r Academy of Management Journal
2020, Vol. 63, No. 1, 96–123.
FOSTERING POSITIVE RELATIONAL DYNAMICS:
THE POWER OF SPACES AND INTERACTION SCRIPTS
MICHAEL Y. LEE
INSEAD
MELISSA MAZMANIAN
University of California, Irvine
LESLIE PERLOW
Harvard Business School
Despite well-accepted understanding that relational dynamics characterized by respect,
openness, and connectedness are critical for healthy team functioning, we know little
about how to foster such dynamics. Drawing on observation and interview data from an
intervention that fostered positive change in the relational dynamics of a global distributed
team, this paper theorizes the mechanisms that enabled a move toward positive
relational dynamics. We find that the intervention brought about relational changes by
not only creating spaces where the team could experiment with new forms of interaction,
but also by utilizing “interaction scripts”—concrete guidelines for interaction
that specify content parameters and participation rules. We establish that the combination
of spaces and interaction scripts was critical for helping the team enact counternormative
forms of interpersonal sharing that led to the emergence of positive relational
dynamics. While existing research has highlighted the importance of spaces for enabling
positive relational change, this paper theorizes the complementary role that interaction
scripts can play in the change process. These findings have implications for research on
positive relationships at work, organizational change, and global and geographically
dispersed teams.
How team members relate to one another affects
how teams perform. A large body of research has
highlighted how relational dynamics characterized
by respect, openness, and connectedness can foster
team creativity (Carmeli, Dutton, & Hardin, 2015),
facilitate team learning (Edmondson, 1999), build
resilience in the face of failure (Stephens, Heaphy,
Carmeli, Spreitzer, & Dutton, 2013), and generate
higher performance (Carmeli et al., 2015; Edmondson,
1999). Yet, we know surprisingly little about how
teams can change relational dynamics in order to foster
these positive outcomes.
In the present paper, we explore how teams can
cultivate and develop what we label “positive relational
dynamics,” or patterns of interaction characterized
by respect, openness, and connectedness.
Using meeting observations and interviews from a
10-week intervention, we analyze how a globally
distributed team moved from interactions characterized
by mistrust, minimal communication, and
disconnection to interactions characterized by respect,
openness, and connectedness.
The intervention had team members set aside
dedicated time each week to meet (both as a full
group and in dyads). In addition, the intervention
provided concrete guidelines for team members to
engage in interactions designed to promote sharing
about personal lives and current work challenges.
Leveraging work that has established the importance
of spaces, or bounded social settings, for enabling
social and organizational change (Bucher & Langley,
2016; Howard-Grenville, Golden-Biddle, Irwin, &
Mao, 2011; Kellogg, 2009; Polletta, 1999), and
drawing on the notion of organizational scripts,
we conceptualize the intervention as consisting of
both spaces and a type of script that we call an “interaction
script.” Interaction scripts are concrete
guidelines for interaction that specify content parameters
and participation rules for interaction. Our
The authors would like to thank Jennifer Howard-
Grenville and three anonymous reviewers for their helpful
comments and excellent guidance during the reviewprocess.
The authors gratefully acknowledge the Division of Research
at the Harvard Business School for providing financial support
for this research.
96
Copyright of the Academy of Management, all rights reserved. Contents may not be copied, emailed, posted to a listserv, or otherwise transmitted without the copyright holder’s express
written permission. Users may print, download, or email articles for individual use only.
observations suggest that the combination of the
spaces and interaction scripts of the intervention
enabled individuals on the team to experiment with
new and socially risky forms of interpersonal sharing
and was key to fostering positive relational dynamics
between team members.
We unpack how the combination of spaces and
interaction scripts engendered new ways of relating
built on respect, openness, and connectedness. In
the early weeks of the intervention, we observed the
team enacting counter-normative forms of personal
and work-related sharing, guided by the interaction
scripts and supported by spaces. In the latter weeks
of the intervention, we observed the team enacting
the scripts with more disclosure and honesty while
simultaneously adapting the scripts in ways that
furthered the development of positive relational
dynamics. We theorize how spaces and scripts work
together to facilitate change: while spaces create
an opening for new dynamics to emerge, interaction
scripts specify and legitimate new forms of interpersonal
sharing that can foster the emergence of
positive relational dynamics.
RELEVANT LITERATURE
Positive Relational Dynamics in Teams
Wedraw on the literature on positive relationships
at work to conceptualize the changes in the team
we observed (e.g., Dutton & Ragins, 2007). While no
universally shared definition exists as to what constitutes
positive relationships at work, a large corpus
of scholarship has emphasized the importance of
respect (Carmeli et al., 2015; Dutton, 2003), openness
(Eisenberg & Witten, 1987; Rogers, 1987), and connectedness
(Dutton & Heaphy, 2003; O’Reilly,
Caldwell, & Barnett, 1989).
“Respectful” interactions refer to those actions
that confer a sense of value and worth to others
(Carmeli et al., 2015: 68). Substantial scholarship has
documented the lack of respect in workplace interactions,
such as incivility, abusive supervision,
public criticism, or rudeness (Andersson & Pearson,
1999; Liu, Liao, & Loi, 2012; Sutton, 2007). Respectful
interactions, on the other hand, place value
on others’ perspectives by listening empathically
(Dutton, 2003), fostering inclusivity (Blatt &
Camden, 2007; Nembhard & Edmondson, 2006),
and affirming and supporting others (Carmeli
et al., 2015). In addition to promoting greater
feelings of worth, respectful interactions have
been linked to positive identity formation (Rogers,
Corley, & Ashforth, 2016), reduction of status differences
(Phillips, Rothbard, & Dumas, 2009), increased
group creativity (Carmeli et al., 2015), and
faster error detection (Vogus, 2004; Weick, 1993).
“Openness” in interactions refers to candid and
frank communication of issues and feelings at work
and encompasses both task-related and personal dimensions
(Eisenberg & Witten, 1987). Task-related
openness involves the sharing of work issues, especially
across hierarchical lines. Scholars have investigated
this phenomenon across a number of
streams of research, including employee voice
(Morrison, 2011), issue selling (Dutton, Ashford,
O’Neill, & Lawrence, 2001), superior–subordinate
communication (Jablin, 1979), and critical upward
communication (Tourish & Robson, 2006). Personal
openness can range from engaging in casual social
interactions in the workplace (Bowen & Blackmon,
2003) to disclosing personally sensitive information
or feelings (Creed, 2003; Phillips et al., 2009). While
personal openness at work does not assume intimate
disclosure, it does presume some level of personal
sharing, social interactions, and emotional expression.
Openness at work has been shown to have positive
impacts on a variety of collective and individual
level outcomes. For instance, openness in taskrelated
communication has positive impacts on
learning and improvement (Detert & Burris, 2007;
Dutton et al., 2001; Edmondson, 1999), decisionmaking
(Morrison & Milliken, 2000), group performance
(Edmondson, 1999;O’Reilly&Roberts, 1977),
and individual health (Cortina, 2008). Openness and
disclosure of personal matters has also been shown
to improve interpersonal relationships (Collins &
Miller, 1994) and to reduce status and power differentials,
and, in doing so, enhance feelings of empowerment
(Ashcraft, 2000; Phillips et al., 2009).
“Connectedness” refers to the strength and quality
of relationships between individuals. Dutton and
Heaphy (2003) conceptualized the quality of a connection
as a function of the amount and range of
emotions expressed, the ability for a relationship to
withstand stress and strain, and its ability to open
new possibilities for action and creativity. These
scholars view connectedness as manifesting in feelings
of mutuality, positive regard, and vitality. Earlier
research on groups studied connectedness under
the guise of constructs such as social integration
and cohesion, which was defined as the degree to
which individuals in a group are “psychologically
linked” or “attracted” to one another (O’Reilly et al.,
1989; Shaw, 1981). Benefits of connectedness include
better health (Dutton & Heaphy, 2003), the
2020 Lee, Mazmanian, and Perlow 97
attainment of valued information and resources
(Burt, 2000), development of positive new identities
(Ibarra, 1993), and improved coordination (Gittell,
Seidner, & Wimbush, 2010).
While respect, openness, and connectedness are
three distinct dimensions of what we label “positive
relational dynamics,” these dimensions have
been shown to be deeply intertwined and mutually
reinforcing. For example, respectful interactions
between group members encourage speaking up
(Nembhard & Edmondson, 2006) and lead to feelings
of greater connectedness (Carmeli et al., 2015). Open
and frequent communication about work and personal
matters can foster greater attraction between
group members (Collins & Miller, 1994) and serve as
a sign of mutual regard (Carmeli et al., 2015). Feelings
of connectedness enable people to speak openly
(Carmeli & Gittell, 2009; Edmondson, 1999) and are
the foundation for respectful interacting (Dutton &
Heaphy, 2003). Scholars of positive relationships at
work emphasize the cyclical and self-reinforcing
nature of interactions characterized by respect, openness,
and connectedness: “Positive relationships in
groups and communities are created through ongoing,
self-perpetuating, and mutually reinforcing acts that
both offer and generate positive energy among their
members” (Kahn, 2007: 281).
Despite the self-reinforcing nature of positive relational
dynamics, they can be difficult to foster.
Status and power differences that pervade organizations
can undermine mutual respect and encourage
individuals to regard others as “less than”
(Galinsky, Magee, Inesi, & Gruenfeld, 2006; Kipnis,
1972). Speaking openly about personal or workrelated
matters requires navigating complex social
norms governing when and how to communicate
such information and can lead to social sanctions
when such norms are violated (Chaikin & Derlega,
1974; Detert & Edmondson, 2011; Morrison, 2011).
Connecting with others at work often involves overcoming
dominant, institutionalized norms of impersonality
(Ashcraft, 2000; Martin, Knopoff, & Beckman,
1998) and organizational politics that foster competition
rather than cooperation (Mayes & Allen, 1977).
While research has explored the antecedents of
positive relational dynamics (e.g., Baker & Dutton,
2007; Nembhard & Edmondson, 2006; Vogus, 2004),
there is strikingly little empirical work on how positive
relational dynamics emerge in the course of
everyday team interactions. Several scholars have
noticed this lack and called for empirical research
to explore this question (Dutton & Ragins, 2007;
Edmondson & Lei, 2014). The research presented
here takes up this challenge by providing empirical
insight into the microdynamics of change as one
team began to develop positive relational dynamics.
In theorizing this empirical story, we draw on two
different analytical constructs: spaces and scripts.
Spaces as a Vehicle for Relational and
Organizational Change
One construct that has been shown to play an important
role in organizational change, including relational
dynamics in groups, is the notion of
“spaces.” Scholars studying how to foster effective
group dialogue emphasize the importance of creating
spaces that can enable people to actively listen,
respect one another, suspend judgments, and
speak honestly (Bohm, 1990; Isaacs, 1999; Senge,
1990). Referred to as “containers,” “holding
spaces,” or “vessels,” such spaces are conceptualized
as social settings separated by physical and
symbolic boundaries from everyday work engagements
where “creative transformation can
take place” (Isaacs, 1999).
Spaces, or bounded social settings, have been
shown to play an important role in fostering change
in organizations. In explaining why one hospital
successfully implemented an institutional mandate
around reducing resident hours and another hospital
did not, Kellogg (2009) theorized the importance of
“relational spaces” in enabling potential reformers to
mobilize, connect, and develop strategies to enact
organizational change. In another study of changing
institutionalized logging practices in the forestry
industry, Zietsma and Lawrence (2010) emphasized
the creation of “experimental spaces.” These spaces
allowed representatives with different goals and
perspectives (timber companies, local residents, and
environmentalists) to try out new ways of working
shielded from institutional pressures. In a study
of routine change made at two hospitals, Bucher
and Langley (2016) found that the existence of
two different types of spaces—reflective and
experimental—enabled hospitals to implement
new complex medical routines. Furnari (2014) theorized
the role of “interstitial spaces”—or social settings
where members of different institutional fields come
together to interact—in facilitating the emergence of
new practices. Finally, in a study of a successful
initiative to bring a sustainability focus into a forprofit
enterprise, Howard-Grenville and colleagues
(2011) theorized the importance of liminality in
enabling the development and incorporation of new
cultural repertoires into the dominant organizational
98 Academy of Management Journal February
culture. Their notion of liminality, a process they label
as “bracketing the everyday,” closely echoes the
concept of spaces in other studies.
Thus, spaces have been identified as important
enablers of a variety of types of change in and across
organizations. Throughout these studies, spaces are
characterized by their separation from team, organizational,
or institutional norms and dominant
modes of working. This separateness comes from
physical, temporal, social, and/or symbolic boundaries,
and is theorized to reduce the salience of
existing norms and patterns of interaction while
fostering reflection, experimentation, and risk taking
(Bucher & Langley, 2016; Furnari, 2014; Zietsma &
Lawrence, 2010).
However, while spaces have been shown to enable
new dynamics to emerge, understanding why and
how particular dynamics emerge in spaces remains
under-theorized. To highlight this point, Furnari
(2014) argues that, while spaces can support the
emergence of new practices, they cannot, on their
own, explain the emergence of new practices. Furnari
(2014: 441) wrote:
The very same features of interstitial spaces that facilitate
the initial emergence of new activities and
ideas hinder their constitution into new practices . . .
the inherently transitional nature of interstitial spaces
means that the new activities and ideas that are tried
out in these settings may easily “get lost,” thereby
making difficult their repetition over time and the
formation of the shared meanings that are necessary
for them to be constituted into new practices.
