA ET AL.
Potential teams were identified for participation at
each organization. Each team was screened to ensure
that it met the definition of a virtual team (i.e., highly
interdependent with some portion of time being phys-
ically noncolocated). Following this high-level screen-
ing, 15 virtual NPD teams were identified for the
study: six teams at the equipment company, five teams
at the computing company, and four teams at the
materials company. Across the teams, the average
team size was approximately 16 members. Telephone
or face-to-face interviews also were conducted with
project managers of each team to gain a deeper
understanding of the nature of work each team was
engaged in, why they were structured as distributed
teams, and the use of ICTs by the teams. These
interviews facilitated survey development.
Project managers solicited participation from the
team members for the online survey, explaining
the purpose of the survey and that participation was
voluntary. A summary of the results was provided
to each project manager, and informal interviews
were conducted to gain additional insights into best
practices. Across the 15 teams, 184 usable surveys
were received (81, 70, and 33 from the Equipment,
Computer, and Material organizations, respectively).
Table 1 describes the sample.
MANCOVA Analysis
To test the contingency hypothesis (H1), multiple
analysis of covariance (MANCOVA) was conducted.
Specifically, differences were tested for in frequency
of use of ICTs across task type, ICT type, and orga-
nization. The MANCOVA results are reported in
Table 2. This analysis allowed us to test for mean
differences on each of the main factors (tasks, ICTs,
organizations), to explore potential interaction effects
among the main effects, and to control for the cov-
ariates team size and degree of colocation. Overall,
the results support the contingency hypothesis
(H1) that ICT use by virtual NPD team members
varies by task type, media type, and the unique situ-
ational factors represented by the three participating
organizations.
The MANCOVA results in Table 2 indicate that
the main effects for tasks, ICTs, and organizations
are all significant at the po.0001 level and all of
the two-way interaction effects are significant at the
po.001 level. The three-way interaction effect for
task  organization  ICT was not significant. Since
CAN YOU HEAR ME NOW? COMMUNICATION IN VIRTUAL PRODUCT DEVELOPMENT TEAMS J PROD INNOV MANAG 145
2009;26:139–155

Table 1. Individual and Team Characteristics (Percentages)

Characteristics Overall (n 5 184) Equipment (n 5 81) Computing (n 5 70) Materials (n 5 33)

Gender
Male 69.5 89.6 50.8 55.2
Female 30.5 10.4 49.2 44.8
Work Experience
Less than 5 years 7.8 14.3 4.3 0.0
5–10 years 11.2 14.3 5.8 15.1
11–15 years 12.8 19.5 4.4 15.2
16–20 years 13.4 6.5 15.9 24.2
21–25 years 19.6 13.0 30.5 12.2
More than 25 years 35.2 32.5 39.1 33.3
Nationality
United States 60.3 58.0 62.9 60.6
Other (17 countries) 39.7 42.0 37.1 39.4
Degree of Team Collocation
o 25% 50.0 22.2 98.6 15.2
25–50% 13.6 17.3 1.4 30.3
50–75% 8.1 8.6 0.0 24.2
76–100% 28.3 51.9 0.0 30.3

