ResearchPDF Available

Distributed Attention and Shared Emotions in the Innovation Process: How Nokia Lost the Smartphone Battle

Authors:

Abstract and Figures

Forthcoming in Administrative Science Quarterly 2015:We conducted a qualitative study of Nokia to understand its rapid downfall over the 2005–2010 period from its position as a world-dominant and innovative technology organization. We found that top and middle managers’ shared emotions during the smartphone innovation process caused cycles of behaviors that harmed both the process and its outcome. Together, organizational attention structures and historical factors generated various types of shared fear among top and middle managers. Top managers were afraid of external competitors and shareholders, while middle managers were mainly afraid of internal groups, including superiors and peers. Top managers’ externally focused fear led them to exert pressure on middle managers without fully revealing the severity of the external threats and to interpret middle managers’ communications in biased ways. Middle managers’ internally focused fear reduced their tendency to share negative information with top managers, leading top managers to develop an overly optimistic perception of their organization’s technical capabilities and neglect long-term investments in developing innovation. Our study contributes to the attention-based view of the firm by describing how distributed attention structures influence shared emotions and how such shared emotions can hinder the subsequent integration of attention, influencing innovation processes and outcomes and resulting in temporal myopia—a short-term focus on product innovation at the expense of long-term product development. Keywords: innovation process, shared emotion, cognition, attention-based view of the firm, smartphone, temporal myopia
Content may be subject to copyright.
Administrative Science Quarterly
1–43
ÓThe Author(s) 2015
Reprints and permissions:
sagepub.com/
journalsPermissions.nav
DOI: 10.1177/0001839215606951
asq.sagepub.com
Distributed Attention
and Shared Emotions
in the Innovation
Process: How Nokia
Lost the Smartphone
Battle
Timo O. Vuori
1
and Quy N. Huy
2
Abstract
We conducted a qualitative study of Nokia to understand its rapid downfall over
the 2005–2010 period from its position as a world-dominant and innovative
technology organization. We found that top and middle managers’ shared emo-
tions during the smartphone innovation process caused cycles of behaviors that
harmed both the process and its outcome. Together, organizational attention
structures and historical factors generated various types of shared fear among
top and middle managers. Top managers were afraid of external competitors
and shareholders, while middle managers were mainly afraid of internal groups,
including superiors and peers. Top managers’ externally focused fear led them
to exert pressure on middle managers without fully revealing the severity of the
external threats and to interpret middle managers’ communications in biased
ways. Middle managers internally focused fear reduced their tendency to share
negative information with top managers, leading top managers to develop an
overly optimistic perception of their organization’s technological capabilities and
neglect long-term investments in developing innovation. Our study contributes
to the attention-based view of the firm by describing how distributed attention
structures influence shared emotions and how such shared emotions can hinder
the subsequent integration of attention, influencing innovation processes and
outcomes and resulting in temporal myopia—a focus on short-term product
innovation at the expense of long-term innovation development.
Keywords: innovation process, shared emotion, cognition, attention-based
view of the firm, smartphone, temporal myopia
Innovation requires highly effective information sharing between various organi-
zational groups. For example, top managers are expected to translate prioritized
1
Department of Industrial Engineering and Management and Department of Management Studies,
Aalto University
2
INSEAD, Singapore
strategic themes into actionable goals for middle management, while the latter
are expected to coordinate with each other and report upwards on the progress
of implementation to enable corrective actions (e.g., March and Simon, 1958;
Ethiraj and Levinthal, 2004). Such integration of attention between organiza-
tional groups (Joseph and Ocasio, 2012) is important for innovation because of
interdependencies relating to a host of factors, including diverse units’ tasks,
technologies, organizational capabilities, and changing market expectations
(e.g., Garud, Tuertscher, and Van de Ven, 2013).
Prior scholars have studied the innovation process and the related communi-
cation and integration of attention mainly from structural (e.g., Henderson and
Clark, 1990; Ocasio, 1997, 2011) and cognitive perspectives (e.g., Gavetti et al.,
2012; Eggers and Kaplan, 2013), but these theoretical perspectives do not fully
explain why various organizational groups often find it hard to integrate their
knowledge through communication. Though the generation of structural channels
to enable interaction between groups such as top and middle managers might
foster comprehensive information sharing between them (Joseph and Ocasio,
2012), the mere existence of such channels may not be sufficient, as people may
still avoid bringing up sensitive issues or refuse to listen to others during their
interaction episodes. Several case studies have described the challenges of com-
panies using various structural arrangements and have alluded to the persistence
of poor mutual understanding between top and middle managers in the innova-
tion process (Leonard-Barton, 1992; Tripsas and Gavetti, 2000; Repenning and
Sterman, 2002; Gilbert, 2005). In a recent noteworthy example, President Obama
said, ‘‘I would not have launched Healthcare.gov if I had known it wasn’t going to
work’’ (The Verge, 2013), while many managers in his administration seemed to
have been long aware of the platform’s technical problems (Economist, 2013;
Washington Post, 2013). Research has also shown that middle managers’ under-
standings often evolve to be quite different from top managers’ perceptions and
intentions (Balogun and Johnson, 2004; Huy, Corley, and Kraatz, 2014).
In addition to being influenced by structures and cognition, intergroup com-
munication processes could be influenced by emotions, which have remained
underexamined in research on organizational innovation (see Garud,
Tuertscher, and Van de Ven, 2013; Anderson, Potoc
ˇnik, and Zhou, 2014, for
reviews). To the extent that the innovation process or its outcomes can have
major consequences for people’s well-being, various groups participating in
innovation may experience strong emotions (cf. Lazarus, 1991; Elfenbein,
2007). Basic research has shown that emotions influence people’s choices and
behaviors (e.g., Izard, 2009; Phelps, Lempert, and Sokol-Hessner, 2014).
Emotions also influence social processes in substantial ways (e.g., Hareli and
Rafaeli, 2008; Niedenthal and Brauer, 2012), and scholars have found that emo-
tions can significantly influence organization members’ thinking and behavior
related to strategy implementation (Huy, 2011; Huy, Corley, and Kraatz, 2014).
Furthermore, emotions often come to be shared within organizational groups
because group members attend to similar things and share their emotions
socially (e.g., Elfenbein, 2014; Menges and Kilduff, 2015). At the same time,
there could be intergroup differences, as each group is primarily focused on its
own task and well-being (e.g., Cyert and March, 1963; Ocasio, 1997). Shared
emotions could thus provide a complementary mechanism for understanding
how organizational groups communicate, coordinate, and act during the innova-
tion process, influencing its outcomes.
2Administrative Science Quarterly XX (2015)
One of the dominant lenses for organizational innovation has its origins in
the Carnegie School (e.g., Simon, 1947; March and Simon, 1958), which in
many ways has sought to answer the question, ‘Given that individuals are
boundedly rational, how should we design organizations?’ A key insight has
been that organizations need both specialization (leading to differentiation) and
integration. Specialization is needed because no individual can handle every-
thing, while integration is needed to combine different viewpoints (see also
Ocasio, 1997; Joseph and Ocasio, 2012). Different levels of differentiation and
integration seem to lead to varying problems. For example, as specialization
increases, employees and managers have to escalate more issues further
upward, which can slow down innovation (Burns and Stalker, 1961). And when
communication channels are optimized for the refinement of current products,
various units may fail to integrate their views during the design of radically new
products (Henderson and Clark, 1990), while others may not see a critical need
for them (Christensen and Bower, 1996).
Unfortunately, the theoretical mechanisms underlying the structural–
cognitive perspective have remained decoupled from emotions. Scholars have
analyzed organizations as systems whose elements process information in an
affect-free way and have remained largely indifferent to how various organiza-
tion members might feel differently about the information they use or react to
it in different ways (see Gavetti et al., 2012, for a review). Yet emotions could
become relevant to the extent that people see the serious implications of the
information they are processing, which triggers emotions when those implica-
tions have personal relevance for them (e.g., Lazarus, 1991). As innovation
often has significant personal relevance, participants in the innovation process
could experience strong emotions in regard to the information they process
and could also influence each other’s emotions as they grapple with various
important situations (cf. Hareli and Rafaeli, 2008). Even though many works
have hinted that emotional reactions might influence people’s behavior inside
structural communication channels during the innovation process (e.g., Burns
and Stalker, 1961; Fang, Kim, and Milliken, 2013; Reitzig and Maciejovsky,
2014), very few, if any, empirical field studies have examined how various
groups’ emotions emerge during the innovation process and what the impacts
are. Hence a deeper and more holistic understanding of the innovation process
would require a joint investigation of how both emotional reactions and
bounded rationality influence the innovation process and its outcomes.
We carried out an inductive study of Nokia’s failure to produce a next-
generation smartphone in response to Apple’s iPhone. This case allowed us to
develop a deeper understanding of the emergence of shared emotions during
the innovation process and their influence on innovation because it represents
an extreme case for theory building. Rapid changes in Nokia’s competitive envi-
ronment put severe pressure on firm members and amplified their emotions,
making the influence of those emotions particularly salient.
INNOVATION AND EMOTION
We follow Elfenbein (2007: 315) in viewing emotion as a process that ‘‘begins
with a focal individual who is exposed to an eliciting stimulus, registers the sti-
mulus for its meaning, and experiences a feeling state and physiological
changes, with downstream consequences for attitudes, behaviors, and
Vuori and Huy 3
cognitions, as well as facial expressions and other emotionally expressive cues.’
A distinguishing characteristic of emotions is that they have a target, which is a
crucial element because the action tendencies associated with emotions relate
to the target of the emotion (e.g., Russell, 2003; Damasio and Carvalho, 2013).