Polletta (1999) makes a similar argument about
the role of spaces in the social movements literature.
She contends that, while scholars have highlighted
the important role of “free spaces” in enabling social
change, “the free space concept simply posits a
‘space’ wherein those dynamics occur, without
specifying how, why, and when certain patterns of
relations produce full-scale mobilization rather than
accommodation or unobtrusive resistance” (Polletta,
1999: 8).
To better understand how the intervention facilitated
the emergence of the relational dynamics that
we observed, we further draw on the construct of
scripts.
Scripts in Organizational Studies
Borrowing from the field of social cognition,
scholars have used the notion of “scripts” to explain
how people develop a shared understanding around
how to act and behave in organizations. According to
these theorists, scripts are internalized cognitive
schemas that direct individuals as to the appropriate
behavior for specific organizational situations or
contexts (Abelson, 1981; Ashforth & Fried, 1988;
Gioia & Poole, 1984; Lord & Kernan, 1987). According
to this work, scripts exist in common interactive
processes such as formal meetings or board presentations
as well as task processes such as production
or go-to-market routines. Scripts can be
either weak or strong, depending on how much detail
is specified in the script (Gioia & Poole, 1984;
Lord & Kernan, 1987). For example, a weak script
might specify expectations of behaviors, and a strong
script would specify both expectations of behavior
and also the sequence of events (Gioia&Poole, 1984).
Scripts serve several functions that are relevant to
these data. First, scripts legitimate behaviors within
organizations. The content of scripts aligns with the
expectations of those in power and thus reinforces
norms around acceptable behavior (Ashforth &
Fried, 1988). Scripts also provide a guide of what
behavior is appropriate in a given situation, helping
newcomers to become socialized into an organization’s
way of doing things (Ashforth & Fried, 1988;
Gioia & Poole, 1984). Finally, scripts conserve cognitive
capacity by supporting automatic enactment
of desired behaviors (Ashforth & Fried, 1988).
We find the concept of scripts to be useful for understanding
the microdynamics of change observed
in this study. Not only did the intervention create
spaces for new behaviors to emerge, but it also included
concrete guidelines for how interactions
should proceed within the spaces of the intervention.
These guidelines functioned like a script,
providing parameters around the content and sequence
of the interactions and reducing the uncertainty
and risk associated with interactions that
were counter-normative for the team.
While we found scripts to be a useful lens for understanding
the change process that we studied,
there are important distinctions between the scripts
provided by the intervention and the notion of
scripts found in the organizational literature. First, in
contrast to the traditional understanding of scripts
that operate in repetitive situations and socialize
people into dominant organizational norms, the
scripts in our study were used in novel situations and
served as a source of experimentation and change.
Second, whereas scripts in the literature are often
taken-for-granted and understood implicitly by organizational
members, scripts in our study were explicit,
communicated as instructions throughout the
2020 Lee, Mazmanian, and Perlow 99
intervention. Third, while organizational scripts
center around task-based routines, such as annual
review processes or manufacturing workflows, the
scripts in our study focused exclusively on guiding
interactions. As a result, we distinguish the scripts in
our intervention from the notion of scripts in the
existing organizational literature by referring to them
as “interaction scripts.”1
RESEARCH SETTING
We studied a technology consulting team that
worked in the health care industry within the technology
consulting arm of TaxCo (a pseudonym), a
large professional services firm employing over
100,000 people. This distributed global team was
responsible for building the data warehousing and
reporting functions for a client headquartered in the
Northeastern United States.
The team itself consisted of six people in the
United States and eight people in India. One
U.S. team member was based at the client site for the
duration of the project. Another U.S. team member
was based in California and rarely traveled to the
client site. The remaining four U.S. team members
worked remotely from three different cities in the
Northeast and traveled every two to three weeks to
the client site. The eight members in India were
based in four different offices throughout India. Four
Indian team members were based in Bangalore, two
were based in Delhi, one in Mumbai, and one in
Hyderabad.
In addition to the senior partner, the team in the
United States also included a partner, a senior
manager (who was the overall project lead), and
three managers. The team in India comprised one
senior manager, one manager, and six engineers.
As is evident from the team organizational chart
(Figure 1), a clear hierarchical divide existed between
the team members in the United States and
those in India. The Indian team members, aside from
the senior manager, were more junior and responsible
for most of the coding and testing work.
The U.S. team members were predominantly managers,
responsible for client communication, project
management, technical design, and oversight.
The senior partner overseeing the team volunteered
the team to participate in an intervention
to improve team effectiveness. For two weeks
prior to the intervention, we conducted on-site interviews
in the United States and India to understand
the issues facing the team. The initial focus of the
intervention was to targetwork–life challenges on the
team, and it quickly became clear during the first two
weeks of fieldwork that “work–life challenges”meant
very different things for the U.S. team members than
it did for the Indian team members. The U.S. team
members questioned the commitment and work
ethic of their Indian colleagues, and they feared that,
if they were not available to help, their Indian colleagues
would sit idle. As a result, the U.S. team
members felt overwhelmed by the volume ofwork and
responsibility to always be available. In contrast, the
Indian team members’ discontent stemmed from
feeling unappreciated and underutilized. They complained
that their U.S. managers gave them only lowlevel
work, did not provide adequate client exposure,
and generally treated them as doers rather than thought
partners. Thus, U.S. team members felt that their
work–life challenges stemmedfrom never being able to
disconnect from work,while Indian teammembers felt
that theirworkwasn’t satisfying and this dissatisfaction
translated as a “work–life challenge”.
In response to these insights, the interventionwas
established. The intervention aimed to help team
members work more effectively together so that
they could share work in new ways. The hope was
that, if the Indian team members could share more
of the workload, the U.S. members would not be so
overwhelmed and the Indian members would feel
more satisfied in their work. The intervention consisted
of two components: “collaborative work
time” (“CWT”) calls and weekly “pulse check”
meetings.
CWT
The first component of the intervention involved
weekly one-on-one calls between junior and senior
team members. These calls were structured to help
the team members get to know each other better as
humans and colleagues, with the hopes of sparking
new forms of collaboration. Team members were
instructed to spend the first 15 minutes of their onehour
call getting to know each other personally. The
remaining 45 minutes of the call were to be spent
1 We also distinguish our notion of script from the
Goffmanian notion of script. For Goffman and others,
scripts are the recurrent activities and patterns of action
characteristic of a given setting that provide the basis by
which individuals reproduce the institutional and interactional
order (Barley & Tolbert, 1997; Goffman, 1983).
Unlike Goffman’s concept of scripts, which are preexisting,
implicit, and taken for granted, interactions scripts
are novel, explicit, and not encoded cognitively.
100 Academy of Management Journal February
collaborating on a “non-routine task.” The task was
framed as work that built skills or knowledge, giving
the Indian team members exposure to more intellectually
stimulating work and giving the U.S. team
membersmore knowledgeable, better-engaged thought
partners.Most weeks, the six junior team members, all
in India, were expected to choose managers they wanted
to speak with and initiate scheduling the call.
Pulse Checks
The second aspect of the intervention asked the
entire team (i.e., all 13members) to hold a weekly, 90-
minute, mandatory team meeting. The goal of these
meetings, called “pulse checks,” was to create an
opportunity for the entire team to engage in meaningful
discussions about work challenges. To facilitate
these conversations, teammemberswere asked to
respondto a set of four questionswhile inthemeeting:
“How are you feeling?”
“How valuable is the work you are doing?”
“How satisfied are you with your learning?”
“Is your operating model sustainable?”
Facilitators provided four response options in the
form of cartoon faces that ranged from smiling to
crying. When given a prompt by the facilitator, each
team member would take turns responding to the
four questions in a round-robin format. They would
identify which face represented their feelings around
each question and provide an explanation for why
they had chosen that face. After each person provided
their ratings and explanations, others were asked to
comment. Throughout the pulse check meeting, the
discussion would evolve organically as team members
surfaced other issues and teammates tried to
address issues that were raised.
Both the CWT calls and pulse checks operated as
spaces, in that they were bounded social settings
temporally and symbolically separated from the
everyday work of the team (Bucher & Langley,
2016; Howard-Grenville et al., 2011; Kellogg, 2009;
Zietsma & Lawrence, 2010). In conjunction, both
the CWT calls and pulse check meetings contained
interaction scripts. We define interaction scripts
as guidelines for interaction that include content
parameters such as topics for discussion, conversation
prompts, direct questions to respond to, and
response options, and participation rules such
as who speaks, for how long, and in what order.
Facilitation
The four facilitators—the third author and three
other individuals—took a role that bridged research
and practice, taking extensive field notes regarding
the team’s interactions and conducting interviews,
while also helping implement the intervention. The
third author both collected data and acted as overseer
of the facilitation team. In the three early pulse
checks, one of the facilitators led the team through
the pulse check questions. In weeks 4 through 6
and in week 8, team members themselves led these
FIGURE 1
Team Structure
United States India
Partner
Sr. Manager (Project Lead)
Manager Manager Manager
Sr. Manager
Manager
2 Engineers 4 Engineers
Sr. Partner*
*Not involved in day-to-day operations of the team.
2020 Lee, Mazmanian, and Perlow 101
discussions and facilitators were present to observe
the conversation. After the first week, one facilitator
was based in India, spending a week at a time at each
of the four different locations where the Indian team
members worked. The other three facilitators spent
time on site at the U.S. location, with at least one of
them on site each time the U.S. team collocated
(every two to three weeks).
METHODS
Data Collection
The facilitation team served as the data collection
team, gathering qualitative data over the course of 12
weeks (two weeks prior to the intervention plus the
10-week intervention). All members of the facilitation
team were trained in ethnographic data collection
by the third author, who also oversaw all aspects
of data collection. Short interviews (generally lasting
15–30 minutes) were conducted each week with
most members of the team and were scheduled based
on team member availability. Interviews were semistructured
and included questions such as “How are
you feeling about your work?” and “What was your
experience this week with the intervention, both in
the one-on-oneCWTcalls and in the previous week’s
pulse check?” In addition, individual interviews
were conducted at the end of the intervention with
each team member (lasting 30 to 60 minutes) to capture
each person’s reflections on his or her experience
with the intervention. In total, 127 interviews were
conducted during the 12 weeks of the study. Weekly
interviewswere not taped but field noteswerewritten
up immediately following each interview. The interviews
at the end of the intervention were captured
via detailed notes taken by facilitators.
In addition, facilitators observed the following
meetings: an initial team meeting in which facilitators
discussed the issues that the team was facing, 10
pulse check meetings, two small-group CWT calls in
week 7 of the intervention (described in more detail
below), and occasional work meetings that were not
directly related to the intervention (e.g., sprint retrospective
meetings). In total, 15 meetings were observed
during the course of the study. All pulse
check meetings were transcribed by facilitators
while the occasional work meetings that were observed
were captured via detailed field notes.
Finally, email communications between members
of the facilitation team and between facilitators and
team members were saved as archival data. Emails
included those sent throughout the intervention as
well as discussions between members of the facilitation
team about how the intervention was going.
Data Analysis
The facilitation team came into the research project
with an interest in work–life challenges. However,
the ways in which people’s lives outside of
work were or were not affected by the intervention
were not salient in these data. Instead,what was most
striking about the intervention was a shift in how
teammates interacted. As a result, the facilitation
team, led by the third author, focused its analysis
on communication patterns and interpersonal
dynamics within the team. They wrote two different
forms of descriptive memos in the initial phase of
analysis—detailed empirical descriptions of each
team meeting and CWT call, and an annotated
timeline of the entire team with reflections on what
happened on a week-by-week basis with key milestones
and empirical description woven into the
team overview.
At this stage, the first and second author (neither of
whom participated in the facilitation or data collection)
were brought onto the project to bring an independent
perspective to data analysis and mitigate
insider bias. Using the techniques of grounded theory
building (Charmaz, 2006; Strauss & Corbin,
1998), the first author conducted independent iterative
textual analysis and open coding of the primary
data—the interview notes, meeting transcripts, and
archival data—as well as a review of the descriptive
memos produced by the facilitation team. The goal of
this exercise was to reexamine the data with a fresh
perspective in order to understand the nature of the
changes in the team’s interpersonal dynamics and
how these changes emerged.
Basedon this analysis, the first author wrote a series
of analyticmemos and shared themwith the other two
authors. The first series of memos revisited the primary
data, in order to explore the nature of the
changes that occurred within the team. The data that
had been coded as “team changes” were reanalyzed
and dimensionalized into subthemes such as “getting
to know each other as people” and “more open discussion
of work issues.” Together, the three authors
then went back to the literature to contextualize the
changes observed in the data and iteratively honed in
on the notion of positive relational dynamics as a
general category for capturing the changes observed.
The second set of memos focused on understanding
how the changes in relational dynamics emerged
through the course of the intervention. Categories
102 Academy of Management Journal February
such as “structured conversational guidance” and
subthemes such as “asking for more specific help”
emerged through the open coding exercise. Fromthis
process, it became clear that the guidelines for interaction
that had been introduced in both the CWT
calls and pulse check meetings played a key role
in changing team dynamics. These memos also interrogated
the nature of the guidelines provided by
the intervention. Through this analysis, and conversation
between all three authors, the distinction
between “content parameters” and “participation
rules” emerged. This distinction, in parallel with
reviewing literature on organizational change and
organizational scripts,helpedushone inonspacesand
interaction scripts as key constructs for understanding
how positive relational dynamics emerged throughout
the intervention.