the higher-order two-way interaction effects are sig- use of ICTs by virtual teams. The significance of the
nificant, those results were interpreted instead of the task  organization and organization  ICT interac-
lower-order main effects. The significant two-way in- tion effects lend support to the perspective offered
teraction effect findings suggest situational context— by social dynamic theories that situational context is
defined in this analysis as the nature of the organiza- also relevant in explaining ICT use by virtual teams.
tion, task type, and ICT type—does in fact matter in Overall, the significant interaction effects suggest a
determining frequency of use of various ICTs. The complex interplay among the task, ICT, and usage
task  ICT interaction effect is significant at the situation as it relates to frequency of ICT use, thus
po.0001 level, thus lending high-level support to supporting H1.
the perspective offered by media capacity theories
that the fit between task and ICT type is related to
Cluster Analysis
Table 2. Multiple Analysis for Covariance (MANCOVA) To better understand the nature of the relationships
for Information and Communication Technologies (ICT) among the contingent factors and ICT use, a series of
Usage Frequency
cluster analyses was performed. Cluster analysis al-
df F Significance lows us to address H2 regarding patterns of ICT use.
The primary purpose of cluster analysis is to classify
Intercept 1 3774.125 .000
Covariates objects based on the characteristics they possess and
Team Size 1 2.326 .127 to find patterns or groups of similar objects (Milligan
Degree of Collocation 1 3.210 .073 and Cooper, 1985). The resulting clusters should
Main Effects
Organization (Org.) 2 34.776 .000 exhibit high within-cluster homogeneity and high
Task 3 115.563 .000 between-cluster heterogeneity. The analytic objective
ICT 8 358.609 .000 was to identify distinct groups of individuals with
Interactions
Organization  Task 6 3.668 .001
similar patterns of ICT usage for the four tasks (i.e.,
Organization  ICT 16 94.604 .000 conveyance, convergence, project management, so-
Task  ICT 24 20.908 .000 cial/relational behaviors). Once patterns of ICT use
Organization  Task  ICT 48 1.297 .082 were identified by task, if and how these distinct
Error 6,050
Total 6,160 groups are related to other organizational, team,
Corrected Total 6,159 and individual characteristics was examined.
R2 47.0% The cluster analysis followed the process recom-
Adjusted R2 46.0%
mended by Milligan and Cooper (1985). For each of
146 J PROD INNOV MANAG
2009;26:139–155
the four tasks, the clustering variables were the usage
frequencies reported for the nine ICTs. The Euclidean
distance measure was used as a measure of similarity.
Based on a review of major clustering methods (ibid.),
a hybrid method was chosen. For each task, a cluster
analysis was run using a hierarchical algorithm (using
Ward’s method). This resulted in clustering solutions
ranging from two to six clusters for each task. To de-
termine the appropriate number of clusters to select
for each task, the dendrograms and cluster profiles
were closely examined. Based on the number of clus-
ters identified, another cluster analysis was then run
using a nonhierarchical (K-means) algorithm. The re-
sulting cluster centroids on each clustering variable
should show significant statistical differences between
clusters. The analyses suggested a four-cluster solu-
tion for the project management task and three-clus-
ter solutions for the conveyance, convergence, and
social/relational tasks.
The results of the cluster analysis by task are pre-
sented in Tables 3, 4, 5, and 6. For each task, section
(a) of the tables include (1) the clustering variable
profiles (ICT use) for each ICT type, (2) analysis of
variance (ANOVA) tests of mean differences in ICT
use across clusters, (3) an ANOVA test of mean
differences in perceived performance across clusters,
and (4) pairwise tests of mean differences across clus-
ters for ICT use and perceived performance. For each
task, section (b) of the tables reports the demographic
cluster composition with chi-square tests of differ-
ences across clusters. Demographic cluster composi-
tion is described in terms of organizational
membership, percentage of colocated team members,
work experience, gender, and nationality.
Overall, the cluster profile results presented in sec-
tion (a) of Tables 3–6 indicate distinct cluster solu-
tions for each task, where each cluster represents a
distinct pattern of ICT use. The significant F-values
for all ANOVA tests on the clustering variables (ICT
usage frequency) in section (a) of Tables 3–6 indicate
significant mean differences across clusters in ICT us-
age frequency. Overall, the cluster analysis findings
suggest distinct patterns of use of different types of
ICTs according to task type; thus, it is concluded that
H2a is supported.
The chi-square tests presented in section (b) of
Tables 3–6 indicate that the patterns of ICT use are
related to situational characteristics. Specifically, for
each task type, differences were tested for in organi-
zational membership, degree of colocation, work
experience, gender, and nationality across the
M. M. MONTOYA ET AL.
cluster solutions, which represent patterns of ICT
use. The chi-square results reported in section (b) of
Tables 3–6 indicate a significant relationship ( po.05)
between pattern of ICT use and organizational mem-
bership and degree of collocation for all four tasks.
Work experience is significantly related ( po.05) to
the pattern of ICT use only for the convergence task.
Gender and nationality are significantly related
( po.05) to the pattern of ICT use for all tasks ex-
cept convergence. In general, the results for the chi-
tests of relationships among the situational character-
istics and patterns of ICTs by task provide evidence in
support of H2b.
Finally, the ANOVA tests for differences in mean
perceived performance scores across clusters are re-
ported in section (a) of Tables 3–6. H3 hypothesized
that the different patterns of ICT use would be asso-
ciated with different levels of perceived performance.
The results indicate significant differences (at the
po.05 and po.10 level) in perceived performance
across clusters for all four tasks. Thus, the results
suggest support for H3.
Discussion of Findings
The following sections discuss the cluster analysis re-
sults by task. The results of the MANCOVA analysis
(Table 2) clearly indicate a contingent relationship, so
the discussion is focused on a deeper exploration of
the distinct patterns of ICT use by task. Tables 3–6
highlight the nature of the patterns of ICT use as
related to key contingent factors.
Project Management Task
For project management, Table 3a shows four distinct
patterns of ICT use by members of virtual NPD
teams. Overall, clusters 1, 2, and 4 demonstrate a
more diverse use of both synchronous and asyn-
chronous ICTs (albeit in different ways) than team
members in cluster 3, who rely largely on two asyn-
chronous ICTs: e-mail and electronic calendars.
Interestingly, clusters 1 and 2 demonstrate a signifi-
cantly higher use of instant messaging (IM) than
either clusters 3 or 4. Though this is a relatively new
ICT, this result suggests that IM is being integrated
into organizational use.
The results in Table 3a also indicate that members
of virtual NPD teams in cluster 1 have significantly
CAN YOU HEAR ME NOW? COMMUNICATION IN VIRTUAL PRODUCT DEVELOPMENT TEAMS J PROD INNOV MANAG 147
2009;26:139–155

Table 3. Cluster Solution: Project Management Task

(a) Cluster Profile

Cluster Mean
ICT Usage ANOVA
Frequencies 1 2 3 4 F-Value Significant Pairwise Comparisons

Face-to-Face 3.44 2.27 2.86 3.82 24.377 [1,2] [2,3] [2,4] [3,4]
DVC 1.28 1.05 1.53 3.17 53.883 [1,4] [2,3] [2.4] [3,4]
Telephone 4.17 4.07 2.41 4.07 40.118 [1,3] [2,3] [3,4]
IM 3.61 4.45 1.38 1.30 149.053 [1,2] [1,3] [1,4] [2,3] [2,4]
Shared Apps 1.67 3.37 1.42 3.75 78.649 [1,2] [1,4] [2,3] [3,4]
Email 4.67 4.17 3.67 4.64 12.320 [1,3] [2,3] [2,4] [3,4]
Voicemail 3.56 2.65 1.61 3.17 29.753 [1,2] [1,3] [2.3] [2,4] [3,4]
Calendar 4.00 3.07 3.59 3.65 2.138
Repository 2.29 4.51 1.39 3.94 120.878 [1,2] [1,3] [1,4] [2,3] [2,4] [3,4]
Perceived 4.21 3.88 3.72 3.66 3.187 [1,3] [1,4]
Performance
Cluster Size 18 46 64 56
 p 5 .10.
 p 5 .05.
 p 5 .01.