For example, fear can make people either get a vaccination or avoid one, depend-
ing on whether the target of their fear is the disease to be prevented or potential
harmful side effects of the vaccination itself. Hence, to understand how emo-
tions might influence the innovation process, it is important to investigate the tar-
get of the emotion in addition to observing whether the emotion is present.
Emotions activate action tendencies rather than fixed actions (Ellsworth and
Scherer, 2003). For example, fear can lead to increased heart and respiratory
rates, freeze or flight behaviors, and ‘‘attentional behaviors,’ leading to atten-
tion to the cause (Damasio and Carvalho, 2013: 145; see also Lazarus, 1991;
Ellsworth and Scherer, 2003). Action tendencies’ main purpose is to help the
individual achieve a favorable person–environment relationship in the particular
situation (Lazarus, 1991; see also Damasio and Carvalho, 2013). By directing
attention to the particular target that triggers the emotion, emotion facilitates
actions that address that particular target, while also temporarily reducing
attention to other matters. Even though emotional reactions are usually adap-
tive, they sometimes cause people to prefer small short-term benefits over
large long-term benefits (e.g., Phelps, Lempert, and Sokol-Hessner, 2014) or to
take actions that avoid tolerable short-term harm but risk causing catastrophic
long-term consequences (e.g., Kahneman, 2011: Part 4).
The emotion that people experience toward a particular target depends on how
they perceive that target’s implications for themselves (e.g., Ellsworth and
Scherer, 2003; Russell, 2003; Izard, 2009). The target need not be physically pres-
ent: recalled and imagined targets can also trigger emotion, as can abstract con-
cepts such as money and power (e.g., Ellsworth and Scherer, 2003; Damasio and
Carvalho, 2013). Appraisal theories of emotion suggest that people appraise spe-
cific dimensions in a target that determine the emotion they come to experience
(Ellsworth and Scherer, 2003). Lazarus (1991) suggested that the appraisal process
has two stages: a primary and secondary stage. For example, people are likely to
experience fear when their primary appraisal suggests potential harm to their well-
being and their secondary appraisal suggests that the outcome of the situation is
uncertain or beyond their control. Even though emotion theories do not fully agree
on the details of the process through which people come to experience emotions,
there is enough alignment that appraisal theories give us an effective tool for
understanding and inferring how and why people experience emotions during the
innovation process (see also Elfenbein, 2007, 2014; Menges and Kilduff, 2015).
Emotional states can be fleeting, such as the fear triggered by a sudden loud
noise, or longer lasting, such as the fear of terrorism (e.g., Izard, 2009). When
an emotion is said to be longer lasting, it usually refers to people experiencing
the same emotion multiple times toward the same target over a long period,
repeating the appraisal process continuously until the situation is resolved
(Ellsworth and Scherer, 2003). In organizational settings, emotional states that
recur consistently over time are likely to have a more systematic effect on the
innovation process than occasional fleeting emotions: though the former have
an iterative, cumulative influence, the latter may be too rare to form accumulat-
ing patterns and may also have partly opposing influences that cancel each
other out when emotions fluctuate.
4Administrative Science Quarterly XX (2015)
Fear, Threat Rigidity, and Shared Emotions during the Innovation Process
The innovation process can be laden with emotion in part because of the uncer-
tain, prospective, yet highly consequential nature of the innovation journey.
Innovators are trying to create things that do not yet exist by leveraging technolo-
gical components of uncertain market potential. Because of this prospective
orientation, future-oriented emotions such as hope and fear (Baumgartner,
Pieters, and Bagozzi, 2008) are particularly likely to arise. Negative emotions tend
to command more attention than positive ones because they evolved to help
people ensure their survival, or at least avoid harm (Baumeister et al., 2001). As a
result, fear—as a future-oriented negative basic emotion—could play a critical
role in the innovation process (see Baumgartner, Pieters, and Bagozzi, 2008;
Izard, 2009). A second reason that we might expect fear to be central to innova-
tion is that other past-oriented negative emotions that people may feel during
the innovation process—such as shame or envy—can turn into fear. When peo-
ple experience such a negative past-focused emotion and then imagine a similar
future situation, they may fear the same negative outcome occurring again (cf.
Dane and George, 2014). For example, one can feel shame after having given a
bad presentation and therefore fear that the next presentation will also fail.
Because fear is related to the perception of threat, the literature on threat
rigidity could inform us. The original threat-rigidity theory hypothesized some
mechanisms that cause organizations to respond in a rigid way to an external
threat (Staw, Sandelands, and Dutton, 1981), including increased emotional
arousal, which narrows top managers’ thinking, constrains communication, and
makes middle managers more dependent on them. The hypothesized mechan-
isms of the threat-rigidity theory remain largely speculative, however, and have
not been empirically verified. The original theory assumes that the whole organi-
zation would perceive the threat in a similar way and develop homogeneous
cognitive-emotional reactions that cause collective rigidity. Empirical tests have
measured only the presence of the threat perceptions with proxies such as
return on assets (Greve, 2011), bargaining cycles (Griffin, Tesluk, and Jacobs,
1995), changes in the studied firms’ funding sources (D’Aunno and Sutton,
1992), and changes in the performance of an acquired unit (Shimizu, 2007)
rather than various groups’ actual perceptions of the threat and resulting beha-
vior inside the organization. This is problematic because some case studies have
suggested that various units in the same organization could differ in their per-
ceptions of whether a new technology constitutes a threat or an opportunity—
or whether it is simply irrelevant (Tripsas and Gavetti, 2000; Gilbert, 2005).
Gilbert (2005) showed that different threat perceptions influenced how indepen-
dent units performed their tasks. These differing perceptions could also influ-
ence the emotions people feel in the different units and how these units
consequently interact with one another in the organizational innovation process.
Such differences in threat perceptions among organizational groups might arise
due to the influence of their varied positions in the organizational structure. If
groups specialize in different tasks and focus on different matters, they probably
perceive things differently and regard some matters as more important than
others. Differing emotions between groups may thus arise (cf. Lazarus, 1991)
as a result of the structural distribution of attention (Ocasio, 1997).
Because of the social psychology of organizational structures, people may
also perceive threats to their status and power within the focal structure, which
Vuori and Huy 5
can trigger strong emotions. A key characteristic of the organizational hierarchy
is that it confers unequal formal status on various organization members
through titles, roles, and responsibilities (e.g., Simon, 1947; Magee and
Galinsky, 2008). Such formal status interacts with informal status to determine
individuals’ power— the extent of their control over resources that other mem-
bers value (Pfeffer, 1981). Many organization members thus value status and
compete with one another to reach or maintain high organizational status
(Rosenbaum, 1979; Magee and Galinsky, 2008) and may feel strong emotions
if they perceive related threats (cf. Lazarus, 1991). Less is known, however,
about how such potential emotions related to seeking status influence their
behaviors during the innovation process.
People’s status also influences the emotions they experience (Ellsworth and
Scherer, 2003: 580). In particular, low-status people may have a hard-wired,
evolutionary-based fear of high-status people because, in the past, individuals
with higher status would have controlled resources critical for survival (Kish-
Gephart et al., 2009). Status differences can also determine the sequence of
emotions felt among individuals (cf. Hareli and Rafaeli, 2008). For example, sta-
tus can turn anger into fear: if an employee expresses anger toward a superior,
he or she is likely to respond in kind, which causes the employee to feel fear
(cf. van Kleef, De Dreu, and Manstead, 2004). Status differences could thus
cause people to differ in how they construe a threat and to experience different
emotions, while people of similar status are more likely to perceive a threat in a
similar way and to experience similar emotions.
Likewise, though the structural distribution of attention likely generates
between-group differences in emotions, it could foster within-group similari-
ties. Members of the same group likely attend to similar things and perceive
that those things have similar implications for them. Group members may
thus experience emotions similar to their peers consistently over time
regarding issues that they see as key to the group’s welfare (Elfenbein,
2014; Menges and Kilduff, 2015). For example, if a group is exposed to fast
time pacing in product development, members may develop a shared fear of
missing deadlines (cf. Brown and Eisenhardt, 1997). Thus, just as group
members may share similar mental models (Mathieu et al., 2000; Healey,
Vuori, and Hodgkinson, 2015), they may also share similar emotions (see
Mackie, Devos, and Smith, 2000; Huy, 2011; Elfenbein, 2014; Menges and
Kilduff, 2015).
Shared emotions would cause many individuals to experience similar action
tendencies (e.g., Elfenbein, 2014; Menges and Kilduff, 2015) and might make
them behave in similar ways as a result of socialization processes (cf. Smith,
Seger, and Mackie, 2007; Huy, 2011). But little empirical research has exam-
ined whether and how macro-level organizational structures and other factors
generate similar or different emotions within and between key organizational
groups and how such shared emotions influence the organizational innovation
process and its outcome.
METHOD
We analyzed Nokia’s rise and fall in the smartphone industry between 2005
and 2010. By the end of 2007, half of all smartphones sold in the world were
Nokias, while Apple’s iPhone had a mere 5 percent share of the global market
6Administrative Science Quarterly XX (2015)
(Gartner, 2009). Nokia had bountiful economic and intellectual resources, had
dominated the industry for years with a stream of innovative models, and was
frequently put forward as a world-class exemplar of strategic agility (Doz and
Kosonen, 2008; Steinbock, 2010). But by the fall of 2010, despite significant
efforts, Nokia had failed to introduce a smartphone to match the iPhone, creat-
ing the perception that it was losing the battle against Apple. Nokia replaced its
chief executive officer (CEO), abandoned software development, and became
a mere hardware provider. It surrendered its position as the leading smart-
phone provider and ultimately exited the mobile phone business.