These insights became the cornerstone of the next
stage of our analysis in which we reanalyzed every
CWT call and pulse check meeting through the lens
of spaces and interaction scripts. This analysis, in
conjunction with examining the literature on spaces,
provided insight into the key mechanisms by which
interaction scripts and spaces facilitated the enactment
of counter-normative ways of speaking and
acting and led to the emergence of positive relational
dynamics.
In the final stage of analysis, we sought to link the
emergence of positive relational dynamics in the
early weeks of the intervention with the overall
changes that we observed by the end of the intervention.
Going back to earlier memos and the
primary data as necessary, we induced aspects of the
change process as a whole—such as the positive response
of other team members, the increased energy
and engagement of the team members, and the adaptation
of scripts—that were key to understanding
how the initial momentumof change grew over time.
These insights led directly to the development of our
theoretical model and the notion of a feedback cycle
embedded within it.
FINDINGS
As the data below suggest, when facilitators first
met the team, relational dynamics were characterized
by a lack of respect, openness, and connectedness.
The lack of respect on the team manifested as
distrust and negative assessments of one another.
U.S. team members described their Indian counterparts
as “lazy” and “unaccountable.” One U.S.
manager, who was from India originally, described
the lack of respect that the Indian team members
experienced from other U.S. team members with
even more poignant language: “The Indian team are
humans, not robots. And it is important for the
U.S. team members to understand that. [The Indian
team members] are not treated as people.”
The lack of openness on the team was evident in
communication patterns that hewed to strict hierarchical
speaking rules and a general sense that
U.S. members spoke and gave directions while Indian
team members received direction and executed
tasks. The project lead, who was based in the U.S.,
remarked that the U.S. members had not engaged the
Indian members beyond requesting concrete tasks:
“We have pushed [the Indian team members] into a
box and they are living very comfortably in it.” An
Indian engineer described a “parent–child”dynamic
between the U.S. and Indian team members in which
the U.S. members would funnel only the barest of
needed information to the Indian members. Another
Indian engineer said, “[The intervention] will give a
lot more perspective to them and to us. Right now,
there is a disconnect.”
A lack of connectedness characterized both the
relationships across the India–United States divide
and relationships within the local teams. Team
members engaged in minimal interaction beyond
what was necessary for completing the work. The
U.S. team members met on-site at the client’s headquarters
once every two or three weeks. Even during
the limited time they were together, they made little
effort to interact socially. They did not go out to
lunch or make small talk in the conference room
where they were “camped” out. The Indian team
members had similar practices. At the Bangalore site,
four team members were based in the same building
but sat on different floors and rarely interacted face to
face. They would even call into team meetings from
their individual desks. One of the engineers in Bangalore
said, “We don’t have lunch together . . . It is
simply a work relationship.” Such a dynamic was
not unusual at TaxCo. Reflecting on the social norms
governing teams at TaxCo, an Indian engineer said of
the team, “We don’t talk about the personal things,
things to get to know each other. In the onshore/
offshore model, we just talk about the work.”
In the two weeks of engagement and planning
prior to the introduction of the formal intervention,
the facilitators observed firsthand some of the dynamics
on the team. Though everyone in the team
was encouraged to speak up to provide input on the
goals of the intervention, U.S. team members dominated
the conversations. Indian team members were
willing to discuss issues they were facing in their jobs
2020 Lee, Mazmanian, and Perlow 103
when speaking one on one with facilitators, but,
when the team met together, they would remain silent,
not bringing these issues to the attention of the
team. When speaking with facilitators, Indian team
members would often express curiosity about the
U.S. team members and their personalities. In contrast,
U.S. team members expressed no interest in the
Indian team members as individuals. Instead, U.S.
members were focused on how much work their
counterparts were doing and questioned how they
spent their time.
Early Weeks of the Intervention: Trying on New
Ways of Interacting
In the section below, we describe how, in the early
weeks of the intervention, the spaces and interaction
scripts of the intervention set the stage for the team to
interact in new ways. Team members made tentative
steps toward getting to know each other personally
and airing work challenges.However, teammembers
were often uncomfortable as they enacted the interaction
scripts. There was a sense of people “trying
on” new forms of interaction in the face of perceived
social risk.
Trying on new ways of interacting in the CWT
calls: Sharing personally. As described above,
CWTcalls comprised both a space and an interaction
script. As dedicated timeswhen team members were
to interact in counter-normative ways, theCWT calls
operated as spaces that were separated temporally
and symbolically from everyday norms and patterns
of interaction. The CWT calls also contained guidelines
for interaction with content parameters—the
injunction to get to know each other personally and
collaborate on work—and participation rules—the
specification that a junior and senior team member
were to spend 15 minutes talking about personal
topics and 45 minutes on a collaborative task. Together,
the content parameters and participation
rules formed an interaction script that enhanced the
symbolic separation of the 60-minute spaces of the
CWT calls and provided direction to team members
on how to engage within these spaces.
When the concept of the CWT calls was introduced,
team members were both excited about the
prospect of having personal conversations with each
other and nervous about what such interactions
might be like. As soon as the intervention was described,
several members expressed a concern that
the instructions to “get to know each other personally”
were too vague and asked for additional support
on how to have these conversations with each
other. For example, in the first pulse check, a U.S.
manager remarked about the prospect of having the
CWT calls
I know we have been talking about the collaborative
time, but do we have any parameters or guidance on
the type of [personal] conversations we should be
having? . . . I know we are career oriented and career
driven, but I think it’s equally important to invest in
[the personal component].
Another U.S. manager said, “I think in the beginning
it would be good to have guidance on the personal
conversations.” An Indian engineer agreed with this
request, saying that “The [Indian team] is not used to
this.”
As a result of the expressed discomfort, the facilitators
sent out weekly emails in the first five weeks of
the intervention with suggested topics, in the form of
explicit conversation prompts, that people could use
to jumpstart the personal portion of theCWTcall. For
example, in the second week of the intervention, the
facilitators sent out the following list in an email to
the entire team:
In case you would like something to guide the 15
minutes of relationship building, I have included a
few questions to get you started:
• Is there a story behind your name? Does your name
mean something?
• How many siblings do you have? Are you the oldest,
middle, youngest, or only child? Who are you closest
to in your family?
• What is one thing about you that people would be
surprised to learn?
• What is the best piece of career advice you have ever
received?
In each following week, a different set of prompts
were sent out by the facilitators. In week 4, the following
email was sent:
Hi all,
As you schedule your Collaborative Work Time for
the week, here are some questions for getting to know
your collaboration partner:
• Where did you grow up?Doyou still live close to that
place?
• What did you study in university?
• How did you get into technology consulting?
Looking forward to hearing how this week’s conversations
go.
104 Academy of Management Journal February
The request for more guidance on how to have personal
conversations reveals just how counternormative
these new interactions were and how
the specificity of the guidelines for interaction supported
the enactment of these new behaviors.
The injunction to get to know each other personally
had its desired effect: after engaging inCWTcalls
for just a few weeks, people reported that they were
getting to know each other better as people, not just
as coworkers. In order to allow for the emergence of a
new interpersonal dynamic, the facilitators did not
ask to listen in to these one-on-one calls. However,
the energy and excitement that came from these
personal interactions was palpable. For example,
after her call in week 4 with the U.S. project lead, an
Indian engineer commented to one of the facilitators,
“Do you see the smile on my face?” The facilitator
described in her notes:
[The Indian engineer] said it was a great call, that they
had had a really wonderful conversation about where
they are from, their career paths, their families, Bangalore
. . . they talked a bit about the current project,
but it was mostly a chance to get to know one another.
[The engineer] was very, very happy about the call.
Similar positive sentiments were expressed across
the team. A U.S. manager said, “[The Indian engineer]
and I had a really good session. It was the first
time we really got to know each other.” After
speaking with the Indian manager in week 3, another
U.S. manager reflected how important it was to get
to know his counterpart and to learn that his wife
would be having a baby soon (which he had no idea
of until this call). Another Indian engineer had a
conversation with a U.S. manager in week 3 in which
they went beyond the original directive and chose to
spend the entire call getting to know each other.
Reflecting on the call, this engineer said with surprise,
“It’s easy and comfortable to talk about things
with [the US manager] . . . The personal connection
with people makes it easier to talk about things related
to work.” The U.S. manager described the same
conversation as oriented around “how the team is
going, our families, where we are from,” and reflected
how “the conversation was more open in
work and personal discussions.”
While the CWT calls’ interaction scripts provided
guidance to team members on how to interact within
the spaces of the calls, the scripts still allowed for
unspecified interactions to emerge. For example, the
conversations in CWT calls in the early weeks also
included discussion of people’s desires for growth
and professional development. In week 3, one of the
engineers in India reported having a “great conversation”
with a U.S. manager in which he shared that
he was interested in doing more technical work besides
documentation and the manager suggested that
the engineer could help her with data modeling. The
U.S. manager reflected on the same conversation:
“It was the first time wereally got to know each other.
I learned about his background and the technical
exposure he wants to get.” These conversations
about professional growth were perceived as valuable
by Indian and U.S. team members. The fact that
they emerged organically reveals that scripts guided
conversations without fully specifying what was
said, leaving open the possibility for unspecified
interactions to emerge that supported positive relational
dynamics.
Thus, in the first few weeks, the spaces and interaction
scripts of the CWT calls guided team
members in engaging in new forms of interaction
centered around personal sharing. Notably, the personal
sharing in the CWT calls was reciprocal, as
both junior and senior members of the team were
sharing with each other.
Trying on new ways of interacting in the pulse
checks: Sharing work challenges. Like the CWT
calls, the pulse checks comprised both a space and an
interaction script. The dedicated time set aside for
each week’s pulse check, as well as the presence of
external facilitators, signaled to team members that
what was to occur in the pulse check was separate
fromeveryday routine interactions. Interaction scripts
further enhanced the symbolic separation of the
90-minute spaces and guided the substance of the
conversation within them. Four specific questionand-
answer prompts (or content parameters) were
provided to foster sharing and encourage the discussion
ofwork challenges. The participation rules—
the round-robin format—aimed to develop a more
balanced and egalitarian speaking dynamic. Notably,
both the content and the format of these interactions
were non-normative for this team.
In the early pulse checks, we observed both junior
and senior members of the team responding to the
pulse check questions by sharing their difficulties
and work challenges. Given the “chin up,” positive
attitude often projected by the U.S. managers prior to
the intervention, the presence of negative responses
to the questions was noteworthy. An Indian engineer
who was new to both TaxCo and the project was
completely overwhelmed in trying to come up to
speed. The team, however, had no idea about her
struggles.When it was her turn to speak up in the first
pulse check, she appeared visibly emotional to those
2020 Lee, Mazmanian, and Perlow 105
in the room with her and responded to the pulse
check question about the team’s operational sustainability
by saying, “‘Learning’: there is so much
learning, I am probably stretching to my limit, I
probably can’t do this forever—I can’t sustain this
pressure for very long.” Such openness immediately
humanized the engineer and affected others on the
team. A U.S. manager turned to the facilitator sitting
next to him with clear concern in his eyes and said,
“She sounds like she is going to cry. I need to talk
to her.”
The team’s response to this disclosure was as notable
as the disclosure itself. A U.S. manager immediately
suggested that the team help take some of the
deliverables off the Indian engineer’s plate while she
comes up to speed on the technologies. Later in the
pulse check meeting, the U.S. project lead said of the
new Indian engineer’s struggles:
We need to make the expectations clear. Give her
some time not only to understand the project, but the
technology and the role. Give her some time to get
familiar—assign tasks wary of the fact she is not on the
same level.
After the pulse check, several team members took it
upon themselves to reach out to this engineer in order
to help.AU.S. manager called her to discuss how
to prioritize her time. These efforts made a real difference;
the following week, the Indian engineer
expressed that she was doing much better and that
her manager was giving her more manageable tasks:
“He is honest with me and tells me what he’s giving
methree days to dowould take him one day . . . I work
at ease, I feel no pressure, I understand what I am
doing.”
Another Indian engineer, who had worked at
TaxCo for a few years but was new to this project
team, was also struggling because the work on this
team did not align with his expertise. In his first pulse
check, he said, “‘How valuable is my work?’
‘Frowning’ to ‘crying.’ I’m not aligned with what I
need.” At the end of the meeting, the U.S. project lead
tried to allay some of his concerns:
[Directed to the Indian engineer:] You come from a
different technical practice. You are moving away
from your technology from the past and concerned
about growth. The reason we brought you here is you
came highly recommended in Oracle—we are very,
very appreciative—a lot of it you can take it back to
Oracle practice. Some of the concepts work here—it
should be a win/win. We’ll try to assign you work
that helps. In my opinion, it shouldn’t be a complete
shift.
In the second pulse check, a different Indian engineer
answered the pulse check questions with
striking honesty. She described the frustrations that
arose from sitting idle while waiting for code to be
delivered for her to test. This lack of a smooth work
stream created a great unevenness in her workload:
“How am I feeling?”: Between “accepting” and
“frowning.” This week is okay . . . the way Agile is,
my work will not be easily distributed over three
weeks. . . . “How satisfied am I with learning?”: Between
“accepting” and “frowning.” I’mlearning from
a testing point of view [but not in other ways]. It’s
not working very well for me right now. “Operating
model”: “Accepting” to “frowning.” [Two other Indian
engineers] both have their own set of responsibilities.