(b) Cluster Composition

Cluster

1 2 3 4
Situational Characteristics n 5 18 n 5 46 n 5 64 n 5 56

Organization Equipment 0 4 37 40
Chi-Square 5 107.438 Computing 11 42 16 1
df 5 6 Materials 7 0 11 15
p-value 5 .000
% Collocated o 25% 11 41 26 14
Chi-Square 5 53.068 25%–50% 3 4 6 12
df 5 9 51%–75% 0 0 6 9
p-value 5 .000 4 75% 4 1 26 21
Work Experience o 5 years 0 2 9 3
Chi-Square 5 23.757 5–10 years 3 4 5 8
df 5 15 11–15 years 0 4 11 8
p-value 5 .069 16–20 years 4 9 8 3
21–25 years 7 10 10 8
4 25 years 4 17 19 23
Gender Male 7 21 47 41
Chi-Square 5 16.939 Female 8 21 12 10
df 5 3
p-value 5 .001
Nationality United States 16 24 43 28
Chi-Square 5 74.140 Non-United States 2 22 21 28
df 5 54
p-value 5 .025

higher perceptions of team performance than
members in clusters 3 and 4. Overall, the results sug-
gest that although the use of an array of synchronous
and asynchronous ICTs is positively related to per-
ceived performance (as evidenced when comparing
clusters 1 and 3), the type of ICT is also important (as
evidenced by comparing clusters 1 and 4). Interest-
ingly, although all clusters used electronic calendars
and there was no significant difference between the
clusters, the results illustrate that the complexity of
project management likely warrants use of multiple
ICTs. In terms of cluster composition, Table 3b
148 J PROD INNOV MANAG M. M. MONTOYA ET AL.
2009;26:139–155

Table 4. Cluster Solution: Conveyance Task

(a) Cluster Profile

Cluster Mean
ICT Usage ANOVA
Frequencies 1 2 3 F-Value Significant Pairwise Comparisons

Face-to-Face 3.21 3.97 2.48 39.820 [1,2] [1,3] [2,3]

DVC 1.58 3.03 1.00 86.443 [1,2] [1,3] [2,3]
Telephone 2.96 3.81 4.03 17.235 [1,2] [1,3]
IM 1.19 1.19 3.82 224.406 [1,2] [1,3] [2,3]
Shared Apps 1.85 3.61 2.74 34.225 [1,2] [1,3] [2,3]
Email 3.64 4.51 4.03 14.332 [1,2] [1,3] [2,3]
Voicemail 1.85 2.80 2.61 10.177 [1,2] [1,3]
Calendar 1.36 2.51 1.94 14.083 [1,2] [1,3] [2,3]
Repository 1.74 3.52 4.17 80.821 [1,2] [1,3] [2,3]
Perceived 3.75 3.67 3.96 2.837 [2,3]
Performance
Cluster Size 47 73 64
 p 5 .10.
 p 5 .05.
 p 5 .01.

(b) Cluster Composition

Cluster

1 2 3
Situational Characteristics n 5 47 n 5 73 n 5 64

Organization Equipment 28 51 2
Chi-Square 5 141.247 Computing 9 0 61
df 5 4 Materials 10 22 1
p-value 5 .000
% Collocated o 25% 16 14 62
Chi-Square 5 89.512 25%–50% 7 17 1
df 5 6 51%–75% 5 10 0
p-value 5 .000 4 75% 19 32 1
Work Experience o 5 years 5 6 3
Chi-Square 5 14.737 5–10 years 4 11 5
df 5 10 11–15 years 9 11 3
p-value 5 .142 16–20 years 5 8 11
21–25 years 5 12 18
4 25 years 16 23 24
Gender Male 31 54 31
Chi-Square 5 10.748 Female 9 15 27
df 5 2
p-value 5 .005
Nationality United States 26 46 39
Chi-Square 5 59.847 Non-United States 21 27 25
df 5 36
p-value 5 .008

indicates that members of clusters 1 and 2 are the Conveyance Task

most dispersed (virtual), whereas clusters 3 and 4 may
be characterized as hybrid in terms of the degree of
collocation. In concert, the degree of colocation and
organizational membership appear to shape ICT
usage for project management.
For conveyance, Table 4a shows three distinct pat-
terns of ICT use. Members of cluster 1 are least di-
verse in terms of the ICTs used for conveyance
activities, relying most on e-mail and face to face
CAN YOU HEAR ME NOW? COMMUNICATION IN VIRTUAL PRODUCT DEVELOPMENT TEAMS J PROD INNOV MANAG 149
2009;26:139–155

Table 5. Cluster Solution: Convergence Task

(a) Cluster Profile

Cluster Mean
ICT Usage ANOVA
Frequencies 1 2 3 F-Value Significant Pairwise Comparisons

Face-to-Face 2.21 3.02 4.19 70.874 [1,2] [1,3] [2,3]

DVC 1.08 1.77 2.33 14.823 [1,2] [1,3] [2,3]
Telephone 2.79 4.10 3.84 27.349 [1,2] [1,3]
IM 1.70 3.61 1.36 72.820 [1,2] [2,3]
Shared Apps 1.51 3.14 2.06 27.999 [1,2] [1,3] [2,3]
Email 2.69 3.78 3.99 24.796 [1,2] [1,3]
Voicemail 1.41 2.16 2.34 8.441 [1,2] [1,3]
Calendar 1.03 2.44 1.44 23.619 [1,2] [2,3]
Repository 1.95 4.03 1.52 137.604 [1,2] [1,3] [2,3]
Perceived 3.92 3.93 3.62 3.970 [1,3] [2,3]
Performance
Cluster Size 39 64 81
 p 5 .10.
 p 5 .05.
 p 5 .01.