Context
The mobile phone industry has grown rapidly during the past 20 years, and its
trajectory has been divided into several distinct technological phases. By 2005,
the industry was moving toward third-generation (3G) radio technologies that
enabled faster Internet connections. In 2007, Apple introduced the iPhone,
which lacked 3G support but still promised mobile Internet, in the eyes of the
mass market, with its large touch screen and advanced user interface enabled
by iOS, an exclusive operating system (OS) based on the OS Apple had been
developing for years for its computers. This move was a major discontinuity for
the whole industry: for the first time, differentiation lay in software rather than
radio technology—see table 1 for key market events and changes in market
shares. This shift presented a growth opportunity to companies such as Apple
and Google, which were already strong in software development; other tradi-
tional players had to scramble to develop software capabilities that were radi-
cally new to them. Samsung, seemingly the only traditional player to prosper
after the software revolution, chose to adopt Google’s Linux-based Android OS
(an open-source software). But Nokia, hoping to ensure sustainable profit,
decided to continue improving its proprietary OS.
Top managers’ belief that Nokia should remain autonomous in software
terms was supported by its past achievements (Cord, 2014). Even its early
phone models had software-enabled functions such as phone books and text
messaging. By 2005, Nokia’s primary smartphone OS was Symbian, which
both internal and external developers generally described as difficult to work
with. Additional complexity arose from Nokia’s simultaneous development of
Internet services to be integrated with Symbian. Nokia had also been exploring
the use of Linux in mobile devices. The result, an OS named MeeGo, was
intended to replace Symbian after several product generations. Nokia’s top
managers thus believed it had all the elements in place to develop good soft-
ware and remain the leading smartphone provider. But consumers’ responses
to Nokia’s phones became increasingly negative between 2008 and 2010
because the improvements in product quality seemed modest compared with
challengers’ offerings.
Data Collection
Our primary data came from private interviews conducted between November
2012 and February 2014. We carried out a total of 76 interviews with Nokia’s
top managers (TMs), middle managers (MMs), engineers, and external experts,
as detailed in table 2. Consistent with Huy (2001, 2011), we defined MMs as
Vuori and Huy 7
those who were two levels below the CEO and one level above engineers and
front-line workers. The interviews were carried out in three rounds, and we
analyzed the data between each round to inform subsequent interviews.
Several key informants who were closely involved with the mobile-phone busi-
ness, including the CEO, vice presidents, R&D managers, and strategy direc-
tors, were interviewed multiple times.
1
We purposefully sampled the
Table 1. Market Shares and Key Events in the Mobile Phone Sector (including Smartphones),
2007–2010*
Company 2007 2008 2009 2010
Nokia 38% (49%) 36% (41%) 36% (39%) 29% (34%)
Apple x% (3%) 1% (7%) 2% (14%) 3% (16%)
Samsung 13% (x%) 15% (4%) 20% (4%) 18% (8%)
RIM (BlackBerry) x% (10%) 2% (16%) 3% (21%) 3% (16%)
Sony Ericsson 9% (x%) 7% (3%) 5% (3%) 3% (3%)
Motorola 14% (x%) 8% (3%) 5% (4%) 2% (5%)
HTC x% (3%) 1% (5%) 1% (5%) 2% (8%)
LG 7% (x%) 8% (2%) 10% (2%) 7% (2%)
Key market events during the year
Apple announces (Jan.)
and launches iPhone
(2.5G radio technology)
in the U.S. (June) and
UK, France, and
Germany (Nov.)
Nokia launches N95
smartphone (with 3G
technology) in March
Nokia announces Ovi
Internet store for
software applications
(Aug.)
Google announces
Android OS (Nov.)
Apple launches iPhone
3G in 26 countries and
AppStore (July)
Nokia launches beta
version of Ovi (Aug.)
Nokia launches its first
touch-phone, 5800
(Oct.)
First Android phone, HTC
Dream, launched (Oct.)
Google announces (Aug.)
and releases (Oct.)
Android Market
Internet store
Nokia announces first
"iPhone killer" N97
(QWERTY keyboard +
touch screen) in Dec.
Apple launches
iPhone 3GS
(June)
Nokia launches
N97 (June)
Samsung
launches its
first Android
phone, Galaxy
(June)
Apple launches
iPhone 4 (June)
Samsung
launches Galaxy
S Android
phone (June)
Nokia announces
(April) and
launches (Oct.)
first high-end,
touch-only
phone to match
iPhone (N8)
*Data are presented as ‘‘all mobile phones (only smartphones)’’; ‘x%’’ means that the market share was too small
to be reported (i.e., included in the category ‘‘others’’). Sources: http://www.gartner.com/newsroom/id/612207;
Chaplinsky and Marston (2011); http://www.gartner.com/newsroom/id/1543014; http://www.gartner.com/
newsroom/id/910112; http://www.quirksmode.org/blog/archives/2011/02/smartphone_sale.html.
1
Nokia had two CEOs during the period of our study. When we speak of ‘‘the CEO,’’ we refer to
the second one, whose tenure as CEO lasted from 2006 to 2010. He worked in several executive
positions at Nokia before becoming CEO. When we speak of ‘‘the chairman,’’ we refer to the first
CEO, whose tenure as CEO lasted from 1992 to 2006. He was chairman of the board from 1999 to
2012. While the chairman was less active in daily management from 2006 onward, we bring up his
role because (1) our informants highlighted how his behaviors had shaped Nokia’s culture, (2) he
still exerted influence as chairman until the end of our study period, and (3) the dynamics that we
describe in the findings had already started during his tenure as CEO.
8Administrative Science Quarterly XX (2015)
interviewees to get both top and middle managers’ descriptions and interpreta-
tions of how their interactions had unfolded and influenced organizational
actions, as well as the perspectives of engineers. We contacted all of them
personally and assured anonymity. An average interview lasted about 80 min-
utes. All the interviews were audio-recorded and transcribed, apart from seven
for which careful notes were taken. Our interview protocol is shown in Online
Appendix A (http://asq.sagepub.com/supplemental). We also conducted over
200 informal conversations and follow-up conversations with the interviewees
and other Nokia informants.
As this is a retrospective field study, a particular concern was reducing the
risk of ex-post rationalization. First, we asked various informants from multiple
hierarchical levels to describe and interpret the same concrete events sepa-
rately, which allowed us to validate the plausibility of their accounts. For
Table 2. Informants and Interviews
Level Person 1st round 2nd round 3rd round
Validated
inductive model?*
0 Chief executive officer 1 1 3 yes
1 Executive vice president (EVP) 1 yes
1 Executive vice president 2 yes
1 Executive vice president1 1 (yes)
2 Technology officer 1 no
2 Senior vice president 1 2 yes
2 Senior vice president 1
2 Senior vice president 1
2 Senior vice president 1
3 Corporate vice president 1 1 1 yes
3 Vice president 1 yes
3 Design vice president 1
3 Product vice president 1 2 yes
3 Product vice president 1
3 Software vice president 1
3 Software vice president 1
3 Software vice president 1 1 yes
3 Software vice president 1 yes
4 Product unit (PU) directors (two individuals) 2 1 1 yes
4 Directors (three individuals) 3 1 1 yes
4 HR directors (two individuals) 1 2
4 Market intelligence director 1 yes
4 Software directors (eight individuals) 5 4 2 3 x yes
1xno
4 Strategy directors (three individuals) 2 2 2 2 x yes
5 Radio engineers (four individuals) 4
5 Software manager 1
5 Strategy managers (four individuals) 2 3 2 yes
5 Supply-chain managers (two individuals) 2
X Top-level consultants (three individuals) 3 2 x yes
*‘‘–’’ means that we did not show the model to the individual due to timing (early stage of the study) or schedule
(short interview).
This EVP emailed us his/her thoughts multiple times and confirmed that top managers lacked understanding of
technological capability, but his/her answers are not included in the count of interviews.
Vuori and Huy 9
example, top managers told us they pressured middle managers for faster per-
formance, and middle managers also told us how they felt pressured by top
managers. Second, we relied on informants who were particularly knowl-
edgeable about the relevant events and for whom the events were person-
ally important, thus improving memory accuracy. Third, we used specific
interview techniques such as precise ‘‘courtroom’’ questioning and event
tracking, which provide accurate and convergent information among infor-
mants (Eisenhardt, 1989). We asked individuals to describe concrete exam-
ples and events, obliging them to rely on their episodic memory, which
provides more comprehensive accounts (Tulving, 2002) and increases recall
accuracy (cf. Fisher, Geiselman, and Amador, 1989; Miller, Cardinal, and
Glick, 1997; Fisher, Ross, and Cahill, 2010). Episodic memories, which may
rely on a separate memory system, complement more-generic recollections
of mental states because they contain specific details of not only what was
happening or felt but also exactly when and where (Tulving, 2002). In this
way our data were triangulated, making them more trustworthy. For exam-
ple, it is more comprehensive when a person remembers a specific instance
of when and where he or she felt fear, rather than simply recalling having felt
generally afraid in the past. Episodic memories may also be more reliable
than retrospective survey measures because they do not oblige informants
to select a discrete category to characterize their memory. We also asked
informants to describe factual manifestations of mental states, which are
more robust to retrospective biases than memories of mental states (e.g.,
Miller, Cardinal, and Glick, 1997). Fourth, we validated subjective accounts
with factual and prospective sources when possible; we reviewed nearly
1,000 articles and books about Nokia, and several informants also shared
confidential internal reports, presentation slides, e-mails, and notes that cor-
roborated our findings.