I don’t have a backup person or a plan in place
right now. I’m trying to bring them up to speed on
the testing process, but they have their own set of
responsibilities as well. It will take some time.
Beyond eliciting honest negative feelings about
how the work was going, which would have been
taboo prior to the intervention, the pulse check
meetings became a space where team members,
particularly managers, heard about issues facing individual
team members and the team as a whole. The
degree to which managers responded to these surfaced
issues in anempathetic and productive fashion
was notable. For example, after this engineer spoke
up about her uneven workload, the U.S. project lead
immediately began to address her concerns:
We’ll try to be creative for the next sprint and look into
the even distribution part. [The two other Indian engineers]
are new to this process, talk to [the Indian
manager], see if he can help you on any of your asks, or
any of the developers here in the United States.
Thus, the pulse checks provided a forum where team
members began to share feelings and raise issues
about how work was going. Furthermore, the pulse
checks encouraged team members, particularly senior
members, to listen empathetically and respond
in productive ways.
Indian team members were not the only ones to
share their work challenges. The degree to which
senior U.S. team members also shared their struggles
in response to the pulse check questions was striking.
In week 1, the U.S. project lead responded to the
question about his satisfaction with his learning by
saying:
[I am between] “accepting” and “frowning.” The nature
of our work is so much driven by the deadlines,
we don’t get to focus on important things. We are
106 Academy of Management Journal February
driven by the urgency factor. I need to invest in myself
. . . but that doesn’t happen for various reasons.
In the same week, the U.S. partner on the project said:
“How am I feeling?” I think I’m somewhere between
“accepting” and the “frown” . . . I got a call from a
partner on the West Coast regarding an oral presentation
Thursday morning—I’m not too sure what
I’m leading . . . I had to clear everything on my calendar
yesterday, have client conversations about it, and
put together a “response” to a client I’ve never met
before, a solution I have no experience with.
Such insight into the daily lives and challenges faced
by senior managers and partners on this project was
new for the rest of the team and humanized those in
power.
Hence, even in the early weeks, the spaces and
interaction scripts of theCWTcalls and pulse checks
enabled the team to engage in new, counternormative
forms of interpersonal sharing. Through
this sharing, team members began to see one another
as humans with lives outside of work who were also
experiencing real challenges at work. The same Indian
team members who just weeks earlier were
described as being treated as “non-human” and
“robots” were not only sharing about their lives and
their experience on the project, they were also being
heard. As the U.S. project lead said after the first
pulse check:
Some of the feedback I heard is really eye opening;
folks are overwhelmed with the volume of work and
new things coming. Agile and [the intervention]
should help them. It’s nice to get the personal aspect
from everyone. Hopefully, on the next pulse check,
there will be more “smiley” faces. I really appreciate
what the intervention is trying to do. It will really
help.
This quote suggests that the team leaders quickly
developed a positive perspective of the communication
and disclosures elicited by the intervention, a
perspective that encouraged additional sharing and
provided participants with the experience of feeling
heard.
These insights and disclosures, along with the
positive response to them, began to shift the relational
dynamics on the team. One indication of this
shift was that, in week 4, the four team members
based in Bangalore, who had previously sat on
different floors of the same office building, intentionally
moved their workspaces to be together.
Attributing the change to the growing sense of connectedness
sparked by the intervention, they chose
to cram together into two cubicles despite this being
an unusual practice for teams at TaxCo. They were so
proud of the change that they sent a picture of
themselves sitting side by side in the two cubicles to
the facilitators. The subteam benefited from colocation
both in terms of work efficiency—“It’s faster
when we’re working in the same space”—and in
terms of feeling more connected to one another. In
his week 4 pulse check, one Indian engineer from
Bangalore responded, “‘My learning’: ‘smiling,’
coming to office regularly and sitting with the
team—I’m getting to know everyone and nothing
went wrong [this week].”
Despite the emergence of respect, openness, and
connectedness in the early weeks of the intervention,
change did not happen immediately. Indian team
members still remainedmostly silent outside of their
pulse check check-ins. In working team meetings not
directly related to the intervention, Indian members
rarely spoke up except when called upon. In week 3,
a U.S. manager remarked that it took “a lot of encouragement
from [another U.S. manager]” to get
Indian members to talk in the meetings. Also, Indian
team members continued to share some work difficulties
in private conversations with facilitators but
did not bring them up in the pulse checks. For instance,
a facilitator wrote in her notes in week 2:
[The Indian manager] has mentioned that it can be
challenging to manage the Indian team members and
a full task load. Didn’t hear anything about that on the
[pulse check] call. And [an Indian engineer] has told
menumerous timeshowhe doesn’t feel like he’s using
much of his brain in his work, and yet he gave an
“accepting” face for “How satisfied are you with your
learning?”
Getting to know each other personally and sharing
work struggles across geographic and hierarchical
divides challenged team norms. Weeks into the
intervention, team members struggled to engage
openly across these boundaries, indicating the difficulty
of changing the team’s relational dynamics.
However, as we describe below, the energy and excitement
generated by the interactions in the early
weeks in the intervention supported deeper enactment
of the scripts in the latter weeks and fueled a
cycle of increasing positive relational dynamics.
Latter Weeks of the Intervention: Deeper
Enactment and Adaptation of Interaction Scripts
In this section, we describe how team members
continued to enact the interaction scripts in the
2020 Lee, Mazmanian, and Perlow 107
spaces of the intervention but with increasing honesty,
openness, and playfulness. They also began to
take ownership of the process, “working” the spaces
of the intervention by experimenting and adapting
the interaction scripts. We view these adaptations as
important in the life of the change process, as they
represented a new level of engagement in the intervention
and led to further emergence of positive
relational dynamics.
Deeper personal sharing. In the intervention’s
fourth week, an Indian manager suggested to a facilitator
that, in order to foster more engagement
from the team, the team should experiment with rotating
facilitation of the pulse check between team
members. He then nominated a particularly outgoing
Indian engineer for the job. The facilitators and the
project lead agreed to this idea, and the nominated
engineer not only agreed to lead the next pulse
check, but also asked if he could add to the script—
asking everybody to tell a funny or embarrassing
story, in addition to the four standard pulse check
questions.
Notably, the Indian engineer’s suggestion came
out of the growing connectedness he had been experiencing
since he and the other team members in
Bangalore started sitting together. In describing his
motivation, he said the team in Bangalore had been
able to joke around a bit that week since they had all
been together, and suggested, “This team is too formal.
If the whole project team were in a team room
together, maybe things would be more casual, more
fun.” This motivation is revealing on multiple levels.
First, it points to the fact that the greater connectedness
from the first weeks of the intervention inspired
teammembers to becomemore engaged and invested
in the change process. Second, it suggests that some
team members began to view the scripts as resources
they could use and adapt to further promote positive
change in the team’s relational dynamics.
The funny or embarrassing story script prompted
new levels of personal sharing within the team. Individuals
shared humorous stories about their children,
dogs, and friends. An Indian engineer started
things off with a story about having terrible handwriting.
It didn’t get a big laugh, but the facilitator
bantered with her about his own bad handwriting. A
U.S. engineer shared how her one-year-old daughter
added a lot of curiosity to her life.AnIndian engineer
told a story about a group of friends who ended up
eating a very small piece of cake together after
expecting it to feed all of them. The energy picked up
when a U.S. manager shared a story of his young
daughter, who was going to be the flower girl in his
cousin’s wedding that weekend. Everyone laughed
as he described having his “hyperactive” daughter
practice walking down the aisle, tossing plastic
Easter eggs instead of flower petals. The senior
manager in India talked about her four-year-old son
who liked to pet the dog next door. After she told him
that he must wash his hands after petting the dog, her
son decided from then on to keep his hands high
above his head around the dog, allowing the dog to
lick his entire face and body instead. “That’s how he
tries to get out of handwashing!” she exclaimed. The
Indian engineer who was facilitating added, “Have
him take a bath!” At that point, the entire team was
laughing. The same engineer then told a story from
his youth about putting a firecracker in a neighbor’s
mailbox and the trouble that he got into as a result.
People were laughing at the stories and engaging
with one another with an energy not observed in
previous pulse checks. At one point, a U.S. manager
asked an Indian team member how her back was
doing, as she had been dealing with back pain. At
another point, an engineer based in Delhi mentioned
that it was nice having another Indian engineer who
was also based in Delhi to sit next to. The engineer to
whom the comment referred responded, “Oh, that’s
sweet.” The U.S. partner gave the Indian engineer
who was facilitating a hard time about his story about
dropping a firecracker in a mailbox. There was a
palpable energy in the room from the personal
sharing and the responses to one another’s sharing.
Communication and interaction was occurring in
new ways across United States–India lines and
around the personal details of each other’s lives.
Team members commented on the pulse check afterward,
reflecting to facilitators that they enjoyed
having “more interactive” time together, that it was
good to have an “informal” piece to the check-in so
that people “start opening up,” and that it was just
plain fun.
Buoyed by the success and energy of this pulse
check, members of the project team chose to continue
to facilitate the pulse checks for the next two
weeks. They passed the baton between teammates on
the India side, and each new facilitator brought their
own addition to the pulse check. In week 5, an Indian
engineer asked each team member to pick the team
member with whom he/she would trade places with
for a week and why. In week 6, another Indian engineer
had the team play the parlor game “two truths
and a lie.”
Notably, the adapted pulse check script in week 5
did not lead to the same level of energetic and lighthearted
communication across hierarchical lines
108 Academy of Management Journal February
that emerged in the week 4 and week 6 pulse checks.
In response to the question of which team member
they would want to trade places with for a day, every
team member chose a U.S. team member. While the
interaction was not perceived or experienced as
overtly negative, it exposed the power and status
divides that persisted within the team. As the facilitator
wrote in her field notes that week, “The team
was not dynamically engaged, even with the personal
question. Everyone wanted to be a [U.S. team
member] . . . Not a soul wanted to be someone in India.”
These interactions also revealed that not all
adaptations of the interaction scripts over the 10
weeks of the intervention had the same positive effect
on the team’s relational dynamics, variation that
we explore later in this paper.
In weeks 6 and 7, the team further experimented
with the interaction scripts in ways that deepened
personal sharing. While reflecting with the facilitators
on how the intervention was going, two Indian
engineers mused over the right group size to foster
more open and connected interactions. According to
the facilitator’s field notes:
This was my last week in Bangalore, and as [two Indian
engineers] were reflecting on [the intervention],
they had the idea of having a small-group pulse
check—they thought it would be more intimate and
give people a better chance to know each other and
really talk about the issues they were facing—it was
too difficult to really get into any meaningful conversation
on the large pulse checks.
The very fact that this was a topic of reflection is
noteworthy, given the initial dynamics on this team.
These team members came up with the idea of a
“small-group CWT call,” and successfully pitched
their idea to the facilitators. In the following week,
the team put this idea into practice.
The small-group CWT calls were initiated in week
7, replacing the one-on-one CWT calls. Facilitators
split the team into two groups with a mix of U.S. and
Indian members in each group. Two Indian engineers,
with feedback from other members of the
team, developed the interaction script for the calls.
One of the two Indian engineers sent the following
email to his small group:
Hi All,
Below is the agenda of the meeting.
The CWT will run for one hour. The conversation
[will] consist of two elements:
1. Relationship building (15 minutes)
The group will spend 10 minutes talking about personal
interests. Get started by sharing about a fun
weekend adventure you have recently had.
2. Share on individual strengths and areas for growth
(45 minutes)
One team member will share about the skills and
strengths he or she brings to the table and the areas he
or she is working to develop. The other team members
will have a chance to react, respond, and share their
own impressions. The group will do this with each
person on the call. The goal of this exercise is to be
more aware of each team member’s abilities so that,
as issues arise in the work, there will be greater understanding
of who is best equipped to respond
to what issues and who can stretch in their development
areas.
As this email indicates, the small-group CWT calls
constituted spaces, in that they were temporally,
socially, and symbolically separated from daily work
interactions. In addition, they contained the elements
of an interaction script: content parameters
(getting to know each other better by sharing a recent
fun weekend adventure and sharing strengths and
weaknesses) and participation rules (half of the team
members in a small group, 15 minutes of individual
personal sharing followed by 45 minutes of sharing
on strengths and weaknesses, and each individual
shares while others respond). In essence,members of
the team adapted the boundaries of the space and
modified the interaction scripts of the weekly CWT
calls in a manner that continued to foster respect,
openness, and connectedness.
The small-group CWT adaptation led to deeper
personal sharing than we observed in the early
weeks. For example, in one of the small-group
CWT calls, everyone was asked what as children
they had wanted to be when they grew up. When a
U.S. manager said he wanted to be the Pope, an Indian
engineer responded by asking the U.S. manager
whether he was still religious. The U.S. manager
took a deep breath, clearly uncertain about how
much to share, and after hesitating said:
Yeah, I mean . . . it’s always an interesting question.
My religious denomination is Roman Catholic, but
there’s always the question of “do you agree with the
church?” And I don’t agree with the church in all
things, but I think at the end of the day religion tomeis
being spiritual, a set of guiding principles and morals,
a way to develop as a person. Church for me is
something we do as a family, it’s a good time to spend
together and grow in our spirituality.