(b) Cluster Composition

Cluster

1 2 3
Situational Characteristics n 5 39 n 5 64 n 5 84

Organization Equipment 14 18 49
Chi-Square 5 87.162 Computing 24 44 2
df 5 4 Materials 1 2 30
p-value 5 .000
% Collocated o 25% 28 51 13
Chi-Square 5 67.641 25%–50% 2 4 19
df 5 6 51%–75% 2 2 11
p-value 5 .000 4 75% 7 7 38
Work Experience o 5 years 1 5 8
Chi-Square 5 20.045 5–10 years 0 8 12
df 5 10 11–15 years 5 2 16
p-value 5 .029 16–20 years 6 7 11
21–25 years 9 14 12
4 25 years 16 26 21
Gender Male 23 35 58
Chi-Square 5 3.957 Female 10 23 18
df 5 2
p-value 5 .138
Nationality United States 27 36 48
Chi-Square 5 45.438 Non-United States 12 28 33
df 5 36
p-value 5 .135

(both more traditional means of communication).
Conversely, clusters 2 and 3 demonstrate more di-
verse and different uses of both synchronous and
asynchronous ICTs. Focusing on clusters 2 and 3,
significant differences can be seen in use for all ICTs,
except for telephone and voicemail. Specifically, mem-
bers of cluster 3 indicate a significantly greater use of
IM and document repositories than members of clus-
ter 2. Conversely, cluster 2 indicates greater usage of
face-to-face and synchronous ICTs—desktop video-
conferencing and shared applications—as well as
e-mail and electronic calendars. Repositories are gen-
erally designed to enable the conveyance or sharing of
project-related documents among team members.
150 J PROD INNOV MANAG M. M. MONTOYA ET AL.
2009;26:139–155

Table 6. Cluster Solution: Social/Relational Task

(a) Cluster Profile

Cluster Mean
ICT Usage ANOVA
Frequencies 1 2 3 F-Value Significant Pairwise Comparisons

Face-to-Face 2.45 3.65 3.12 11.182 [1,2] [1,3] [2,3]

DVC 1.70 1.22 1.23 4.713 [1,2] [1,3]
Telephone 3.79 4.13 2.39 70.331 [1,3] [2,3]
IM 4.21 3.16 1.15 106.373 [1,2] [1,3] [2,3]
Shared Apps 3.30 1.32 1.20 99.248 [1,2] [1,3]
Email 3.69 3.66 2.81 14.963 [1,3] [2,3]
Voicemail 2.19 2.68 1.45 28.877 [1,2] [1,3] [2,3]
Calendar 2.82 1.23 1.23 43.406 [1,2] [1,3]
Repository 4.18 1.83 1.25 123.381 [1,2] [1,3] [2,3]
Perceived 4.05 3.87 3.66 3.875 [1,3]
Performance
Cluster Size 29 62 89
 p 5 .10.
 p 5 .05.
 p 5 .01.

(b) Cluster Composition

Cluster

1 2 3
Situational Characteristics n 5 29 n 5 62 n 5 89

Organization Equipment 9 7 61
Chi-Square 5 64.435 Computing 20 35 15
df 5 4 Materials 0 20 13
p-value 5 .000
% Collocated o 25% 24 40 25
Chi-Square 5 35.592 25%–50% 2 7 16
df 5 6 51%–75% 1 4 10
p-value 5 .000 4 75% 2 11 38
Work Experience o 5 years 0 3 10
Chi-Square 5 12.342 5–10 years 5 6 9
df 5 10 11–15 years 3 6 14
p-value 5 .263 16–20 years 5 10 8
21–25 years 5 17 13
4 25 years 9 20 32
Gender Male 15 29 68
Chi-Square 5 19.774 Female 11 28 12
df 5 2
p-value 5 .000
Nationality United States 13 45 52
Chi-Square 5 54.086 Non-United States 16 17 37
df 5 36
p-value 5 .027