Data Analysis
We used the method described by Gioia, Corley, and Hamilton (2013) to guide
our data analysis. Novel understanding can often be gained by carefully investi-
gating how various participants of an organizational process experience events,
and the authors suggested some practices that bring ‘‘qualitative rigor.’’ As is
typical of inductive research, our analytical process was iterative and over-
lapped with data collection, but several phases can be recognized. During these
phases, we developed increasingly refined inferences of novel theoretical
mechanisms from our data.
We started with open coding (Strauss and Corbin, 1998), which included
reading the interview transcripts and marking codes to describe the content of
the interviews. We coded nearly 2,000 segments of data. The initial codes cov-
ered various topics, such as ‘MMs criticizing TMs for being ‘lost’,’’ but we also
included emotion-related themes such as ‘perceived lack of emotionally
appealing communication.’’ We further categorized such first-order codes into
more-abstract theoretical dimensions. We also wrote numerous memos during
the process to develop theoretical insights.
As we iterated between coding and data, our theorizing evolved in four main
steps: (1) Nokia’s failure to innovate caused by TMs’ inaccurate understanding
10 Administrative Science Quarterly XX (2015)
of capabilities; (2) Nokia’s failure to adjust during the innovation process caused
by inauthentic TM–MM interactions; (3) the interim outcome, temporal
myopia—excessive focus on short-term innovation even though long-term
activities might lead to more beneficial outcomes (Levinthal and March,
1993)—caused by TMs’ externally focused fear and MMs internally focused
fear and low external fear; and (4) Nokia’s attention structures that in part eli-
cited such fears, which created decoupling interactions between TMs and
MMs, causing temporal myopia.
We refined our coding procedures according to our evolving understanding
(Strauss and Corbin, 1998). To code emotions more systematically, we used
the appraisal theories of emotions (Lazarus, 1991; Ellsworth and Scherer, 2003;
see also Huy, 2011; Huy, Corley, and Kraatz, 2014), which indicate that emo-
tions result from specific appraisal processes and each basic emotion is associ-
ated with a prototypical appraisal pattern. For example, fear is elicited when
people perceive potential harm in their relationship with their environment and
consider their coping potential as low or uncertain. We thus inferred fear when
people described their concern that undesirable things might happen to them—
that is, uncertain future harm—and did not feel they had full control over the
threat or the ability to eliminate it. In addition, the informants sometimes expli-
citly described various emotions they had felt, by using phrases such as ‘I was
afraid that . . .,’ or they recalled physiological reactions such as shaking or a
racing heartbeat and their context, which allowed us to infer specific emotions.
When multiple informants described having experienced a similar emotion and/
or reported that colleagues experienced the same emotional reaction associ-
ated with a similar appraisal, we inferred that shared emotions were present.
Labeling emotions felt jointly but individually as ‘‘shared’’ is consistent with the
way ‘‘shared’ is used in research on shared mental models that uses a compo-
sitional view (see Kozlowski and Klein, 2000; Mathieu et al., 2000). Initially, we
recognized a variety of emotions from the informants’ accounts, but we ulti-
mately focused on fear because it was the emotion most commonly described,
and it explained most plausibly the key dynamics that we describe in our induc-
tive model.
Beyond coding, we also used a practice similar to constant comparisons in
grounded theory (Strauss and Corbin, 1998) to identify differences and similari-
ties among various data segments. For example, by comparing an MM’s fear
accounts with those of other MMs, we could see that they had experienced
similar emotions associated with similar appraisals, thus suggesting shared
emotions. In addition, when MMs’ expressed targets of fear differed from
those of TMs, we ultimately inferred the difference between internally focused
and externally focused fear and started to analyze their antecedents and conse-
quences. This iterative process resulted in the data structure presented in fig-
ure 1, which shows how the theoretical concepts we developed are grounded
in the empirical data.
We used a factual timeline of events, theoretical logic, and the informants’
descriptions, which contained various events at different points in time and
relationships between the events, to identify the temporal dynamic of our pro-
cess model, i.e., reflecting how some factors influence subsequent factors
over time. For example, we created several causal-loop diagrams to map the
links between concepts that were abstracted from the data. Likewise, we
Vuori and Huy 11
matched the informants’ fear-related statements with their descriptions of TM–
MM interactions through axial coding (Strauss and Corbin, 1998).
Checking the Model’s Validity
We conducted a three-step member-check process to increase the validity of
our inductive model (Shah and Corley, 2006). First, we presented the key ele-
ments of the model individually to 23 informants who were spread across differ-
ent groups and functions. As shown in table 2, 21 of these informants, including
TMs and software VPs and directors, concurred that our model accurately
Figure 1. Code-aggregation diagram.
1st Order Codes 2nd Order Themes Aggregate Dimensions
Middle managers’
low external fear
Shared fears
Middle managers’
high internal
fear
• MeeGo MM perceiving that iPhone is a bad phone
• Symbian MM feeling that Symbian is better than other OS
• Symbian MM saying numbers looked fine
• MM hearing EVP say iPhone not a threat
• MM says limited exposure to customers gives wrong feeling
• PU director fears losing job unless PU promises too much
• MeeGo MM feels other units will fight if MM points out problems
• Symbian MM fears TMs will shout if MM shares problems
• Heart racing in a meeting due to fear
Top managers’
high external fear
• TM’s intense fear reaction to iPhone
• TMs worried about competitors’ development
• TM thought “Oh shit” after Google presentation
• Strong short-term pressure for quarterly results
• TM’s fear about short-term Symbian sales
Structural,
attention-related
antecedents of
shared fear
Structural-
behavioral
antecedents of
shared fear
Middle managers’
dependence on
organizational status
• MM’s money and fame made him “not care” anymore
• MM could not afford to take a risk
• New person needs to adjust to the system
Intuitive fear
reactions
TMs’ history of
aggression
• “The chairman ruled with fear”
• Many seem to fear the chairman
• Outsiders report that people fear the chairman
• Chairman shouting often
• People share stories of the chairman‘s aggression
• EVP#3 described as an aggressive person
TMs’ and MMs’
different appraisals
Structural
distribution
of attention
• TM primary appraisal about external threat
• MM primary appraisal about internal threat
• MM has limited visibility to customer
• TM visited customers often
• MM describes how boss was the ”customer”
TMs’ over-optimistic
perception of
organizational
capabilities TM–MM assessment
gap
Little
corrective
action
• TM setting tough deadline
• TM not going to negotiate
• TM pressuring to make things faster
• TM favoring ”new people” who promise more
• TM thinks people are changing rapidly
• EVP thinks touchphone is almost ready, when delayed a lot
• EVP surprised by the real status of development
• Not reducing product-specific modifications for Symbian
• Not cutting extra work from Symbian
• Not hiring new software people for long-term development
Innovation
underperformance
(temporal
myopia)
Emphasis on devices
over software
• Software suffered due to a device launch
• Operating system fragmented due to many devices
• Resources moved from OS development to devices
Declining
product
quality
• N95 success in 2007
• Delays in the development of 5800 in 2008
• N97 fiasco in 2009
MMs’ over-optimistic
reporting
• Making demo to fake progress
• Hiding negative usability test results
• Waiting for someone else to speak first about problems
TM–MM decoupling
interaction
TMs’ low
technological
competence
• No software engineers in TMT
• Comments that CEO does not understand software
• TM admits having low software expertise
TM pressuring MM
for fast performance
12 Administrative Science Quarterly XX (2015)
described what had occurred at Nokia.
2
Second, we sent a four-page summary
of our key findings to 331 TMs and MMs who worked for Nokia during our study
period, with an invitation to send us feedback, ‘‘especially if you disagree with
the findings.’’ We received 37 written responses, all but three confirming our
key findings. Several MMs also shared additional examples of how they had felt
fear or how fear had prevented them from reporting negative information to
TMs, even when they knew it would have been the right thing to do. Third, we
conducted 27 follow-up phone calls and face-to-face conversations during the
revision process to double-check that our informants had indeed meant that they
had felt fear as an emotion (as opposed to using ‘‘fear’’ as shorthand for a cogni-
tive evaluation of a situation) when describing their own experiences and their
interpretations of others’ experiences. Twenty-two (81 percent) of them con-
firmed that they had indeed meant the emotion fear and that this emotion had
influenced their behavior. Hence Nokia informants overwhelmingly agreed with
the key elements of the inductive model that we present below, in particular the
role of externally focused and internally focused fear.
FINDINGS
Various types of fear experienced by Nokia’s TMs and MMs caused a decou-
pling interaction pattern between the two groups that harmed Nokia’s innova-
tion process. Two factors influenced TMs’ and MMs’ fears: Nokia’s structural
distribution of attention produced differences in appraisals and shared fears
among TM and MM groups, and TMs’ past aggressive behaviors—and widely
shared stories about them—triggered intuitive fear reactions in MMs because
they had a lower position in the organizational hierarchy than TMs.
Structural, Attention-related Antecedents of Shared Fears
Structural distribution of attention. Nokia’s organizational structure
included the division of labor between TMs and MMs such that the former had
a focus on the external environment and the latter on the implementation of
TMs’ directives (March and Simon, 1958; Joseph and Ocasio, 2012). TMs
focused on the external environment through direct interaction with customers
and competitors and also through staff channels (corporate market intelli-
gence). A TM told us that he had plenty of interactions with customers, espe-
cially large network operators such as AT&T and France Telecom, and the CEO
publicly admitted that he was ‘‘paranoid about all the competition’
(MarketWatch, 2007).