2020 Lee, Mazmanian, and Perlow 109
This type of disclosure about religious beliefs represented
a substantial deepening in the intimacy of
the personal sharing in the team. In experimenting
with how to foster deeper personal engagement by
changing the number of people on the calls and
expanding the time spent discussing personal and
professional goals, the team moved beyond being
passive recipients of the intervention to become designers
of their own change process.
Deeper sharing of work challenges. In the latter
weeks of the intervention, the team also adapted the
scripts in the pulse checks with the aim of inspiring
deeper sharing of work challenges. In pulse check
responses throughout the early and middle weeks of
the intervention, team members frequently brought
up the challenge of feeling crunched for time and
not having enough hours to complete work. In fact,
multiple team members raised challenges around
hours worked in their responses to the pulse check
questions in weeks 3 through 7. After hearing repeated
frustration about this issue of hours, the
facilitators tried to initiate a discussion about hours
in the week 5, 6, and 7 pulse checks. While these
attempts generated some discussion, the conversations
yielded little concrete resolution. The lack of
resolution was due, in part, to the limited time
remaining after the round-robin responses to the
pulse check questions. At the end of the week 7
pulse check, the team finally addressed the issue
head on. A U.S. manager expressed his desire for
the team to dive deeper into the underlying problem
with hours:
I just wanted to put out one point that everyone is
coming up with time, but it’s really great to know
that—why they are not getting time? Is it because of
too much work and they already committed, as per
estimation, but are not able to do it? There has to be a
reason, right? It’s not that 24 hours has squished to 20
hours, that’s not it. So, if everybody can come up with
how to talk about time and how it can be improved,
rather than just mentioning time.
Motivated by a desire to engage in a real discussion
about the hours challenge, the same U.S. manager
suggested an adaptation to the following week’s
pulse check script. Rather than have everyone answer
the four standard pulse check questions in
a round-robin style, he instead wanted to focus
the entire team’s time, energy, and attention on the
issue of hours for the week 8 pulse check. The
U.S. manager facilitated the meeting and opened by
encouraging people to express their concerns about
hours:
We want to be thinking about how we can work
smarter, not harder . . . If we are not finding time to do
what we need, we need to talk about it—not just the
reason but how we can make adjustments around the
issue of time.
After weeks of generalized discussion of the topic,
the substance and productivity of the week 8 pulse
check discussion was remarkable. Not only did the
India side raise concrete issues around hours, the
U.S. project lead and managers responded in a way
that indicated they acknowledged the challenges
the Indian members were facing and genuinely
wanted to resolve these issues. In response to the
U.S. manager’s encouragement, two Indian engineers
pointed out that the time required for testing
their code was not included in time estimates for
tasks in the team’s project planning.Thismeant that
they had to put in more hours than was expected to
finish their tasks. A third Indian engineer questioned
the usefulness of the “showcase” meetings
that the team held each week to show the client the
progress the team made. The entire team was expected
to be on these two-hour calls, evenmembers
who weren’t presenting their work. To accommodate
the client, the calls took place in the evenings,
India time. He said:
Basically, we have four to five showcase meetings . . .
they will help me understand the other activities in
the project, but, if I have a time crunch or my plate is
full, I wish I could drop off that call and focus on my
work.
He also questioned the value of meetings that had
been created among the Indian team members to
make sure that the Indian manager was aware of
issues before they were raised in the meeting with
the entire team. Finally, a fourth Indian engineer
raised an issue related to the challenges the team
had with conferencing software that often took a
long time to set up: “Sometimes, Citrix is really
slow; there are days when we have to restart our
laptop for Citrix and it takes 15 minutes or 20
minutes to start a call.”
Not only were Indian team members openly sharing
their issues and challenges, the U.S. managers
heard the challenges and developed concrete actions
to try to address these issues. For instance, the
U.S. project lead agreed that it may not make sense
for everyone to be on the call for the showcase
meetings and even admitted that it “was probably
poor planning on my part.” The team also discussed
the problem of not including estimates of time spent
110 Academy of Management Journal February
testing code in their sprint planning. Based on this
discussion, the team decided to allocate more time to
testing and to start documenting howmuch time was
being spent on testing so that they could develop
accurate estimates for future efforts. The U.S. project
lead said:
What we need to account for is testing support, that’s
the short-term solution we can account for [Indian
engineer’s] time or somebody’s time to support the
tester to navigate through the process, so we account
for that, that is one thing. Once we account for that,
other things will start to fall off their plate. You’re
right, we’re not accounting for that anywhere, but it is
an important task, yet we all hear from Scrum calls
that work is not getting done because of support for
testers.
Awareness of a positive change in relational dynamics
was shared throughout the team. As one
Indian engineer reflected after the pulse check,
“Everyone is talking—the team is becoming more
vocal and speaking up about their perspectives.”
Thus, in the latter weeks of the intervention, the
positive change in relational dynamics continued
to build through the deeper enactment and adaptation
of the intervention’s interaction scripts.
Team members, buoyed by the momentum from
the early weeks, worked the spaces of the intervention
and redesigned the interaction scripts
to further the change effort. In so doing, they connected
more deeply as people, not just as colleagues,
and tackled persistent work challenges.
As a result, new forms of interacting and relating,
which were counter-normative before the intervention,
increasingly became accepted and embraced
by the team.
Changes in Team Relational Dynamics
Throughout the course of the 10-week intervention,
we witnessed a shift in the level of respect,
openness, and connectedness on the team. The
emergence of positive relational dynamics on this
team was evidenced by both the actions and reflections
of those involved.
Respect. The increase in respect among team
members was evident in the degree to which team
members increasingly valued and supported one
another, especially across geographical and hierarchical
divides. An example of the increased respect
within the team occurred in week 9, when the U.S.
team lead suggested a change in the time of one of
the team’s calls. In an email to the team, he wrote:
Team—As you might have started to realize that we
are literally fighting for meeting time during the
day . . . I am thinking if it will be a good idea to move
the [meeting name] calls to the night (India morning)
so that we can free up the prime-time meeting slots.
I have given a few options, letmeknowthe consensus.
Also, “No” is also an acceptable answer!!
Several Indian team members pointed to this incident
as indicative of a broader shift in the way the
U.S. team members were engaging with the Indian
team members. In week 10, an Indian engineer reflected,
“Usually the U.S. [team] just tells us when it
is. There’s more two-way dialogue now.” Another
Indian engineer also noted a change in the way the
U.S. project lead engaged with the Indian team
members: “[The U.S. project lead] is now more open
to the team’s perspectives and ideas.” A third Indian
engineer compared this team to those he had been on
in the past: “In this project, the first thing is [the Indian
team members] have been heard—the feeling
that your voice is heard by the leadership.”
Openness. The development of openness within
the team was evident in the increased sharing about
personal matters and work issues. In the meetings
leading up to the intervention’s start and during the
first few weeks of the intervention, everyone on the
team was encouraged to speak up to provide input on
the goals of the intervention. However, the senior
U.S. members of the team dominated the conversations
during this early period and Indian team
members would often remain silent.
By the end of the intervention, the team was
sharing more openly, surfacing and addressing issues
across the group. The Indian team members
became more willing to speak directly and reflected
that they had become “more equal in terms of communication
and the kind of work that is happening.”
As one Indian engineer said in her week 10 reflection,
“It’s definitely increased our communication
together.” Another Indian engineer reflected:
[The intervention] has definitely helped the team come
together. People are more expressive, there is a platform
to express opinions and problems. Any one of us
would not have taken the step to do that on our own.
This reflection points to the role of spaces in bringing
the entire team together outside of everyday interactions
and the critical component of interaction
scripts that enabled the expression of “opinions and
problems” in a productive manner. The “platform”
referred to is, in essence, the spaces and interaction
scripts of the intervention.
2020 Lee, Mazmanian, and Perlow 111
Connectedness. The increased connectedness
within the team was reflected in the way the team
members began to connect with one another as
people, not just as coworkers. In the CWT calls and
the adapted pulse check meetings, team members
learned about each other’s personal lives and engaged
with each other in both light-hearted (telling
funny stories) and earnest (talking about one’s religious
views) ways. After the intervention ended, an
Indian engineer reflected:
The team became more close-knit after all those
weeks … All those icebreakers on the pulse check,
they may seem like stupid questions, but at the end of
the day they really help—everyone was laughing, and
that really helps.
At one point in week 9, the workload on the team
became intense. It was at this moment, when an
earlier incarnation of this team may have developed
frustrations with their global counterparts, that the
team’s greater connectedness became evident. Instead
of each member of the team retreating into his
or her own workload, an Indian engineer stepped
forward and volunteered to help the U.S. manager
with a complex task that the manager did not have
the time to work on. During this period, an Indian
engineer shared, “The team is working really hard to
help each other out to make sure everything is finished
for the deadline. We’re working together to
see that we didn’t miss anything.” Also in week 9,
the U.S. manager emphasized the change in the way
the team was collaborating during this time of high
workload and increased stress:
If we hadn’t had [the intervention], we would still be
in small, small teams and passing specs across the
wall from one side to the other and not this level of
[team-wide] collaboration . . . like the email from [the
Indian engineer this week] saying “I am available.”
Those kinds of things are happening because of [the
intervention]–better understanding, knowing people
better on personal level, understanding career aspirations,
and feeling much more comfortable reaching
out asking for help or asking for work.
To be sure, team relational dynamics are the
product of numerous small exchanges, and not every
single interaction observed was uniformly positive.
In fact, how team members dealt with exchanges that
they experienced as less than positive was also an
important part of the change process. For example, in
the week 8 pulse check, the most junior Indian engineer
raised that one of her challenges was feeling
like part of her day was less productive because she
was waiting for direction. After the call, her manager
admonished her for sharing this on the call and asked
her to run things by him before she brought them to
the entire team. The junior engineer turned to one of
her Indian colleagues to make sense of this exchange.
As she relayed to a facilitator, the colleague told her
that “she had every right to say what she did and that
she should always have that right going forward.”
This response reassured her, and, a week later, she
was adamant in her weekly check-in with the facilitator
that the team had become more open and
connected. She emphasized that she hoped that the
intervention would continue, indicating her continued
faith in her colleagues and the intervention. This
incident, while rare in our data, reveals the difficulty,
if not impossibility, of fostering relational dynamics
that are uniformly positive. It also suggests that, once
a team has made a shift to more respectful, open, and
connected forms of interaction, these positive relational
dynamics can help contextualize and mitigate
difficulties when they arise.
Across the board, teammates were struck by the
dramatic growth of respect, openness, and connectedness
in the team during the intervention. Table 1
provides additional data documenting the emergence
of positive relational dynamics that occurred
over the course of the 10 weeks.
FOSTERING POSITIVE
RELATIONAL DYNAMICS
Based on our findings, we induced a theoretical
model that distills how the intervention—a set of
spaces and interaction scripts—facilitated the emergence
of positive relational dynamics on a global team.
Below, we present each component of our model.
Spaces and Interaction Scripts Facilitated the
Enactment of Sharing Interactions
During the course of the intervention, we observed
the team interacting in new ways. Specifically, we
saw evidence of team members sharing aspects of
their personal lives and work challenges across
geographic, cultural, linguistic, and hierarchal divisions.
We also witnessed team members responding
to each other’s sharing in positive ways. We label
the combination of a sharing behavior and a positive
response as a “sharing interaction.” Understanding
how the intervention facilitated the enactment of
sharing interactions requires insight into how spaces
and interaction scripts worked together to transform
the team’s taken-for-granted ways of interacting.
112 Academy of Management Journal February
TABLE 1
Changes in Positive Relational Dynamics over the Course of the Intervention
Early weeks: Low levels Middle weeks: Growing levels Latter weeks: High levels
Respect:
Affirming,
Supporting,
Listening,
Helping
“I’m more excited for the [Indian team
members] . . . Before, it was like
a parent–child, funneling
information.” (Indian senior
manager)
“[A U.S. manager] struggles. He is a
pretty involved person who gets
upset when things are not done on
time . . . He is accepting these
realities . . . I have seen a softening of
his tone. I amencouraging him to ask
what we can do to help when things
are not done in time.” (U.S. senior
manager)
“It was the first time I ever heard [U.S.
managers] ask us to think about what
would work best on our end as a
team.” (Indian engineer)
“I echo [the U.S. senior manager, who
gave a low score on being heard]—I
want to challenge the team that,
when issues are raised, people
respond to them.” (U.S. manager) “It’s important to have time to
appreciate the work of the team. It’s
good to have more interactive time
together and have other activities
with the team.” (Indian engineer)
“So many things are going on—just
working, working, working. The
team is reviewing each other’s work,
stepping up, and generally
functioning at high throttle together.
We’re working together to see that
we didn’t miss anything.” (Indian
engineer)
“Right now, [the U.S. manager] is
passing the commands to the Indian
team members and you can’t just
pass the commands . . . He is in order
giver mode. He needs to become
more engaged and to help offshore.”
(U.S. senior manager)
“[A U.S. manager] ‘blocked time to talk
informally’ on Friday (his initiative).
She feels he is the mentor she’s been
looking for.” (Facilitator notes,
conversation with Indian engineer)
“The team is functioning more like
a team now and helping each other
if a task is not completed.” (U.S.
manager)
“I had a great conversation with [the
U.S. senior manager] last week and
will be helping out on one of his firm
initiatives this week. He gave a lot of
advice about a lot of things. He asked
me all the questions I have for him
and took time to answer each one.”