Moreover, they allow for ‘‘different time, different Convergence Task

place’’ access by dispersed team members. Similarly,
IM can be deployed for relatively quick conveyance
interactions (i.e., ‘‘question and answer’’) among team
members. Thus, cluster 3’s greater usage of these ICTs
may have led to more focused conveyance efforts and
positive effects on perceived team performance.
For convergence, Table 5a indicates three distinct
patterns of ICT use. Overall, the results suggest that
face-to-face and more traditional ICTs—telephone
and e-mail—still play key roles with regard to con-
vergence-related communications. Compared with
CAN YOU HEAR ME NOW? COMMUNICATION IN VIRTUAL PRODUCT DEVELOPMENT TEAMS
clusters 1 and 2, members of cluster 3 appear to be
complementing the convergence efforts with signifi-
cantly higher usages of IM, shared applications, and
repositories. The results in Table 5a indicate that clus-
ter 3 had significantly lower perceptions of team per-
formance than members of clusters 1 or 2, but there
was no significant difference in perceived performance
between clusters 1 and 2. This suggests that multiple
communication methods that can lead to positive per-
ceptions of performance. In terms of cluster compo-
sition, Table 5b indicates that clusters 1 and 2 are the
most dispersed, likely explaining why face-to-face use
is significantly less than cluster 3, whose composition
is a hybrid mix with regard to degree of collocation.
As noted earlier, the convergence task is the only task
for which work experience is significantly related to
the pattern of ICT use, whereas gender and national-
ity are not significantly related. Perhaps not surpris-
ingly, this suggests that experience is a key factor in
convergence-related communications.
Social/Relational Task
For social/relational tasks, Table 6a shows three dis-
tinct patterns of ICT use. The results suggest that, in
additional to more traditional means, IM is emerging
as a way to advance the development of team rela-
tionships. Table 6a indicates that members of cluster 1
had significantly higher perceptions of team perfor-
mance than cluster 3. The results suggest that use of
multiple ICTs with varying characteristics appears to
be positively related to perceived performance. In
terms of cluster composition, Table 6b indicates that
members of clusters 1 and 2 were the most dispersed,
with cluster 3 a hybrid in terms of the degree of col-
location, explaining cluster 3’s significantly greater
use of face-to-face interaction for development of
social/relational ties.
Implications for Practice
Computing and networking technology continue to
evolve at an astounding rate. In today’s business
environment, balancing travel costs against the need
to fully leverage global resources are top-of-mind
concerns for NPD managers. This section discusses
the practical implications of the present study for
firms deploying virtual NPD teams.
J PROD INNOV MANAG 151
2009;26:139–155
Guidelines for Practice: Work Practices
This study’s results illustrate that simply making ICTs
available does not necessarily lead to use. Past re-
search has shown that usefulness and ease of use drive
intentions to adopt any new technology (e.g., Davis,
1989; Venkatesh and Davis, 2000). Organizations
must work to engender positive attitudes toward ex-
isting or new ICT that will lead to use. Mechanisms
such as training and use testimonials can be used to
clearly illustrate the usefulness of a new ICT (e.g., IM)
to a virtual NPD team. Actively influencing attitudes
is particularly pertinent when the ICT is intended to
replace (or complement) traditional means that team
members are more familiar and comfortable with.
Importantly, when attempting to engender positive
attitudes, training programs will need to go beyond
knowing the technical nuances (i.e., ‘‘how it works’’)
of any given ICT. Instead, virtual NPD team mem-
bers need to know why an ICT is useful and when it is
most appropriate to use relative to specific tasks and
development projects.
As both the use of virtual NPD teams and the array
of available ICTs continue to grow, so does the need
to manage communication and collaboration in this
environment. Along these lines, three guidelines are
offered for good practice that may facilitate more
effective ICT deployment, thus leading to more pos-
itive perceptions about virtual NPD team perfor-
mance (Duarte and Snyder, 2006). First, members of
virtual NPD teams should establish standards for
availability and acknowledgement of communica-
tions. Specifically, these standards address when and
how dispersed team members will be available for
collaboration and how quickly they will respond to
messages. By agreeing to how quickly members will
respond to messages, this may eliminate the need for
so-called ‘‘assault’’ communications when members
appear nonresponsive. For example, one team in the
study recently agreed that all e-mail-based communi-
cation would be acknowledged within 24 hours. Stan-
dards around communication can also provide
foundation for establishing trust among virtual
NPD members, essential to successful completion of
virtual NPD team projects (McDonough et al., 2001).
Second, it is important to recognize that synchro-
nous communication —even for virtual NPD teams—
plays an important role, particularly for convergence
and social/relational tasks. Despite challenges, syn-
chronous communication is regularly used by virtual
teams, especially at the time of project kick-off
152 J PROD INNOV MANAG
2009;26:139–155
(Boutellier et al., 1999; Malhotra and Majchrzak,
2004; McDonough et al., 1999). It helps build strong
relationships, trust, and commitment to the team, and
it enables the resolution of differences, problem solv-
ing, and decision making. When budgets and time
constraints of travel make face-to-face infeasible (par-
ticularly for the whole team), members of virtual
NPD teams should be encouraged to use the full
range of synchronous ICTs to enable as close a rep-
lication as possible to face-to-face communication.
These results illustrate that when NPD members are
highly dispersed they tend to use an array of synchro-
nous ICTs (i.e., IM, telephone, shared applications) to
replace face-to-face interaction, thus creating ‘‘func-
tional equivalents’’ to face to face. Moreover, ICTs
can be used in combination to further expand func-
tionality of any one media (e.g., audio conferences can
be combined with a shared application via NetMeet-
ing or a document repository to review and edit pro-
ject-related documents by dispersed team members).
In addition, when synchronous ICTs are infeasible or
difficult, one option is to designate at least one mem-
ber of the virtual NPD team as the project liaison
across dispersed sites. For example, at the Materials