3
Nokia had clearly defined processes for managing
2
We showed a late version of the figures in this paper and explained them, after which the infor-
mants were asked whether they felt the model was a reasonable description of what had happened
at Nokia, in light of their own experiences. Twenty-one participants agreed with the model, and two
expressed doubts: one who had a significant role in radio-technology development before 2007 and
did not witness decoupling interactions between TMs and MMs in his or her area of activity, and
one who claimed that Nokia did not have problems in software quality.
3
When we refer to the data from our private interviews, we have anonymized the individuals such
that top managers (CEO and three executive vice presidents) are marked as TM, senior vice presi-
dents and vice presidents as UMM#1.#N (upper middle managers), and directors and managers
as MM#1.#N (middle managers). We also indicate the primary focus area of the individuals, such
as ‘‘software,’’ with the quotes when this has theoretical relevance. When we use data from public
sources, we refer to the actual titles of the people.
Vuori and Huy 13
emerging strategic issues (e.g., Kajanto et al., 2004) and strategic agility (Doz
and Kosonen, 2008), according to the CEO, even using ‘cultural anthropolo-
gists . . . to find out what’s important in people’s lives because everything
changes so quickly’ (Korn-Ferry Institute, 2010). Nokia’s active sensing of the
market was manifest in its response to the iPhone threat about one year before
its launch, which it knew about despite Apple’s famed secrecy. In other words,
cognitive inertia (Fransman, 1994; Tripsas and Gavetti, 2000) cannot be seen as
a plausible or sufficient explanation for Nokia’s failure. As members in senior
strategic positions at Nokia revealed:
We knew [the iPhone] was coming out about a year in advance. We had pretty good
specifications for it. . . . The first thing [a MM] flagged up [in a market review] was
that we didn’t have touch screens or touch-screen development. . . . That message
went right up to the top of the organization, and was well received too. [The CEO]
forwarded [the] email to his subordinates. Our market analysis showed that our big-
gest competitive weakness was the lack of touch-screen products; he agreed, and
wrote ‘‘Please take action on this.’ (MM#1, business)
Confidential prospective documents that were shown to us confirmed that the
TMs did indeed recognize this issue, gave it a high priority, and included it in
their requests to MMs. UMM#1, who was working close to the CEO, likewise
described how the CEO put clear pressure on MMs so that they would imple-
ment what he thought would be essential:
Oneofthefirstthings[theCEO]broughtupwasthetouchscreen....Hefeltitwas
thenextbigthing...Hebroughtitupwiththeexecutivegroupeverywayhecould.
And he spoke directly with technical middle management. ...Ineverysingle
executive-group meeting, they went over our outlook with the touch screen. And this
was right after he was made CEO [a year before iPhone launch] . . . I recall many
such cases vividly, where he brought up the right concern [what Nokia should do], he
discussed it directly with the technical middle and upper management, he went
straight to the topic, put pressure on people, put it up in all possible goals [for MMs],
followed up on it in every single meeting.
A TM further elaborated in a private interview with us that he saw it as TMs’
task to translate market expectations into clear goals for the organization: ‘‘The
tight schedule and challenges come from the markets. They put time pressure
on you, and you need to pass it on to the organization.’ Given that Nokia failed
to develop an innovative offering to match the iPhone, despite TMs’ intentions
and directives, we need to better understand how the implementation process
unfolded within Nokia’s structural and historical context.
MMs’ focus of attention was predominantly on the implementation of
TMs’ orders rather than the external environment and was split between
two key groups: product units (PUs) in charge of defining specific products and
software units in charge of developing Nokia’s two OSs, Symbian and MeeGo,
and modifying them to meet the needs of specific phone models based on
requests from the PUs. Although the primary communication channel flowed
from TMs to PUs to software units, the software units also had direct interac-
tions with TMs, both formally and informally. This suggests that a purely struc-
tural explanation—lack of opportunities for interaction (cf. Henderson and Clark,
1990; Ocasio, 1997)—does not explain the assessment gap between TMs and
14 Administrative Science Quarterly XX (2015)
software MMs that we describe below. The PUs’ attention was focused on
TMs’ requests and narrow market segments corresponding to the specific
products they were developing. Nokia had five PUs, each responsible for a spe-
cific market segment such as music phones. Within each unit, there were sev-
eral programs, such that Nokia had ‘‘40 to 50 product programs per year’’
(UMM#2), and program leaders were tasked with implementing their respec-
tive programs. Each PU focused on delivering phones as directed by TMs. The
key aspects that TMs emphasized for PUs were product segmentation, cost,
and schedule. The software units’ attention was focused on satisfying both
TMs’ and PUs’ requests. As our informants related:
The pressure was not personified in any one [person], but came from the whole
executive group. . . . For me personally and my team, the pressure was the message
that ‘‘we need these phones, we need those phones, we have to have releases in
such and such a time.’ (UMM#3, software)
Those [key technological decisions] were all taken to the executive group and they
approved the technical choices....ItwasliketheTMstoldusthat‘‘now we’re using
[specific software], end of discussion, this is what we’ll do.’ . . . Each morning we
had a new strategy and each morning there was a new angle. It was very much
dependent on what someone had said to [EVP in charge of the unit] on the previous
day—that was the new direction. (UMM#4, software)
Even in the early 2000s, the sales organization had already been changed such that
the link to the customer was rather unclear. . . . There was a layer between the cus-
tomer and the units developing the software. . . . This blocked transparency and also
motivation. (MM#7, software)
Organizational norms—an element of attention structures (Ocasio, 1997)—also
intensified MMs’ internal focus. Nokia’s TMs believed that MMs’ narrow inter-
nal focus would ensure effective implementation, and therefore they discour-
aged MMs’ external focus. For example, after Apple had introduced the
iPhone, the CEO said publicly, ‘I don’t think that what we have seen so far
(from Apple) is something that would any way necessitate us changing our
thinking’’ (IntoMobile, 2007). Likewise, when MMs asked critical questions,
the TMs urged them to keep their attention focused on implementing their
tasks: ‘‘. . . the answer typically was that ‘this is the direction we have chosen.
Don’t challenge me, but focus on implementation’’ (MM#2).
TMs’ and MMs’ different appraisals. As a result of the structural distribu-
tion of attention, different appraisals generated different emotional reactions
among TMs and MMs. Whereas the TMs’ appraisals of the person–
environment relationship focused on the external environment, MMs’ apprai-
sals mainly focused on factors inside Nokia. TMs recognized early on that
external conditions beyond Nokia’s control threatened its survival. As one of
them described:
[In 2007], Sergey Brin, one of the founders of Google, took something out of his
pocket that I’d learned to call a ‘‘smartphone.’’ He put it on the desk, then he pulled
out another, and a third. Then he turned to the audience and said, ‘‘In two or three
years’ time, these displays’’—that’s what he called them—’’won’t add any value. All
the value will be in the cloud, and these displays will just be a window on the cloud.’’
Vuori and Huy 15
I’m looking at him and I’m smiling, of course—there were 500 people watching. But
I’m thinking, ‘‘Oh shit!’’ because he was spelling out Google’s strategy: make the
smartphone less value-adding and move the intelligence to the cloud [which would
make Nokia irrelevant as its core business was making ‘‘these displays’’].
In contrast, MMs’ appraisals focused on threats perceived inside Nokia.
Though PUs also made sense of their respective narrow market segments,
their appraisals often focused on what TMs wanted from them. These apprai-
sals included a perception of potential harm to their well-being and uncertainty
over their ability to cope with it. For example, an MM from a PU (MM#3) told
us: ‘‘The people who told the truth [about the feasibility of schedules put] their
reputations on the line. They ran that risk.’ Similarly, the focus of software
MMs’ appraisals was on the threats and uncertainty relating to TMs’ requests.
UMM#5 noted that ‘‘if you did not comply [with TMs’ unreasonable request to
maintain a steady pace despite mounting technical challenges], you would be
labeled as a loser.’ In addition, Nokia’s tendency to change its organizational
structures frequently to ensure effective resource allocation made MMs in
both PUs and the software organizations perceive heightened threats to their
well-being inside the organization (a primary appraisal of fear) and low ability to
cope with such uncertainty (a secondary appraisal).
4
As a software MM told
us: ‘‘Reorganizations were difficult. People had to reapply for their jobs at
Nokia. ...Isentmyapplication.Thenyougotaninvitation to an interview . . .
Itwasquitetough....Atthatpoint,youstarted to wonder whether you were
working at Nokia or not. I felt like I wasn’t—I didn’t have a job or a position’
(MM#4, software). Conversely, we found very little evidence of MMs threat-
related appraisals related to the external environment beyond PUs’ considera-
tions of how well a specific product would sell in its segment. Our informants
described very few appraisals related to the external environment:
Our view of our competitors’ products’ usage was completely distorted in 2005–
2008. People didn’t know how good Android was, or the iPhone....So,acertain
small group knew, but it wasn’t known throughout the company how good the com-
petitors’ products are becoming. The group of people who really knew the pain was
way too small. (MM#3)
Structural–Behavioral Antecedents of Shared Fear
TMs’ history of aggression. A second factor that triggered fear among
Nokia’s MMs was its TMs’ history of aggressive behaviors toward MMs, which
was widely known inside the organization. When conducting our interviews,
we were struck by how often informants referred to the aggressive behavior of
Nokia’s chairman, who was also the CEO from 1992 to 2006. He was
described as ‘‘extremely temperamental’ (UMM#6) and shouting at people ‘at
the top of his lungs’ (top-level strategy consultant) ‘so hard that [the target’s]
ballsshrank...infrontof15otherVPsandSVPs’(MM#5,software). Thus ‘‘it
was very difficult to tell him things he didn’t want to hear’ (top-level HR consul-
tant). Our informants also described several other TMs as highly aggressive.