(Indian engineer)
Openness:
Sharing,
Disclosing,
Speaking up
“I don’t discuss things with others. I
prefer to think about a problem on
myown and solve it myself.” (Indian
engineer)
“The CWT calls are one of the best
ways to get people talking and
breaking the ice.” (Indian engineer)
“I like the CWT and meetings with the
whole team. It’s definitely increased
our communication together.”
(Indian engineer)
“She agrees that the Indian team is
quiet, that it takes ‘a lot of
encouragement from [a U.S.
manager]’ to get them to talk. She
also thinks part of the issue is that
they are more junior; ‘I, too, didn’t
have the confidence to speak up
when I was more junior.’”
(Facilitator notes, conversation with
Indian senior manager)
“The team is doing better than
last week . . . the dynamics are
improving . . . The team was very
vocal on the call [with an Indian
senior manager] and were talking
to each other.” (Indian engineer)
“We’re doing good, all been speaking
up, give inputs that we hadn’t really
been doing. Most people have
issues—the same ones that you
would find at other companies or
with other people—but we talk
about them.” (Indian engineer)
“When offshore has spoken up in the
past, the pushback from [the U.S.
side] has been very hard. I think this
makes people more hesitant to speak
up and to feel as if their voice is not
being heard.” (Indian engineer)
“Offshore doesn’t generally speak
much on the calls but the team is
starting to speak up on calls the last
two weeks.” (Indian engineer)
“‘Last three to four weeks, everyone
talking’—they said they’ve seen a
change in the last three to four weeks
as far as everyone on the team really
willing to be open and share their
concerns.” (Facilitator notes,
conversation with two Indian
engineers)
“From the business perspective,
people were talking about
themselves, trying to be more
honest. Important things came
out like technical skill sets,
development needs, and soft skills
that people want to develop. It will
help us get good at managing those
types of things. We also gave
suggestions—‘if you are doing this,
you may want to try this instead.’We
gave suggestions for next steps.”
(U.S. manager)
“Since we’ve [been on the project]
here, the point was more of one-way
communication. Now, everyone is
speaking and highlights [their]
concerns openly.” (Indian engineer)
2020 Lee, Mazmanian, and Perlow 113
The intervention’s spaces—the temporally and
symbolically bounded CWT calls and pulse check
meetings—created the opening for new forms of interaction
to emerge. The separation of the spaces
from the everyday interactions of the team rendered
existing norms and patterns of interaction less salient.
As a result, team members came into the CWT
calls and pulse checks knowing that the typical
norms might not apply. The symbolic and temporal
separation of the spaces also meant that team members
could not rely on habitual patterns of interacting
as a guide for how to behave in the spaces. Instead,
spaces encouraged individuals to set aside their
preconceived notions and bring an open mind to the
situation at hand. Furthermore, the intervention’s
spaces were referred to explicitly as a place for the
team to interact with each other in new and different
ways, supporting experimentation and risk taking.
However, spaces on their own do not provide direction
for how a team might interact within the
space or what type of experimentation could occur.
Taking a theater metaphor, spaces can be conceived
as a new stage, a place where people don new
clothing, set aside taken-for-granted assumptions,
and experiment with a new style of interaction.
However, an empty stage on its own leaves open
infinite possibilities—thus, the value of having
a script. Whereas spaces created the opening for
change, interaction scripts provided direction and
shape to the change process. These data suggest that
interaction scripts shaped interactions within spaces
through three mechanisms.
Interaction scripts guided experimentation into
specified forms. Within the boundaries of spaces,
scripts provided specific guidance on what to discuss
(content parameters) and how to discuss it
TABLE 1
(Continued)
Early weeks: Low levels Middle weeks: Growing levels Latter weeks: High levels
“[Indian engineer] still feels that it is
inappropriate to voice his issues on
the team calls since it is not an issue
that impacts everyone . . . He said he
would be more comfortable bringing
up issues in a small group setting
because he doesn’t want to ‘bug’ the
team by ‘generalizing’ things that
don’t apply to everyone.” (Facilitator
notes, conversation with Indian
engineer)
Connectedness “There should be some kind of
motivation for the team, a team
outing, a team lunch. It would
definitely help us. Something from
the on-site team motivating us . . .We
are only professionally connected.”
(Indian engineer)
“I’mseeing a change that is a very good
sign . . . [The intervention] is really
helping to bring the Indian team
members together and getting
them to increase the level of
communication.” (U.S. senior
manager)
“[The U.S. senior manager] thought
there were clear benefits in
improving one-on-one rapport,
collaborating across the United
States and India and understanding
what work is on each other’s plates.”
(Facilitator notes, conversation with
“Note that [Indian engineer] was U.S. senior manager)
expecting his first child and yet very
few people knew about it on the
team.” (Facilitator notes)
“It was such a delight talking to [the
U.S. senior manager]. I now fear him
a little less than I did before the call.”
(Indian engineer)
“I feel good—I know [U.S. manager] as
a person, I now know certain things
about [the U.S. senior manager] that I
didn’t know before. We are more
bonded within the team.” (Indian
engineer)
“[Indian engineer] says that there are
many team members that she does
not know, and thus would like to
focus on getting to know more of
them during the CWT calls.”
(Facilitator notes, conversation with
Indian engineer)
“[Speaking about the small-group
CWT calls:] Since I am on the BO
team, I don’t interact as much with
ETL—we got a chance to know
each other. Plus, there was a light
moment, to help to chill out with
each other . . . Overall, we got to
know each person—not just
professionally, but personally.”
(U.S. manager)
“It is difficult to differentiate [the
intervention] from the way that the
team now operates. There has been a
positive impact on collaboration
since the experiment began.” (U.S.
manager)
“I feel really disconnected from the
team.” (Indian engineer)
114 Academy of Management Journal February
(participation rules). The specificity and explicitness
of the interaction scripts were key to reducing
uncertainty during the change process and structuring
what type of experimentation and risk taking
was to happen in the spaces of the intervention. The
value of the specificity of the scripts was evident
when team members continued to ask for concrete
conversation prompts to guide their personal sharing
in the early weeks of the intervention.
The scripts’ concrete guidelines also made team
members aware of potential disconnects between
ideal sharing behaviors as specified by the scripts
and how the team was actually interacting. For example,
in week 3 of the intervention, the Indian
senior manager expressed to the U.S. project lead
that the Indian team members were still not speaking
up in the pulse checks as much as desired and that
senior members of the team needed to continue
to encourage them to express their feelings. Such
feedback points to how the explicit guidelines of the
scripts provided a benchmark by which team members
could assess their interactions and encourage
each other to more fully engage in sharing behaviors.
Interaction scripts legitimated counter-normative
forms of interaction. Whereas the spaces of the intervention
reduced the salience of existing interaction
norms, scripts seeded the development
of new norms. In this study, scripts expanded the
scope of what was legitimate sharing and, in doing
so, challenged expectations of what was appropriate
behavior. Prior to the intervention, sharing
personal information and work challenges violated
implicit team norms and entailed social
risk. The scripts, however, explicitly asked individuals
to engage in such sharing. In doing so,
the scripts reduced the social risk of broaching
such topics, making them legitimate topics of
conversation.
Interaction scripts established an expectation of
shared risk taking. Interaction scripts also fostered
risk taking by establishing an expectation that everyone
would engage in the same risky behaviors.
Classic research in game theory points to the difficulty
for a single person to take a risk that only yields
benefit if others also take the same risk. In strong
organizational cultures, an individual challenging
normative assumptions about what, when, and how
to speak is likely to be censured, mocked, or completely
ignored. Interaction scripts, however, mitigated
that risk through the promise of collective
engagement. It was established up front that every
single member of the team was expected to take the
risk (i.e., participate in the intervention activities).
As a result, the collective action problem was alleviated
and benefits accrued for each group member.
Notably, the same mechanisms through which
spaces and scripts facilitated sharing of personal and
work-related matters also facilitated consistently
positive responses to disclosures. Positive responses
manifested as empathetic listening, reciprocal sharing,
and active responding to the expressed challenges.
First, the fact that these interactions were
occurring in a space separate from the everyday
created latitude for the team to take the time to hear
each other. Second, the participation rules of the
scripts (e.g., round-robin turn taking in which people
were given a turn to talk and clear expectations of
when to listen) set the stage for collective sharing.
This guided reciprocity and structured listening
limited immediate and unreflective reactions to
disclosures. Third, because scripts legitimated the
new sharing behaviors, the chances that any single
disclosures would be viewed as socially deviant
were reduced. Instead, individuals were primed to
give each other leeway and respond positively to
sharing.
Figure 2 depicts how spaces and interaction
scripts facilitated the enactment of sharing interactions.
The intervention spaces created the opening
for change by reducing the salience of existing
norms, fostering greater mindfulness of one’s interactions,
and facilitating experimentation. Scripts
provided shape and direction to the change process
by specifying both the what and the how of the interactions,
establishing the legitimacy of such behaviors,
and creating an expectation that all would
engage in these new risky behaviors. Notably, the
mechanisms by which spaces and interaction scripts
facilitated sharing interactions were mutually reinforcing.
The symbolic and temporal separation of the
spaces from everyday patterns of interaction facilitated
the enactment of the new interactions specified
by the scripts. Conversely, the existence of scripts
that explicitly specified forms of experimentation
and counter-normative interaction further enhanced
the symbolic separation of the spaces.
Positive Cycle of Emerging Positive Relational
Dynamics and Deeper Sharing
As the intervention progressed, the sharing interactions
sparked the emergence of a virtuous
feedback cycle of relational dynamics built on
respect, openness, and connectedness. Building
from Figure 2, which details the link between
spaces, interaction scripts, and sharing interactions,
2020 Lee, Mazmanian, and Perlow 115
Figure 3 outlines our full model of how positive relational
dynamics emerged. This model illustrates
how sharing interactions led to a feedback cycle
of increasing respect, openness, and connectedness.
Below, we unpack the elements and links in
this cycle.
First, the sharing interactions directly established
respect, openness, and connectedness. The sharing
interactions fostered respect by humanizing individuals
across hierarchical, cultural, linguistic,
and geographic distances. These interactions helped
team members see one another as people, with lives,
interests, and passions outside of work. The sharing
interactions established openness by virtue of the
wider range of topics—both personal and work
related—that the team shared in pulse checks and
CWT calls. And the sharing interactions fostered
connectedness through the positive responses to
personal and work-related disclosures. Taking the
form of reciprocal sharing, empathetic listening,
and active responding, positive responses fostered a
sense that the group cared about each other’s work
experiences and who they were as people. In these
ways, the sharing interactions established the beginnings
of positive relational dynamics on the team.
Second, the respect, openness, and connectedness
that emerged from early sharing interactions generated
energy and engagement within the team, fueling
a cycle of growing positive relational dynamics. The
energy was evident in the excitement and satisfaction
that individuals displayed after their CWT
calls and during the fun, personal pulse check
conversations in weeks 4 and 6. The increased engagement
by the team manifested in the team’s
adaption of the initial scripts, starting in week 4. Such
adaptation of the interactions scripts indicated that
team members were progressively investing more
of their time, energy, and belief in the possibility of
change.
The increased energy and engagement of the team,
in turn, inspired individuals to fully embrace the
scripts in latter weeks. Thus, the enactment of sharing
interactions, facilitated by the intervention’s
spaces and interaction scripts, initiated a cycle of
growing respect, openness, and connectedness on
the team—a cycle fueled by increased energy and
engagement in the change process.
One important feature of our theoretical model is
that not every type of interaction script will necessarily
lead to the emergence of positive relational
FIGURE 2
How Spaces and Interaction Scripts Facilitated Enactment of Sharing Interactions
Spaces: Separation from everyday norms and patterns of interaction
Sharing personally
and about work
challenges
Interaction scripts
that specify collective
interpersonal sharing
(content parameters and
participation rules)
Positive response
(empathetic listening,
reciprocal sharing,
active responding)
Sharing Interactions
+
Space creates opening for change
by
1. Reducing salience of norms
and patterns
2. Increasing attention to
immediate situation
3. Fostering experimentation
and risk taking
Scripts support change by
1. Guiding experimentation
into specified forms
2. Legitimating counter-normative
forms of interacting
3. Creating expectation of
shared risk taking
Mutually
reinforcing
116 Academy of Management Journal February
dynamics. In this particular intervention, the content
parameters—interpersonal sharing about personal
or work challenges—and participation rules—
collective engagement across cultural and hierarchical
team divides—of the interaction scripts were
important to the observed outcomes of the intervention.
The collective interpersonal sharing
specified by the interaction scripts in our study were
key to humanizing individuals, fostering positive
responses to disclosures, and generating positive
relational dynamics.
In addition, not every single instance of interaction
scripts fostered sharing interactions that were
equally successful at facilitating respect, openness,
and connectedness—notably, the week 5 pulse
check when team members were asked with whom
they would like to trade places. One feature of the
adaptation of the week 5 pulse check script was that
it prompted social comparisons within the team. The
content parameters in other weeks did not prompt
social comparisons, but instead asked people to
share a personal story or to discuss a work challenge.
This variation suggests that interaction scripts that
elicit social comparisons are less likely to humanize
team members and, as a result, are less likely to foster
respect, openness, and connectedness. This variation
also highlights how spaces on their own would
likely not have led to a transformation in the team’s
relational dynamics.
The Role of Script Adaptation to the Change
Process
The organic adaptations of the scripts, prompted
by suggestions from the team members to the facilitators,
played a critical role in the change process.
Such adaptations indicate that, while the scripts
were initially developed via a top-down process,
their evolution occurred through a bottom-up process.