organization, a team designated one member to serve
as the liaison, tasking this member with periodic
travel to the various dispersed sites. This approach
can address some budgetary constraints and help
drive relationship-building benefits for all dispersed
members.
Finally, virtual NPD team members should be
encouraged to agree on a plan of ICT use and have
senders take responsibility for prioritizing communi-
cations. Employees are inundated with numerous
messages on a daily basis (e.g., e-mail, voicemail,
IM pings). Informational messages, requests for ac-
tion, request for meetings, and so forth all end up in
the inbox. A project manager at the Materials orga-
nization somewhat jokingly suggested his title was ‘‘e-
mail manager’’ as he spent the first hour of every day
sorting through hundreds of e-mail communications
to determine what messages were relevant (or not) to
his NPD team project activities and what was urgent
or required immediate attention. A mutually agreed
on plan of ICT use and standards helps establish how
members will use various ICTs for various tasks as
they consider what is appropriate and necessary. Pri-
oritizing use of ICTs and even messages within a
specific ICT can facilitate task processes. For exam-
ple, after multiple instances of important/urgent
messages falling through the ‘‘e-mail crack,’’ teams
M. M. MONTOYA ET AL.
at the Materials organization agreed that when mem-
bers were not available (due to time differences or
other commitments), for urgent/time sensitive mes-
sages, voicemail would be used. To make prioritiza-
tion efforts work, senders must take responsibility by
carefully considering the urgency of the message, the
desired response time frame, and which team mem-
bers need to be engaged. Developing mutually agreed
on standards for ICTs requires consideration of access
issues and local constraints. Although many organi-
zations today are moving to globally common tech-
nology platforms and network infrastructures, some
organizations still have incompatible infrastructures
that preclude the use of some ICTs by all virtual team
members.
Selecting ICTs to Support Virtual NPD Teams
For organizational decision makers who may be se-
lecting ICTs to support virtual NPD teams, it is im-
portant to understand the nature of the NPD process
and the needs of those working in the process. This
research highlights key interrelationships to take into
consideration when making decisions about collabo-
rative system features and capabilities.
The results regarding the complex interactions
among task type, ICT type, and situational factors
suggest that there is not one ‘‘right’’ approach for en-
abling virtual NPD work. However, some generaliza-
tions can be drawn from the patterns of the cluster
analysis results, follow-up interviews, and past re-
search. These generalizations are presented in Table
7. Generalizations such as these can be useful, but it is
important to note that they do not always hold. For
example, lean media can become richer as people
work together over time. Decision makers should
carefully consider whether a particular ICT generally
supports the way the virtual NPD team members, in-
dividually or collectively, carry out the NPD process.
Team members may be fine working with multiple
ICTs simultaneously, whereas others prefer to work
with one ICT at a time to limit their sense of chaos.
For example, one team at the Equipment organization
used teleconferencing along with a shared computer
application to review documents ‘‘live’’. When pre-
sented with this option, another team at Equipment
suggested they would prefer to review the documents
on their own time, as they felt using the shared ap-
plication required increased coordination and would
‘‘distract from the conversation.’’ ICTs intended to
CAN YOU HEAR ME NOW? COMMUNICATION IN VIRTUAL PRODUCT DEVELOPMENT TEAMS
Table 7. ICT Task–Fit Generalizationsa
ICT Best Task–Fit
Telephone  Interaction that requires real time give and
take
 Complex problem solving
 Interpersonal issues
 Interaction that requires multiple individuals
(conference calling)
 Addressing team effectiveness and
functioning
Video  Interaction that requires real time give and
Conferencing take
 Interaction that requires multiple individuals
 Complex problem solving
 Addressing team effectiveness and
functioning
 Interaction that requires visual social cues
Voicemail  Information sharing
 Scheduling and project management
E-mail  Non-real-time information sharing with
entire team or subgroup
 Project updates
 One-to-one or one-to-many communications
 Searchable archives of contributions
Document  Non-real-time information sharing with
Repositories entire team or subgroup
 Sharing documents or other information
 Information storage and retrieval
 Different time information sharing with the
entire team
 Project updates
 Collecting comments
a
Developed from follow-up interviews and Duarte and Snyder (2006).
support virtual NPD work should allow for variance
in individual and collective work styles. Unfortu-
nately, the work environment that many people are
faced with today often includes a hodgepodge com-
bination of independent tools and technologies. De-
cision makers should carefully consider existing
organizational practices and NPD processes, includ-
ing interoperability limitations and the need for train-
ing on ICT use.
Firms must also plan for growth of the team, in-
cluding beyond the enterprise, when selecting ICTs to
support virtual NPD work. Although the present
study has focused attention on the internal virtual
NPD team, it is a growing trend for NPD work to
include suppliers and other external partners, some of
whom may require equal access to the core NPD team
and project-related information at certain points in
the development process. As the needs of the virtual
NPD team are considered, it may be useful for deci-
sion makers to look for integration potential with
other collaboration tools. Newer aggregation technol-
ogies (e.g., Web-based portals) can provide a single
J PROD INNOV MANAG 153
2009;26:139–155
point of entry to access information and other indi-
viduals. Unified interfaces offer not only convenience
but also opportunities for customization and they en-
able seamless integration across a range of tools so
there are no barriers to switching between ICTs or
using ICTs in concert. In addition to portals, seamless
integration of ICTs is available with newer technolo-
gies like MSN Messenger or Groove, which allow us-
ers to start audio conversations or video-conferencing
sessions, to chat, to send files, to share applications
and whiteboards, and even to send messages to mobile
devices. Integrated ICTs make it easier for virtual
NPD members to deploy specific capabilities relative