UMM#7 told how one of the EVPs once ‘‘pounded the table so hard that
4
Structural changes included both major ones (2004 and 2008) and more minor ones, such as
establishing new teams and units and rotating people in different roles, which happened yearly.
16 Administrative Science Quarterly XX (2015)
pieces of fruit went flying,’’ and MM#6 said the same EVP ‘‘had nothing but
poison running through his veins.’
MMs’ intuitive fear reactions. Research suggests that aggressiveness
often automatically triggers fear in social situations (e.g., Hareli and Rafaeli,
2008), especially when the aggressor has a higher status than the target of the
aggression (e.g., van Kleef, De Dreu, and Manstead, 2004); that emotions can
become associated with the places where they have been felt (such as meet-
ing rooms with TMs) (Damasio, 2003: chap. 2); and that hearing stories of oth-
ers’ previous aversive experiences with their leaders can intuitively activate
fear toward authority figures (Kish-Gephart et al., 2009). Most of our informants
reported having experienced or heard stories of the chairman or other Nokia
TMs behaving aggressively, and thus intuitive fear of TMs seems to have
gripped Nokia’s MMs:
[The chairman] had the habit that if someone said that ‘‘things aren’t going so well,’’
then after that the person would be doing very poorly. [The chairman] had a distinc-
tive style so that everyone had to tell him that things were going very well. (MM#7,
software)
[The chairman] was very cold and it seemed to me that many Nokians feared him.
(Reporter close to Nokia)
The atmosphere of fear was created through speech. The worst one was the presen-
tation that [the chairman] gave at Tampere [a city in Finland] about R&D expenses. . . .
He said that his only mistake had been to give us too much money, and that despite
that, our products still weren’t good enough, and we weren’t making them fast
enough. On top of that, he said that if things continued as they were, 15% of the peo-
ple in that seminar hall would be gone by the time he came back next year. . . . This
was my first encounter with [the chairman], and it left a permanent negative feeling.
(MM#4, software)
One TM noted that ‘[MMs who interacted with me] were nervous. . . . cer-
tainly you could sense . . . people’s natural fear of their lord and master.’’ But
he also admitted that he had not adequately realized the implications of MMs’
fear. The aggressive behaviors of the chairman and some EVPs activated a gen-
eric fear toward authority figures at Nokia, in particular all powerful TMs.
Shared Fears
In contrast to conventional accounts (e.g., Henderson and Clark, 1990; Ocasio,
1997; Eggers and Kaplan, 2013) that depict engineers and managers as rela-
tively unemotional, our informants described having felt strong emotions.
Nokia’s TMs felt high external fear triggered by sources outside the organiza-
tion such as competitors and shareholders, whereas MMs mainly experienced
little external fear and high internal fear triggered by sources inside the firm
such as superiors and peers. Our informants described experiencing fear of the
same targets repeatedly—when the targets were encountered, recalled, or
imagined—rather than feeling fear continuously. These fears had largely a
structural rather than personal origin, meaning they were related to the focal
groups’ formal position in the organizational structure. We describe these vari-
ous fears below and present additional data in Online Appendix B, table B1.
Vuori and Huy 17
TMs’ high external fear. Nokia’s TMs experienced high fear toward exter-
nal entities in regard to both the long- and short-term survival of the organiza-
tion. One described an intense fear reaction relating to long-term survival when
reacting to news of the iPhone: ‘When [internal market intelligence] news of
the iPhone arrived [in fall 2005], I asked which OS they were using. When I
found out it was iOS, it made my hair stand on end. iOS was a bombshell. . . .
It was shocking news. . . . iPhone was an extension of Mac—a Mac computer
with radio added. They’d been building the applications and the OS for 35 or 40
years’’ (TM). A second TM confirmed, ‘‘I do identify with your statement that
the top management felt external fear,’’ and a third one validated that TMs
experienced shared fear for the long-term well-being of their firm. In the short
term, TMs’ shared fear was related to the threat of cannibalization of Nokia’s
current OS, Symbian. If the firm did not keep posting strong quarterly sales,
the board of directors and financial markets might punish TMs. As a TM said,
‘‘Management is under huge pressure from investors. . . . If you don’t hit your
quarterly targets, you’re going to be a former [executive] very fast.’’ Though
TMs knew Nokia needed a better OS (MeeGo) to match iOS, they also knew
that developing it would take several years. But they were afraid that publicly
acknowledging Symbian’s inferiority to iOS would torpedo sales figures and
shatter internal morale in the short term:
I couldn’t say [publicly] that Symbian was no good and that we had to replace it with
MeeGo as soon as possible, because I was afraid of the [negative effect on] Symbian
sales. . . . Our organization had to have faith in it—you must believe in the gun you’re
holding, because there’s nothing else. It takes years to make a new OS. That’s why
we had to keep the faith with Symbian. (TM)
MMs’ high internal fear. Most of the MMs we interviewed or spoke with
informally admitted that they had feared their superiors. MM#3 told us that
‘‘they were rather scary moments when you had to go against [TMs],’’ and
software MM#8 told us, ‘I felt fear,’’ and indicated that the target of his fear
was TMs. Another informant said that shared fear among MMs toward TMs
‘‘describes the company very well’’ (MM#9, software). Some informants
expressed manifestations of internal fear, such as MM#4 (software) who,
when reporting fear of his superiors during vertical interactions, wailed loudly,
almost breaking into tears.
Nokia MMs also seemed aware of the fear felt by their colleagues; many
described not only their personal fears but also the tendency of many individu-
als to feel a similar fear (i.e., shared fear). MM#10 told us how ‘the atmo-
sphere of fear emerged’ in Nokia, and MM#11 described shared fear using the
metaphor of a highly feared, dictatorial national leader: ‘General sentiment and
communication ended up like North Korea.’ Software MM#7 noted that MMs
in general ‘‘were afraid of being caught.’’ Some informants also remembered
specific occasions when they could infer their colleagues’ fear from observable
cues or when they and their colleagues had spoken explicitly about fear. For
example, software MM#12 told us that his colleague ‘‘was typically open and
clear, but in certain meetings [with higher-level leaders], he became very quiet
and when he spoke his voice was shaking.’’ Software MM#7 said he had sug-
gested criticizing a TM’s decision, but his colleague ‘‘said that he didn’t have
18 Administrative Science Quarterly XX (2015)
the courage; he had a family and small children.’ MM#13 explained that an
R&D MM had perceived that his superior was afraid of TMs; the R&D MM
approached MM#13 privately after a meeting and complained: ‘‘[My boss in
R&D] has got no balls. He won’t push this [technological issue] upstairs.’’
Software MM#8 said he and his colleague once discussed their shared fear
toward TMs in a private gathering at a Helsinki bar, and MM#14 said she had
multiple conversations about ‘‘reduced trust and higher worries and fear with
my colleagues since the late nineties.’’
As well as fearing their superiors, many MMs described feeling afraid of
their colleagues in other units:
In those forums, there was no way you would accuse someone of not telling the
truth. [Culturally, it was not acceptable to] criticize someone else’s story in any
way. . . . No one wanted to fight their battles in front of [the CEO] and others,
because you knew that if you put someone [else] down, they’d put you down the
first chance they got. (UMM#8, software)
I should’ve been much, much more courageous. And I should’ve made a lot more
noise, should’ve criticized people more directly. . . . I could’ve made more of an
impact. And it would’ve been breaking the consensus atmosphere. . . . Nobody
wanted to rock the boat, especially [among] the middle management [level]. ...I
didn’t want to be labeled as a mean person who was constantly criticizing the hard
work of others. . . . I should have been braver about rattling people’s cages. (MM#1)
Beyond the key antecedents of fear described above, another reason for MMs
fear toward their colleagues was that many MMs perceived that other MMs
behaviors were self-centered and dysfunctional, using descriptions such as
‘‘focused only on self-interest and own career development’ or ‘‘deceptive.’’
The MMs we interviewed further described how they felt threatened by these
behaviors because they might harm the status and career progression of more-
authentic individuals. These reactions seem consistent with earlier research
showing that perceptions of political behaviors generate primary appraisals of
threat that elicit negative emotional states (e.g., Chang, Rosen, and Levy,
2009).
Prospective documentary data provided corroborating evidence of internal
fear at Nokia. In a book written in 2008, a former Nokian described Nokia’s
‘‘atmosphere of fear’’ (Palmu-Joronen, 2009: 12). A news article from 2007
was entitled ‘‘Fear is driving Nokians to unionize’ and described individuals’
fear of losing their jobs or being exploited (Taloussanomat, 2007a). A consul-
tant’s report in 2007 (mentioned by the EVP of HR in Taloussanomat, 2007b)
and a book written in 2009 by a former software and design director (Risku,
2010: 38, 61) both noted that a lack of courage was a particular problem at
Nokia. Two Nokians’ personal notes written between 2005 and 2007 directly
indicated that they had perceived others feeling fear; one indicated that he had
personally felt fear. Several blog posts and online discussions that included
Nokia MMs, IT professionals, and other observers close to Nokia also included
statements that described or indicated fear (Finnsanity, 2007; Uusisuomi,
2008; Jarmostoor, 2009). Two confidential reports written near the end of our
study period also indirectly indicated that MMs might feel fear toward TMs.