The adaptations of scripts led to some of the
most consequential moments in the team’s evolving
relational dynamics, indicating that, when there is
an opportunity for bottom-up adaptation of scripts,
valuable experimentation may occur.
FIGURE 3
How Spaces and Interaction Scripts Fostered Positive Relational Dynamics
Sharing personally
and about work
challenges
Interaction scripts that
specify collective
interpersonal sharing
Positive response
to sharing
Cycle of positive relational
dynamics
Spillover of positive
relational dynamics in
everyday interactions
Adapting scripts
Spaces: Separation from everyday norms and patterns of interaction
Enabling: Leadership Support and External Facilitators
Respect,
openness,
connectedness
Energy and
engagement
Sharing Interactions
Figure 2
Figure 2
Everyday
2020 Lee, Mazmanian, and Perlow 117
Spillover of Positive Relational Dynamics Outside
of the Intervention Spaces and Scripts
The emergence of positive relational dynamics
within the spaces of the intervention led to improved
dynamics outside of these spaces. Examples of
spillover included the team members moving to sit
together in week 4. Further, their sitting together fed
back into how the team engaged within the spaces of
the intervention. Another example of spillover occurred
in week 8, when the U.S. project lead asked
the Indian team members for their input in changing
one of their meeting times, and in week 9, when a
team member became aware of how busy certain
colleagues were and offered to pitch in and help
those with more on their plate. These examples
highlight the fluidity of the boundaries between the
intervention spaces and the everyday interactions of
the team as well as the positive feedback in relational
dynamics across these boundaries.
Enabling Conditions: Leadership Support and
External Facilitators
These data suggest that, in order to realize the
generative potential of spaces and interaction scripts
for fostering positive relational dynamics, two enabling
conditions should be present: leadership
support and external facilitation.
Leadership support. As noted above, one of the
key functions of spaces and interaction scripts is
that they mitigate the social risk of enacting new
behaviors. However, the legitimating power of
spaces and interaction scripts do not solely reside in
the spaces and scripts themselves. The authority of
the team and organizational leaders who sanctioned
and supported the intervention imbued the spaces
and scripts with external legitimacy. Such legitimacy
was particularly important in the initial stages
of the intervention, when team members were deciding
how to enact the scripts and whether or not to
engage in what they perceived to be socially risky
behaviors. The engagement and support of the
U.S. senior manager and partner were critical for signaling
to the other teammembers that itwas legitimate
and safe to interact with one another in new ways.
Team and organizational leaders supported the
intervention in multiple ways. They sanctioned the
intervention as an official firm priority by ensuring
that time spent on the intervention could be logged as
part of each team member’s target hours toward firm
initiatives. Just as importantly, the project lead and
senior partner participated in the intervention itself.
Their participation and engagement in the sharing
interactions reinforced the sense that everyone on
the team would be taking the same risks, thus reducing
the risk for any one person.
External facilitation. Facilitators also supported
the change process in multiple ways. First, the presence
of the external facilitators in the intervention
enhanced the symbolic separation of the
intervention’s spaces and the everyday work of the
team. Their presence signaled to team members that
team norms would not necessarily apply in these
spaces. Second, the neutral status of external facilitators
mitigated the impression that participating
in the intervention was a top-down imperative of
management. The presence of facilitators, as outsiders
to the team’s formal power structure, bolstered
the sense that everyone on the team was in
the change effort together. Third, facilitators provided
in situ feedback and encouragement to support
script enactment. For example, following a
team member’s response to the pulse check questions
in week 5, the facilitator said, “Thank you so
much. That was terrific. So we are hoping that all of
your openness that existed in the first half of the call
will carry over in the second half.”
DISCUSSION AND IMPLICATIONS
This research documents one way that organizations
can foster positive relational dynamics on
teams.We found that the combination of spaces and
interaction scripts can enable the emergence of respect,
openness, and connectedness in teams. While
spaces create the opening for new relational dynamics
to emerge, interaction scripts provide content
parameters and interaction rules that can guide
team members to interact in ways that generate respect,
openness, and connectedness.
These findings contribute to the literature on
positive relationships at work. Existing research in
this area highlights the importance of relationships
characterized by respect, openness, and connectedness
to team and organizational functioning (Carmeli
et al., 2015; Dutton & Heaphy, 2003; Edmondson,
1999). While scholars have suggested different potential
enablers of positive relational dynamics, such
as inclusive leadership (Fletcher & Ragins, 2007;
Nembhard & Edmondson, 2006) and relational skills
and practices (Baker&Dutton, 2007; Quinn, 2007), few
studies have empirically investigated the processes
and mechanisms by which relational dynamics can
be transformed. This present study deepens our understanding
of how positive relational dynamics can
118 Academy of Management Journal February
be fostered. We illustrate how, by carving out spaces
and providing interaction scripts that encouraged
collective interpersonal sharing, a distributed global
team developed a virtuous cycle of respect, openness,
and connectedness. This study also provides empirical
support for the positive feedback loops that
scholars of positive relationships at work have theorized
are inherent to the evolution of positive relationships
(Kahn, 2007; Quinn, 2007), and the role of
energy as a catalyst for these feedback loops (Dutton &
Heaphy, 2003; Quinn, 2007).
This study also deepens our understanding of
when interpersonal sharing can lead to positive relationships
at work. Existing research in psychology
highlights the positive relationship between personal
sharing and relational strength in general social
contexts (e.g., Collins & Miller, 1994). However,
studies of sharing in work contexts have noted the
negative impact that personal disclosure can have in
these settings (Detert & Edmondson, 2011; Morrison,
2011). The present study supports the importance of
normative context in influencing how interpersonal
sharing is received, but also suggests that the preexisting
normative context is not fully determinant.
The preexisting team norms in this study were
hardly favorable to the sharing of personal matters or
work challenges. Instead, the temporary opening in
the normative context of the team, facilitated by the
interaction scripts and spaces of the intervention,
was sufficient for the interpersonal sharing to be received
positively.
In addition to contributing to the research on
positive relationships at work, the concept of interaction
scripts contributes to research on change in
organizations. Prior research has highlighted the
importance of spaces, or bounded social settings, for
enabling change in institutional practices, routines,
group dynamics, and organizational culture (Bucher
& Langley, 2016; Furnari, 2014; Howard-Grenville
et al., 2011; Isaacs, 1999; Kellogg, 2009; Zietsma &
Lawrence, 2010). However, the notion of spaces,
which are, by definition, neutral as to what happens
within them, fails to provide insight into the dynamics
and mechanisms of change (Furnari, 2014;
Polletta, 1999).
This study introduces the notion of interaction
scripts to deepen our understanding of how and why
certain changes in interaction patterns can emerge
within a space. Furthermore, this study identifies the
mechanisms by which interaction scripts complement
spaces in affecting relational change: namely,
they help a team overcome the uncertainty, social
risk, and collective action problem associated with
deviating from team norms and engaging in risky
interpersonal sharing behaviors.
Notably, when we revisited the empirical research
that establishes spaces as a vehicle for change in
organizations, we found evidence of additional
structures resembling our concept of interaction
scripts that contributed to the change processes.
For instance, in Howard-Grenville and colleagues’
(2011) study of cultural change, they note the use of
“liminal spaces” that appear to incorporate interaction
scripts. One of the key liminal spaces in their
study is a strategy retreat that introduced celebration
and games (content parameters) and invited participation
from all hierarchical levels (participation
rules).AsHoward-Grenville et al. (2011: 530) explains:
Actors gave careful thought to how they would facilitate
and invite interaction, intentionally bringing to
the fore moments, events, or occasions in which
people could interact differently, suspend the usual
organizational social dynamics, and allow role-based
interactions to recede in importance.
In other words, these actors were consciously incorporating
scripts to structure interactions within
the space to bring about new ways of relating.
Similarly, Bucher and Langley’s (2016) study of
routine change describes how both reflective and
experimental spaces facilitated successful changes
in medical routines. Upon examination, both kinds
of spaces in their study contained structured guidelines
for interaction. For example, in reflective
spaces, individuals were explicitly directed to
review the old routines while envisioning new
routines (i.e., content parameters), and, in one of
their cases, an external facilitator was even brought
in to guide the reflection. In experimental spaces,
the new envisioned medical routines provided the
structured guidelines governing how the teams
would interact within the space. In Kellogg’s (2009)
study of institutional change at two hospitals, in
addition to relational spaces where reformers could
meet to interact and mobilize, additional guidelines
shaped reformer’s interactions in ways that supported
the change process. For example, directors of
surgery gave directives to doctors to discuss how to
change practices in order to comply with the new
hours policy (content parameters) and, in afternoon
rounds, juniors and seniors were both expected to
speak and raise issues (content parameters and participation
rules).
That guidelines for interaction with content parameters
and participation rules are present in several
existing studies of organizational change highlights
2020 Lee, Mazmanian, and Perlow 119
that spaces are often accompanied by additional
structures that support the change process. However,
this hidden role of interaction scripts as a unique
contribution to fostering change has been undertheorized.
For example,while Furnari (2014) recently
suggested that, in addition to spaces, successful interaction
rituals—or interactions characterized by
mutual attention andemotional energy—are needed to
support the emergence of new practices, he did not
address how successful interaction rituals come to be.
Our notion of interaction scripts serves as one possible
means for facilitating successful interaction rituals
within spaces.
This work also contributes to the literature on
distributed and global teams. Research on global
teams has highlighted the challenges of overcoming
cultural, hierarchical, linguistic, and geographic
distance to achieve team effectiveness (Cramton,
2001; Espinosa, Slaughter, Kraut, & Herbsleb, 2007;
Herbsleb & Mockus, 2003; Hinds & Bailey, 2003;
Hinds & Mortensen, 2005; Metiu, 2006). Scholars
note the importance of fostering closeness within
global teams to manage coordination challenges.
Current work suggests that in-person site visits are
critical for fostering this familiarity (Hinds &
Cramton, 2014). According to this research, site
visits allow individuals to work together in vivo and
provide informal time for getting to know one another
outside of work (Hinds & Cramton, 2014). Our
study suggests an alternative mechanism for fostering
closeness in global teams that may not require
extended time with each other on-site. In these data,
spaces and interaction scripts fostered positive relational
dynamics even while the team remained
distributed. The team described in this research
never met face to face as an entire group.
This research has implications for practitioners
and leaders seeking to foster positive relational dynamics
in teams. Practitioners should think about
how to structure personal sharing and open discussion
of work issues in service of fostering respect,
openness, and connectedness and should consider
using spaces and interaction scripts as a resource.
The recent research on choice architecture in behavioral
economics provides a useful analog for
understanding the potential implications of interaction
scripts for practitioners. Choice architecture
is the study of how designing the decision
environment can influence or “nudge” individuals
to make decisions that have better outcomes for the
individual and/or society, such as making 401(k)
contributions a default option in order to increase
retirement savings (Benartzi & Thaler, 2007; Thaler
&Sunstein, 2012). Interaction scripts can be viewed a
form of “interaction architecture”; just as choice architecture
can be used to nudge individuals to make
more optimal decisions, interaction scripts may
serve as a way to nudge group members to interact
with each other in more productive and beneficial
ways.
This study was limited in temporal scope. While
the relatively short duration of the study (two-week
assessment plus a 10-week intervention) speaks to
the power that spaces and interaction scripts can
have in fostering new dynamics, it would be worthwhile
to examine how team relational dynamics
evolve over a longer timeframe. To borrow Lewin’s
classic framework on the process of group change
(Lewin, 1947), we observed the “unfreezing” and
“changing” stages of change but likely did not observe
the full process of the team “re-freezing”
around new norms or patterns of interaction. Do
teams continue to utilize spaces and interaction
scripts even after the new patterns of interaction are
firmly established, or do they no longer need them?
Future research could explore the temporal dynamics
of relational change over time and the evolving
role that spaces and scripts play throughout a longer
process of relational change.
Future research is needed to further understand
the conditions under which spaces and interaction
scripts are likely to foster positive relational dynamics.
Specifically, future research could explore
whether there are certain content parameters and
participation rules of interaction scripts that make
them more or less effective at fostering positive relational
dynamics. For example, what level of personal
sharing is needed to foster connectedness?
What might be the downside of too much personal
sharing (e.g., Martin et al., 1998)? Also, while the
intervention we studied comprised spaces and
scripts, future research could explore under what
conditions bounded spaces might be sufficient to
foster positive change without the addition of interaction
scripts. Alternatively, can scripts effectively
facilitate change outside of spaces? Finally, the
question of whether external facilitation is needed in
order for spaces and scripts to enable change is an
open question. Notably, the interaction scripts in
Howard-Grenville and colleagues’ (2011) study were
developed and utilized by insiders, suggesting that
external facilitators may not always be necessary for
scripts to support change.
Lastly, future research is needed to explore the
relevance of interaction scripts for fostering other
types of change. As noted, interaction scripts can be
120 Academy of Management Journal February
found in many other case studies of change in and
across organizations but have yet to be theorized or
identified as an important enabler of change. How
might the relevance and functioning of interaction
scripts be different when the change is in organizational
routines, practices, or cultural repertoires
rather than in relational dynamics? What are the
boundary conditions for when interaction scripts are
useful for facilitating such change?
It is notoriously difficult for teams to change negative
patterns of interacting and relating. Our research
shines light on one resource for facilitating the
emergence of positive relational dynamics in teams.