to the task at hand rather than having to switch be-
tween disparate systems.
Finally, it is also important to consider and plan
for desired levels of access for the core NPD project
team, the extended team/enterprise (if relevant), and
visitors. Security and identity management are neces-
sary features of an ICT, particularly if there is sensi-
tive data or information that should be accessible only
by select team members. Project managers at the
Equipment organization, for example, noted that ac-
cess management was particularly important for de-
velopment projects where suppliers were involved in
the NPD process at certain points. This also suggests
the importance of planning ahead for how the team’s
content will be managed and shared.
In the end, when selecting ICTs to support virtual
NPD teams decision makers must remember that
technology is simply the medium and not the mes-
sage of virtual work. Although ICT features and ca-
pabilities allow firms to access, embed, and transfer
knowledge, the real challenge lies within the complex
interplay among content, context, and the individuals
who pull the NPD process pieces together. As these
results suggest, virtual NPD members tend to display
different patterns of use depending on the task and
situational characteristics. It may be that the nature of
virtual work is as much a function of organizational
processes and culture—and thus as slow to change—
than it is simply an issue of making collaborative ICT
systems available. This is not meant to imply that
virtual NPD work cannot be changed, but rather it
suggests a slow process of change.
Future Directions and Conclusions
Given the increasing use of virtual NPD teams,
coupled with the proliferation of ICTs intended to
154 J PROD INNOV MANAG
2009;26:139–155
support them, the need for research to understand
how and why teams use ICTs is essential. Without this
understanding, organizations may continue to sup-
port or invest in ICTs that do not effectively support
the work of virtual NPD teams. These results indicate
a complex task–context–ICT use contingency. This
study’s results suggest that virtual NPD team mem-
bers use a wide range of available ICTs, including
face-to-face communication as well as ICTs specifi-
cally designed to support virtual work. This study
provides insight into the patterns of ICT use where
multiple ICTs are available simultaneously. Although
the subjects were not randomly assigned to treatments
(meaning causal statements cannot be made), the re-
sults do provide better understanding of how virtual
NPD team members use ICTs to support different
tasks and some of the drivers of the differences.
Importantly, the study bridges the perspectives of
media capacity theories and social dynamic media
theories. Some of the findings are consistent with the
prescriptive ‘‘fit’’ perspective of media capacity theo-
ries. For example, media capacity theories would say
that synchronous ICTs are the best fit for convergence
tasks. But the results also clearly indicate different
patterns of ICT use where rather unexpected combi-
nations of ICTs are used to accomplish convergence
tasks in highly dispersed teams when certain synchro-
nous ICTs are not available options. The study dem-
onstrates how social dynamic media theories can
explain the divergences from the predictions of media
capacity theories. The rational mapping of the ‘‘ideal’’
match between task and ICT characteristics pre-
scribed by media capacity theories simply cannot al-
ways be achieved in practice. Unlike most studies of
virtual teams, the study reflects the reality that virtual
NPD team members can and do use multiple ICTs in
combination to achieve the functional equivalency of
the ‘‘ideal.’’ Social dynamic theories help us under-
stand why there are different patterns of ICT use
within and across tasks.
Future research is needed in several areas to further
understanding of how virtual NPD teams work and
how to best enable effective and efficient communi-
cations. Future research is needed to assess objective
outcomes, including project success and productivity.
A relationship was observed between the patterns of
ICT use and members’ perceptions of team perfor-
mance from a process perspective. But this study did
not relate ICT use to project outcomes, which could
be defined in terms of market outcomes (sales, profit,
share) or productivity outcomes (budget, time to mar-
M. M. MONTOYA ET AL.
ket, efficiency). For example, it may be that different
patterns of ICT use are more efficient. Some past re-
search on group support systems (Fjermestad and
Hiltz, 1998) suggests that various means of commu-
nication have process losses/gains associated with
them in terms of work accomplishment. Future re-
search is needed to explore the relative efficiency and
effectiveness of alternative patterns of ICT use.
Future research is needed to investigate how differ-
ent patterns of ICT use are related specifically to
stages in the product development process and types
of projects (e.g., newness of the product or newness of
the market). Though four broad categories of tasks
were included that describe typical behaviors of an
NPD team, future research is needed to relate these
behaviors to a generalized NPD process and project
characteristics to understand differences in communi-
cation behaviors and ICT use. It may well be that
there is an even more complex contingency at play,
including task, context, NPD process stage, degree of
newness of the project/market, and ICT use.
Future research should also assess the durability of
patterns of ICT use over time and generalizability
across types of organizations. For example, there may
be other factors or richer definitions of situational
factors to consider such as language differences, team
leadership, organizational culture, and organizational
structure/strategy that may differentially influence
patterns of ICT use (Massey et al., 2001; McDono-
ugh et al., 1999). These results indicate that national-
ity and gender are significant contingent factors. But
the nature of these data did not allow for examination
of why culture or gender affects ICT use. Also, the
results suggest that organizational membership mat-
ters, but this study did not capture detailed charac-
terization of the firms. Future research should
characterize organization types and assess the gener-
alizability of these findings across organization type.
In conclusion, this study provides some insight into
the ICT practices of virtual NPD teams. Proactively
dealing with the challenge of moving beyond tradi-
tional ways of collaborating is essential to leveraging
dispersed knowledge resources and capitalizing on
ICT investments.
References
Allen, T.J., Lee, D., and Tushman, M. (1980). R&D Performance as a