There were also several reports in the Finnish media in 2008 about a rumor
that Nokia was threatening to relocate its headquarters from Finland if a law
Vuori and Huy 19
that would allow companies to read their employees’ e-mails was not passed
(Wikipedia, 2015). The law became known as ‘Lex Nokia’’ (The Register,
2009). For many of our informants, this law symbolized the atmosphere of fear
at Nokia (see also Murobbs, 2006, for related online discussions by Finnish IT
professionals). Our informants also consistently noted that ‘no one was stupid
enough’’ (MM#15, software) to mention their true feelings in e-mails because
they suspected that the company was monitoring them.
MMs’ low external fear. Threat-rigidity theory would predict that when a
competitor poses a threat, the whole organization responds to it with emotional
arousal (cf. Staw, Sandelands, and Dutton, 1981). This did not happen at Nokia.
Although they deeply feared internal entities, most Nokia MMs experienced
only modest external fear. A senior strategy consultant explained that Nokia’s
MMs had little external fear that other firms were better because they believed
Nokia had superior capabilities relative to other organizations:
They [Nokia MMs] were overconfident in their own abilities. From a consultant’s
point of view, it showed in how they treated [the consultant], whatever function you
went into Nokia . . . to talk about, like, ‘‘Hey, have you thought about this issue like
this?’’ Nokia has always been one of the most arrogant companies ever towards my
colleagues, who were partners back then. It showed up in how everything was
always perfect, you know, Nokians [thought that they] were the best in class. And
they didn’t even want to hear how you could think about things differently. They just
went into their cocoon and patted themselves on the back.
Software UMM#8 noted, retrospectively: ‘[TMs] should have created healthy
dissatisfaction among people a lot faster.’’ As MMs’ attention was focused on
intra-organizational factors, the manner in which they assessed competitors’
products seemed biased by what was perceived as relevant for Nokia. MMs
downplayed the competitive gap by comparing Nokia’s future developments
against competitors past products: ‘‘[Competitors’ products] were always
compared to what we knew that we were doing next. There was not much to
learn from. . . . In comparison to our [future products] they were no better’’
(MM#16). Many MMs also made comparisons along dimensions that favored
Nokia’s products. For example, several MMs told us they had felt that ‘‘the
iPhone was a bad phone; it didn’t even have 3G [radio technology]’’ (UMM#3,
software). Some wondered whether ‘‘Apple paid magazine editors to sex up
the iPhone in their fawning articles’ because ‘‘they believed [that the iPhone
hype] was undeserved, irrational’’ (Cord, 2014: 86). These quotes indicate that
many Nokia MMs genuinely believed that their products were superior to
Apple’s.
MMs remained aware of the macro measures of their company’s perfor-
mance, but these measures—which again reflected the past rather than the
present—also suggested that there was ‘nothing to worry about’ (MM#3). As
quarterly evaluations by the stock market were important for Nokia, positive
market news strongly influenced MMs’ appraisals of external competition and
calmed their external fears. As UMM#7 noted, ‘[We sold] the classic smart-
phones with old-fashioned keyboards in such goddamn huge numbers that
some people questioned why we should even make touch-screen phones—
’we’re doing fine as it is.’’
20 Administrative Science Quarterly XX (2015)
TMs’ actions also contributed to MMs’ low external fear. TMs’ public rheto-
ric downplayed the threat of new entrants, including Apple and Google, to
maintain confidence among employees and customers that Nokia would con-
tinue to prosper. For example, the CEO downplayed Google’s Android strategy
publicly in November 2008: ‘We have had that [platform and applications] for
10 years. Definitely, we are ahead’ (Symbian-Freak, 2008). Likewise, an EVP
stated publicly that the iPhone is ‘‘an interesting product but it is lacking a few
essential features, such as 3G, which would enable fast data connections’
(Taloussanomat, 2007c).
5
MMs’ dependence on organizational status. In addition to the primary
antecedents of shared internal fear described above, many MMs’ dependence
on high status inside the organization amplified the influence of those primary
antecedents on MMs’ internal fear. Not surprisingly, many MMs attached high
importance to their social status inside Nokia and thus perceived negative reac-
tions from colleagues as personal threats. In addition, as many MMs lived in
Finland, home to few globally successful high-tech companies, they could see
no better job opportunity than being part of the industry world leader. Working
for Nokia conferred a prestigious external reputation or image. As our infor-
mants explained:
[Many people change jobs to advance in their careers, but] if you’ve been with Nokia,
especially as the level of SVP, VP, or even a director, where the hell are you going,
really? You’re going nowhere, so you circle around within the company. Unless
you’re willing to move to the US or Asia, you have no [better] job opportunities.
(Senior strategy consultant)
[Nokia’s status was so high that] if you had a Nokia business card, you could get a
meeting with any CEO. Any CEO. . . . Everyone had good jobs; no one wanted to
leave Nokia at that point. . . . Critique [of the company] was seen as negative; the
mindset was that if you criticize what’s being done, then you’re not genuinely com-
mitted to it. (MM#17)
For most MMs, Nokia membership was a central contributing factor to their
personal status. But for a few other MMs, being part of Nokia was deemed
less important. These rare MMs had already accumulated sufficient personal
achievements and material resources to have a sense of individual autonomy
and self-worth beyond what could be conferred by their organizational member-
ship. They also experienced lower internal fear than their peers. For example,
one software leader who had worked at Nokia for over 15 years, made
5
A fourth reason behind MMs’ low external fear might have been their beliefs about their own
superiority relative to the members of other organizations (cf. Hiller and Hambrick, 2005; Chatterjee
and Hambrick, 2007). Even though our interview data did not directly indicate that Nokia MMs had
these beliefs, two contextual factors could plausibly produce them (cf. Hayward and Hambrick,
1997). First, Nokia was for a long time among the most desired employers in Finland and therefore
recruited only the ‘‘best of the best,’ making it possible that Nokians saw themselves as superior
to others. Second, Nokia’s long success and market power may have made others flatter Nokians
in both business and private settings, reinforcing Nokians’ belief that they were better than others.
Note that we refer to MMs perceiving Nokia as a collective to be superior relative to other organiza-
tions but not to how individual MMs saw themselves relative to other MMs or TMs inside Nokia,
who shared the same positive reinforcement from the outside and were also among the best of
the best.
Vuori and Huy 21
groundbreaking innovations, and earned millions of euros was one of this select
group. He was described by several informants as very direct in confronting
TMs, and our interview with him confirmed this perception. But he was ‘‘com-
pletely sidelined’ (UMM#6), and his views—as well as the views of the few
other critical MMs—were dismissed during collective interaction processes.
Table B2 in Online Appendix B provides additional examples of individuals who
were part of this critical group and how their views were dismissed. The table
also includes more typical examples of MMs who had low to medium personal
achievement at Nokia and felt high internal fear.
TM–MM Decoupling Interactions
In combination, TMs’ external fear and MMs’ internal fear and low external fear
produced an interaction pattern between these groups that increased the
decoupling between TMs and MMs’ perceptions of how quickly Nokia could
develop new software and introduce smartphones to match the iPhone. During
these interactions, TMs, driven by external fear, exerted heavy pressure and
imposed challenging demands on MMs. This pressure came over and above
the high internal fear that MMs were already experiencing. Instead of pushing
back, MMs generally acquiesced to TMs’ demands and continued reporting
optimistic progress despite mounting implementation difficulties. Online
Appendix table B3 provides additional data on these interactions.
TMs’ increasing pressure on MMs for fast performance. Based on the
extant literature, we might expect top-level executives to appreciate the tech-
nological complexity involved in innovation processes (cf. Kaplan and Tripsas,
2008; Dougherty and Dunne, 2012) and therefore to be careful not to let their
own impatience compromise product quality (cf. Brown and Eisenhardt, 1997).
But at Nokia we found the opposite pattern, which arose because a previously
unacknowledged factor—shared external fear—influenced TMs’ behaviors.
Fearful that Nokia would lose its world dominance and post weak results in the
next quarter, TMs tended to act urgently, an action tendency associated with
fear (Lazarus, 1991), and put high pressure on product and software units to
develop new, innovative products rapidly.
First, TMs put direct pressure on MMs to perform faster. One TM noted, ‘It
was clear that we [TMs] feared the iPhone. So we told the middle managers
that they had to deliver touch-phones quickly.’’ Another TM said, ‘‘The pressure
we put on the [Symbian] software organization was insane, because the com-
mercial realities were also pressing. You must have something to sell. And of
course, when the pressure is on the R&D teams, really intensely, of course
they feel that they don’t have enough time.’ A high-level leader from the
MeeGo organization also reflected:
We spoke of a delay of at least six months, if not a year. We talked about it, it was
brought up . . . [but TMs] just said ‘let’s go, you just have to run faster, we have to
dothis.’...They said [some short-term commitments] mustn’t delay us, but we still
had to meet them. They probably didn’t realize that every time they asked us to do
something, someone actually had to do it.
22 Administrative Science Quarterly XX (2015)
Beyond the verbal exhortations, TMs also applied pressure for faster perfor-
mance by MMs through personnel selection. Although many MMs in the soft-
ware groups accepted TMs requests, a few did push back, as Online
Appendix table B2 shows. Predictably but not ideally, TMs favored MMs who
provided reassuring reports, thus temporarily alleviating TMs’ external fear, and
sanctioned MMs whose accounts validated TMs’ external fear:
Question: Why didn’t you explain to TMs that software development would be com-
promised if you had to develop as many phones in the given time period as the TMs
wanted?
UMM#9, software: Someone else always said yes [to unfeasible TM demands]. All I
got was the information—that’s how it was. And then my responsibilities were cut.