We found that the combination of spaces and interaction
scripts can guide groups to collectively
manage social risk and experiment with new forms
of interpersonal sharing. In doing so, they plant the
seeds of respect, openness, and connectedness that
can grow and ultimately transform a team’s relational
dynamics.
REFERENCES
Abelson, R. 1981. Psychological status of the script concept.
American Psychologist, 36: 715–729.
Andersson, L. M., & Pearson, C. M. 1999. Tit for tat? The
spiraling effect of incivility in the workplace. Academy
of Management Review, 24: 452–471.
Ashcraft, K. L. 2000. Empowering “professional” relationships:
Organizational communication meets feminist
practice. Management Communication Quarterly, 13:
347–392.
Ashforth, B., & Fried, Y. 1988. The mindlessness of organizational
behaviors. Human Relations, 41: 305–329.
Baker, W., & Dutton, J. E. 2007. Enabling positive social
capital in organizations. In J. E. Dutton & B. R. Ragins
(Eds.), Exploring positive relationships at work:
Building a theoretical and research foundation:
325–346. Mahwah, NJ: Lawrence Erlbaum Associates.
Barley, S. R., & Tolbert, P. S. 1997. Institutionalization and
structuration. Organization Studies, 18: 93–117.
Benartzi, S., & Thaler, R. H. 2007. Heuristics and biases in
retirement savings behavior. Journal of Economic
Perspectives, 21: 81–104.
Blatt, R., & Camden, C. T. 2007. Positive relationships
and cultivating community. In J. E. Dutton & B. R.
Ragins (Eds.), Exploring positive relationships at
work: Building a theoretical and research foundation:
243–264. Mahwah, NJ: Lawrence Erlbaum
Associates.
Bohm, D. 1990. On dialogue. New York, NY: Routledge.
Bowen, F., & Blackmon, K. 2003. Spirals of silence: The
dynamic effects of diversity on organizational voice.
Journal of Management Studies, 40: 1393–1417.
Bucher, S., & Langley, A. 2016. The interplay of reflective
and experimental spaces in interrupting and reorienting
routine dynamics. Organization Science, 27:
594–613.
Burt, R. S. 2000. The network structure of social capital.
Research in Organizational Behavior, 22: 345–423.
Carmeli, A., Dutton, J. E., & Hardin, A. E. 2015. Respect as
an engine for new ideas: Linking respectful engagement,
relational information processing and creativity
among employees and teams. Human Relations, 68:
1021–1047.
Carmeli, A., & Gittell, J. 2009. High-quality relationships,
psychological safety, and learning from failures in
work organizations. Journal of Organizational Behavior,
30: 709–729.
Chaikin, A. L., & Derlega, V. J. 1974. Variables affecting the
appropriateness of self-disclosure. Journal of Consulting
and Clinical Psychology, 42: 588–593.
Charmaz, K. 2006. Constructing grounded theory: A
practical guide through qualitative research. London,
U.K.: SAGE.
Collins, N. L., & Miller, L. C. 1994. Self-disclosure and
liking: A meta-analytic review. Psychological Bulletin,
116: 457–475.
Cortina, L. M. 2008. Unseen injustice: Incivility as modern
discrimination in organizations. Academy of Management
Review, 33: 55–75.
Cramton, C. D. 2001. The mutual knowledge problem and
its consequences for dispersed collaboration. Organization
Science, 12: 346–371.
Creed, W. E. D. 2003. Voice lessons: Tempered radicalism
and the use of voice and silence. Journal of Management
Studies, 40: 1503–1536.
Detert, J. R., & Burris, E. R. 2007. Leadership behavior and
employee voice: Is the door really open? Academy of
Management Journal, 50: 869–884.
Detert, J. R., & Edmondson, A. C. 2011. Implicit voice
theories: Taken-for-granted rules of self-censorship
at work. Academy of Management Journal, 54:
461–488.
Dutton, J. E. 2003. Energize your workplace: How to
create and sustain high-quality connections at work.
Hoboken, NJ: John Wiley & Sons.
Dutton, J. E., Ashford, S. J.,O’Neill, R. M.,&Lawrence, K. A.
2001. Moves that matter: Issue selling and organizational
change. Academy of Management Journal, 44:
716–736.
Dutton, J. E., & Heaphy, E. D. 2003. The power of highquality
connections. In K. S. Cameron, J. E. Dutton,
2020 Lee, Mazmanian, and Perlow 121
&R.E.Quinn(Eds.), Positiveorganizational scholarship:
Foundations of a new discipline, 3: 263–278. San Francisco,
CA: Berrett-Koehler.
Dutton, J. E., & Ragins, B. R. E. 2007. Exploring positive
relationships at work: Building a theoretical and
research foundation. Mahwah, NJ: Lawrence Erlbaum
Associates.
Edmondson, A. 1999. Psychological safety and learning
behavior in work teams. Administrative Science
Quarterly, 44: 350–383.
Edmondson, A. C., & Lei, Z. 2014. Psychological safety:
The history, renaissance, and future of an interpersonal
construct. Annual Review of Organizational
Psychology and Organizational Behavior, 1: 23–43.
Eisenberg, E. M., & Witten, M. G. 1987. Reconsidering
openness in organizational communication. Academy
of Management Review, 12: 418–426.
Espinosa, J. A., Slaughter, S. A., Kraut, R. E., & Herbsleb,
J. D. 2007. Team knowledge and coordination in
geographically distributed software development.
Journal of Management Information Systems, 24:
135–169.
Fletcher, J. K., & Ragins, B. R. 2007. Stone center relational
cultural theory. In B. R. Ragins & K. E. Kram (Eds.),
Handbook of mentoring at work: Theory, research,
and practice: 373–399. Los Angeles, CA: SAGE.
Furnari, S. 2014. Interstitial spaces: Microinteraction settings
and the genesis of new practices between institutional
fields. Academy of Management Review,
39: 439–462.
Galinsky, A. D., Magee, J. C., Inesi, M. E., & Gruenfeld, D. H.
2006. Power and perspectives not taken. Psychological
Science, 17: 1068–1074.
Gioia, D. A., & Poole, P. 1984. Scripts in organizational
behavior.Academy ofManagementReview, 9: 449–459.
Gittell, J. H., Seidner, R., & Wimbush, J. 2010. A relational
model of how high-performance work systems work.
Organization Science, 21: 490–506.
Goffman, E. 1983. The interaction order: American Sociological
Association, 1982 presidential address.
American Sociological Review, 48: 1–17.
Herbsleb, J. D., & Mockus, A. 2003. An empirical study of
speed and communication in globally distributed
software development. IEEE Transactions on Software
Engineering, 29: 481–494.
Hinds, P. J., & Bailey, D. E. 2003. Out of sight, out of sync:
Understanding conflict in distributed teams. Organization
Science, 14: 615–632.
Hinds, P. J., & Cramton, C. D. 2014. Situated coworker
familiarity: How site visits transform relationships
among distributed workers. Organization Science,
25: 794–814.
Hinds, P. J., & Mortensen, M. 2005. Understanding conflict
in geographically distributed teams: The moderating
effects of shared identity, shared context, and spontaneous
communication. Organization Science, 16:
290–307.
Howard-Grenville, J., Golden-Biddle, K., Irwin, J., & Mao, J.
2011. Liminality as cultural process for cultural
change. Organization Science, 22: 522–539.
Ibarra, H. 1993. Personal networks of women and minorities
in management: A conceptual framework. Academy
of Management Review, 18: 56–87.
Isaacs, W. 1999. Dialogue: The art of thinking together.
New York, NY: John Wiley & Sons.
Jablin,F.M.1979. Superior–subordinate communication:The
state of the art. Psychological Bulletin, 86: 1201–1222.
Kahn, W. A. 2007. Meaningful connections. In J. E. Dutton
& B. R. Ragins (Eds.), Exploring positive relationships
at work: Building a theoretical and research foundation:
189–206. Mahwah, NJ: Lawrence Erlbaum
Associates.
Kellogg, K. C. 2009. Operating room: Relational spaces and
microinstitutional change in surgery. American Journal
of Sociology, 115: 657–711.
Kipnis, D. 1972. Does power corrupt? Journal of Personality
and Social Psychology, 24: 33–41.
Lewin, K. 1947. Frontiers in group dynamics: Concept,
method and reality in social science: Social equilibrium
and change. Human Relations, 1: 5–41.
Liu, D., Liao, H., & Loi, R. 2012. The dark side of leadership:
A three-level investigation of the cascading
effect of abusive supervision on employee creativity.
Academy of Management Journal, 55: 1187–1212.
Lord, R. G., & Kernan, M. 1987. Scripts as determinants of
purposeful behavior in organizations. Academy of
Management Review, 12: 265–277.
Martin, J., Knopoff, K., & Beckman, C. 1998. An alternative
to bureaucratic impersonality and emotional labor:
Bounded emotionality at The Body Shop. Administrative
Science Quarterly, 43: 429–469.
Mayes, B. T., & Allen, R. W. 1977. Toward a definition of
organizational politics. Academy of Management
Review, 2: 672–678.
Metiu, A. 2006. Owning the code: Status closure in distributed
groups. Organization Science, 17: 418–435.
Morrison, E. W. 2011. Employee voice behavior: Integration
and directions for future research. Academy of
Management Annals, 5: 373–412.
Morrison, E. W., & Milliken, F. J. 2000. Organizational silence:
A barrier to change and development in a pluralistic
world. Academy of Management Review, 25:
706–725.
122 Academy of Management Journal February
Nembhard, I. M., & Edmondson, A. C. 2006. Making it safe:
The effects of leader inclusiveness and professional
status on psychological safety and improvement efforts
in health care teams. Journal of Organizational
Behavior, 27: 941–966.
O’Reilly, C. A., III, Caldwell, D. F., & Barnett, W. P. 1989.
Work group demography, social integration, and turnover.
Administrative Science Quarterly, 34: 21–37.
O’Reilly, C. A., & Roberts, K. H. 1977. Task group structure,
communication, and effectiveness in three organizations.
Journal of Applied Psychology, 62: 674–681.
Phillips, K. W., Rothbard, N. P., & Dumas, T. L. 2009.
To disclose or not to disclose? Status distance and
self-disclosure in diverse environments. Academy of
Management Review, 34: 710–732.
Polletta, F. 1999. “Free spaces” in collective action. Theory
and Society, 28: 1–38.
Quinn, R. W. 2007. Energizing others in work connections.
In J. E. Dutton & B. R. Ragins (Eds.), Exploring positive
relationships at work: Building a theoretical and
research foundation: 73–90. Mahwah, NJ: Lawrence
Erlbaum Associates.
Rogers, D. P. 1987. The development of a measure of perceived
communication openness. Journal of Business
Communication, 24: 53–61.
Rogers, K. M., Corley, K. G., & Ashforth, B. E. 2016. Seeing
more than orange: Organizational respect and positive
identity transformation in a prison context. Administrative
Science Quarterly, 62: 219–269.
Senge, P. 1990. The fifth discipline: The art and practice
of the learning organization. New York, NY: Currency
Doubleday.
Shaw, M. 1981. Group dynamics.NewYork, NY: McGraw-
Hill.
Stephens, J. P., Heaphy, E. D., Carmeli, A., Spreitzer, G. M.,
& Dutton, J. E. 2013. Relationship quality and virtuousness:
Emotional carrying capacity as a source of
individual and team resilience. Journal of Applied
Behavioral Science, 49: 13–41.
Strauss, A., & Corbin, J. 1998. Basics of qualitative research:
Techniques and procedures for developing
grounded theory. Thousand Oaks, CA: SAGE.
Sutton, R. I. 2007. The no asshole rule. London, U.K.:
Sphere.
Thaler, R. H., Sunstein, C., & Balz, J. P. 2012. Choice architecture.
In E. Shafir (Ed.), The behavioral foundations
of public policy: 1–13. Princeton, NJ:
Princeton University Press.
Tourish, D., & Robson, P. 2006. Sensemaking and the distortion
of critical upward communication in organizations.
Journal of Management Studies, 43:
711–730.
Vogus, T. J. 2004. In search of mechanisms: How do HR
practices affect organizational performance? (Unpublished
doctoral dissertation). University of Michigan,
Ann Arbor.
Weick, K. E. 1993. The collapse of sensemaking in organizations:
The Mann Gulch disaster. Administrative
Science Quarterly, 38: 628–652.
Zietsma, C., & Lawrence, T. B. 2010. Institutional work in
the transformation of an organizational field: The interplay
of boundary work and practice work. Administrative
Science Quarterly, 55: 189–221.
Michael Y. Lee (mike.lee@insead.edu) is an assistant professor
of organizational behavior at INSEAD. His research
focuses on novel and innovative approaches to organizing
that facilitate individual, team, and organizational
effectiveness.
Melissa A. Mazmanian (m.mazmanian@uci.edu) is an
associate professor of Informatics at the School of Information
and Computer Sciences, University of California,
Irvine. Her research revolves around the use of
information and communication technologies in relation
to daily work practice, communication patterns, and the
experience of personal and professional time in the digital
age.
Leslie A. Perlow (lperlow@hbs.edu) is the Konosuke Matsushita
Professor of Leadership in the Organizational Behavior
Unit at Harvard Business School. Her research
focuses on the microdynamics of work. She documents
existing work patterns, identifies ways teams and organizations
can change their work patterns, and studies the
change process.
2020 Lee, Mazmanian, and Perlow 123