Function of Internal Communication, Project Management, and
the Nature of the Work. IEEE Transactions on Engineering Man-
agement 27:2–11.
CAN YOU HEAR ME NOW? COMMUNICATION IN VIRTUAL PRODUCT DEVELOPMENT TEAMS
Boland, R., Tenkasi, R., and Te ’eni, D. (1994). Designing Information
Technology to Support Distributed Cognition. Organization Sci-
ence 5(3–4):456–475.
Boudreau, M., Loch, K., Robey, D., and Straub, D. (1998). Going
Global: Using IT to Advance the Competitiveness of the Virtual
Transnational Organization. Academy of Management Executive
12(4):120–128.
Boutellier, R., Gassman, O., Macho, H., and Roux, M. (1999). Man-
agement of Dispersed Product Development Teams: The Role of
Information Technologies. R&D Management 28(1):13–25.
Burke, K. and Chidambaram, L. (1999). How Much Bandwidth Is
Enough? A Longitudinal Examination of Media Characteristics
and Group Outcomes. MIS Quarterly 23(4):557–580.
Carlson, J.R. and Zmud, R.W. (1999). Channel Expansion Theory and
the Experiential Mature of Media Richness Perceptions. Academy
of Management Journal 42(2):153–170.
Cohen, S.G. and Bailey, D.E. (1997). What Makes Teams Work:
Group Effectiveness Research from the Shop Floor to the Execu-
tive Suite. Journal of Management 23(3):239–290.
Daft, R., Lengel, R., and Trevino, L. (1987). Message Equivocality,
Media Selection, and Manager Performance Implications for
Information Systems. MIS Quarterly 11(3):355–366.
Davis, F.D. (1989). Perceived Usefulness, Perceived Ease of Use, and
User Acceptance of Information Technology. MIS Quarterly
13(3):319–340.
DeSanctis, G. and Poole, M.S. (1994). Capturing the Complexity in
Advanced Technology Use: Adaptive Structuration Theory. Orga-
nization Science 5(2):121–147.
Duarte, D. and Snyder, N. (2006). Mastering Virtual Teams: Strategies,
Tools, and Techniques that Succeed. San Francisco: Jossey-Bass.
Fjermestad, J. and Hiltz, S. (1998). An Assessment of Group Support
Systems Experimental Research: Methodology and Results. Jour-
nal of Management Information Systems 15(3):7–149.
Fulk, J. (1993). Social Construction of Communication Technology.
Academy of Management Journal 36(5):921–950.
Fulk, J. and Boyd, B. (1991). Emerging Theories of Communication in
Organizations. Journal of Management 17(2):407–446.
Fulk, J., Schmitz, J., and Steinfield, C.W. (1990). A Social Influence
Model of Technology Use. In: Organizations and Communication,
ed. J. Fulk, and C. Steinfield. Newbury Park, NJ: Sage Publica-
tions, 117–140.
Kirkman, B., Rosen, B., Tesluk, P., and Gibson, C. (2004). The Impact
of Team Empowerment on Virtual Team Performance: The Mod-
erating Role of Face-to-Face Interaction. Academy of Management
Journal 47:175–192.
Majchrzak, A., Rice, R., Malhotra, A., King, N., and Ba, S. (2000).
Technology Adaptation: The Case of a Computer-Supported Inter-
organizational Virtual Team. MIS Quarterly 24(4):569–600.
Malhotra, A. and Majchrzak, A. (2004). Enabling Knowledge
Creation in Far-Flung Teams: Best Practices for IT Support
and Knowledge Sharing. Journal of Knowledge Management
8(4):75–88.
Marks, M.A., Mathieu, J.E., and Zaccaro, S.J. (2001). A Temporally
Based Framework and Taxonomy of Team Processes. Academy of
Management Review 26(3):356–376.
Massey, A.P. and Montoya-Weiss, M. (2006). Unraveling the Tempo-
ral Fabric of Knowledge Conversion: A Model of Media Selection
and Use. MIS Quarterly 30(1):99–114.
Massey, A.P., Montoya-Weiss, M., Hung, C., and Ramesh, V. (2001).
Global Virtual Teams: Cultural Perceptions of Task-Technology
Fit. Communications of the ACM 44(12):83–84.
J PROD INNOV MANAG 155
2009;26:139–155
McDonough, E.F., Kahn, K.B., and Barczak, G. (2001). An Investi-
gation of the Use of Global, Virtual, and Collocated New Product
Development Teams. Journal of Product Innovation Management
18(2):110–120.
McDonough, E.F., Kahn, K.B., and Griffin, A. (1999). Man-
aging Communication in Global Product Development Teams.
IEEE Transactions on Engineering Management 46(4):375–
386.
McGrath, J. (1991). Time, Interaction, and Performance (TIP): A The-
ory of Groups. Small Group Research 22(2):147–174.
Milligan, G.W. and Cooper, M.C. (1985). An Examination of Proce-
dures for Determining the Number of Clusters in a Data Set. Psy-
chometrika 50(2):159–179.
Montoya-Weiss, M., Massey, A.P., and Song, M. (2001). Getting It
Together: Temporal Coordination and Conflict Management in
Global Virtual Teams. Academy of Management Journal
44(6):1251–1262.
Orlikowski, W.J. and Gash, D.C. (1994). Technological Frames: Mak-
ing Sense of Information Technology in Organizations. ACM
Transactions on Information Systems 12(2):174–207.
Pinto, M.B., Pinto, J.K., and Prescott, J.R. (1993). Antecedents and
Consequences of Project Team Cross-Functional Communication.
Management Science 39:1281–1298.
Powell, A., Piccoli, G., and Ives, B. (2004). Virtual Teams: A Review of
Current Literature and Directions for Future Research. Database
for Advances in Information Systems 35(1):6–36.
Robey, D., Boudreau, M., and Rose, G.M. (2000). Information Tech-
nology and Organizational Learning: A Review and Assessment of
Research. Accounting, Management, and Information Technology
10(1):125–155.
Russ, G., Daft, R., and Lengel, R. (1991). Media Selection and Man-
agerial Characteristics in Organizational Communications. Man-
agement Communication Quarterly 4(2):151–175.
Schmidt, J.B., Montoya-Weiss, M., and Massey, A.P. (2001). NPD
Decision-Making Effectiveness: A Comparison of Individuals,
Face-to-Face Teams, and Virtual Teams. Decision Sciences
32(4):575–600.
Schneider, B. (1990). Organizational Culture and Climate. San Fran-
cisco: Jossey-Bass.
Short, J.A., Williams, F., and Christie, B. (1976). The Social Psychol-
ogy of Telecommunications. New York: Wiley.
Song, X., Montoya-Weiss, M., and Schmidt, J. (1997).
Antecedents and Consequences of Cross-Functional Cooperation:
A Comparison of Marketing, R&D, and Manufacturing
Perspectives. Journal of Product Innovation Management 14(1):
35–47.
Venkatesh, V. and Davis, F.D. (2000). A Theoretical Extension of the
Technology Acceptance Model: Four Longitudinal Field Studies.
Management Science 46(2):186–204.
Walther, J. (1992). Interpersonal Effects in Computer-Mediated Inter-
action. Communication Research 19(2):52–90.
Warkentin, M.E., Sayeed, L., and Hightower, R. (1997). Virtual Teams
versus Face-to-Face Teams: An Exploratory Study of a Web-Based
Conference System. Decision Sciences 28(4):975–996.
Webster, J. and Trevino, L. (1995). Rational and Social Theories as
Complementary Explanations of Communication Media Choices:
Two Policy-Capturing Studies. Academy of Management Journal
36(6):1544–1572.
Zack, M. (1993). Interactivity and Communication Mode Choice in
Ongoing Management Groups. Information Systems Research
4(3):207–261.