Because there was always some lunatic who promised they’d do all these ten fine
and wonderful things within the timeframe given by TMs . . . even though it wasn’t
true at all [that it would be possible]. . . . TMs trusted these people when they said
it’s going to work out. They had blind faith. The management team . . . knew a lot of
people—but they picked some young, fast-talking guy who said, ‘I have this little
trick, I’ll fix this thing.’’
Nokia TMs further sated their appetite for optimistic news by favoring ‘‘new
blood’’ who displayed a ‘‘can-do’ attitude. TMs confirmed that they had stressed
the need for new, typically younger staff, whom TMs perceived as having the
new skills and motivation needed to develop Nokia’s capability. A TM told us that
he ‘‘kept saying that we need new people,’’ and another TM added:
Some people can explain things more clearly, like ‘‘this is what we should be doing.’’
And others seem indecisive and talk nonsense. And then there’s the plus that some-
one had a clear plan that these things should be done. Of course you listen to those
people more, especially in a crisis, than to people who theorize issues endlessly. It
comes down to action in the end.
As longer organizational tenure was typically associated with lower internal
fear, however, Nokia TMs inadvertently paid less attention to MMs who had
lower internal fears and were thus more honest about organizational limitations
and proposed action plans that were less simplistic and less ambitious. TMs’
external fear might have led them to perceive MMs who reported problems as
pessimists or naysayers and thus to discount their warnings.
MMs’ over-optimistic reporting. Based on the extant literature, we might
expect that if TMs issued suboptimal directives because they lacked opera-
tional knowledge, MMs would help TMs improve their decisions through joint
sensemaking (e.g., Joseph and Ocasio, 2012). This did not happen at Nokia.
Instead, MMs’ internal fear made them promise to deliver what TMs were
requesting, even though they privately doubted the feasibility of TMs’ requests.
Though political perspectives might describe such behaviors as calculated cog-
nitive approaches for gaining power (e.g., Pfeffer, 1981), our analysis reveals
an alternative or complementary explanatory mechanism: MMs reacted emo-
tionally in particular situations as fear drove them to make over-optimistic pro-
mises and reports. These empty promises brought short-term emotional relief
to both MMs and TMs: MMs did not have to admit their limitations and risk
Vuori and Huy 23
sanctions in the short term, while TMs’ fear of losing to external competition
was allayed. In addition, low external fear convinced MMs that there was little
need to have a difficult debate with TMs about a non-critical issue. Thanks to
MMs’ submissive stance and over-optimistic reporting, TMs developed the per-
ception that their ambitious requests were feasible and the company’s long-
term development was viable.
MMs in product units explained to us how they behaved in situations that
triggered fear for their own well-being and access to resources inside Nokia:
[A central factor influencing whose proposal got resources was] the timeframe—how
long it took [to develop the product]. You can get the resources by promising some-
thing earlier, or promising a lot. It’s sales work really, you make an offer [TMs] can’t
refuse. It’s a challenge to get resources if there are competing teams, so [you over-
sell]. When you promised something, the most important thing was the momentary
fulfillment of needs; saying ‘‘yes’’ here and now. The benefit of doing that was so
much greater than saying ‘it can’t be done’’—[even if you were] right. (MM#3)
The message about each product area had to be kept positive so [the units] would
be allowed to continue [to operate]. If the message was that this hasn’t progressed
with these resources, [MMs] would be afraid that [their project] would be discontin-
ued. . . . One source of fear was that many good projects had been discontinued. It
was known [throughout the company] that Nokia had discontinued many good prod-
uct concepts that could have been successful—for one reason or another. (MM#16)
Some MMs also described how intense internal fear silenced them during
review meetings: ‘I remember one meeting where they had an insanely opti-
mistic deadline for one subproject that was critical for [the whole product]. . . . I
thought I should point it out, but the thought of challenging [TMs] made my
heart race, and then I just kept quiet’’ (MM#8).
MMs in the Symbian software organization felt pressure from both TMs and
product unit directors. Fear made them agree to TMs’ demands, even when
they did not believe they could be met. The typical interaction sequence was
as follows: MMs voiced their concerns, TMs continued putting pressure on
them, and MMs complied:
The software people tried to say that things weren’t going so well, but the pressure
to give the ‘‘right’ answer [e.g., the next software version would be ready by a given
date] to top management was high. Concerns were ignored during the conversation.
The interaction was fragmented. If you were too negative, it would be your head on
the block. If you said something couldn’t be done, then it’s about whether we should
replace you. (TM)
They [higher level MMs in charge of specific aspects of software] accepted interface
concepts as targets [as requested by TM] without going over whether it could be
done and how fast. They accepted the concept and set the target that it’d be in
stores within a certain time, so it was ‘go for it’’—and then the guys started thinking
about how they’re gonna build it. (MM#18, software)
They [TMs] thought that if they just put pressure on the product-development organi-
zation, they would execute it. So maybe the product-development area should have
been more assertive and said, ‘What you’re asking for is impossible.’’...Inproduct
development, people didn’t have the courage to say, ‘‘Listen, it’s like this. We can’t
give you anything more.’ In Nokia’s R&D, the culture was such that they wanted to
please the upper levels. They wanted to give them good news ...notarealitycheck.
24 Administrative Science Quarterly XX (2015)
There was a lot of top-down guidance. We’d be showing a schedule, then when we
talked to middle management, we’d ask, ‘‘Why did you promise that? Why did you
say yes?’’‘‘Well, because it will be a hassle if we say it’s late . . . we can’t say that.’’
We needed a truth commission so people could tell them things without the fear of
punishment, to say what was really going on. (UMM#7)
When problems started appearing, MMs did not inform TMs of potential delays
or missing product features. Fearing TMs immediate negative reactions, they
remained silent or filtered information:
[In our unit] we wouldn’t say anything about this [anticipated delay to TMs] because
we thought that [another unit] was going to slip first. So we’d let them slip first, and
then we could say, ‘‘Oh well, in that case we’ll take an extra six weeks.’ So there
was an element of brinkmanship and trying not to be . . . the one to slip first.
(UMM#10, software)
The information flowed downwards, but the information did not flow upwards. Top
management was directly lied to. They were misled. I remember examples when
you had a chart, and the supervisor told [you] to move the data points to the right [to
give a better impression]. Then you did that and your supervisor went to present it to
the higher-level executives. There were situations where everybody [on our level]
knew things were going wrong, but we were thinking, ‘‘Why tell TMs about this? It
won’t make things any better.’’ We discussed this kind of choice openly. And it was
possible to give them embellished reports because they did not understand the soft-
ware. (MM#19)
MM informants also reflected how their behavior had been driven by fear
rather than cold, affect-neutral self-interest calculation:
Question: Would you say that this [not telling about problems] was because of pure
cognitive self-interest calculation?
MM#12: No....Iknewthatitwouldhitmeharderlater.Iknew it would be better
to talk about it early. But I couldn’t make myself do it. . . . It was like going to the
dentist; you know that the situation is only going to get worse, but you still postpone
it. . . . I think it was fear.
Question: When people gave optimistic promises, do you think they did it calculat-
ingly, out of political self-interest, or was it an emotional reaction?
MM#13: In those situations, there were lots of people talking in a meeting and things
moved really quickly. You just went with your gut. It’s very complex. You can’t work
out how each and every person will react, and weigh up what would be best for you
in the long run. It’s more based on feeling.
TM–MM Assessment Gap
As a result of decoupling interactions between TMs and MMs, an assessment
gap between the two groups arose. TMs’ perceptions of what Nokia would be
capable of doing differed substantially from MMs’ perceptions, which in retro-
spect were closer to reality. Table B4 in the Online Appendix provides addi-
tional data demonstrating the assessment gap.
TMs’ over-optimistic capability perception. TMs relied heavily on reports
from various MM groups to stay updated on the speed of development of the
Vuori and Huy 25
OS software. Even though Nokia’s TMs had an accurate understanding of the
market needs, their interactions with MMs gave them an over-optimistic under-
standing of Nokia’s capabilities. As Nokia TMs noted:
I only realized [I was getting filtered reports] later on. There was a change in the busi-
ness environment: we got rid of the hardware competitors and had software and
internet competitors instead. But our organization’s area of expertise was [not in
these new areas]. People [in Nokia] learned to speak the [new technical] language
quickly; they became quasi-experts. They gave the impression that they understood
[this new area], but [I realized later] it was only skin deep. I was too ready to believe
that these people could change. (TM)
The products were always late, but they were never late in [reporting] conversations.
They came out one or two years late, but in conversations it was always that [our
smart] phone would be ready in one or two months. And because it was said that it
would be ready in one to two months, you never initiated bigger improvements that
would have required six months. That would have created a long delay based on the
understanding that prevailed then. You always imagined that the products would
come out soon. (Another TM)
Beyond noting that MMs’ communication was positively biased, Nokia’s TMs
also reflected that their own emotional states might have influenced how they
interpreted signals from MMs and contributed to the emergence of their over-
optimistic capability perception:
Making a big change takes such a long time that you don’t want to believe it. . . .
This is related to admitting things to yourself. So even if you understood on an intel-
lectual level, you still won’t accept it. . . . No matter what level you were at—
employee, supervisor, middle or top management—it’s extremely difficult to truly
accept [that you cannot fix Symbian rapidly], even though you might understand it
intellectually. (TM)
One reason we believed we could [develop software capabilities and smartphones
as described by some MMs] was that the alternative [scenario was] horrible. The
alternative, which Nokia [was forced] to adopt later, was to focus only on hardware
and buy the software outside. [We would become] a pure hardware manufacturer.
(Another TM)
Because TMs had an over-optimistic perception of capability development,
they continued pressuring MMs to work faster because of their external fear of
competition. A vicious cycle leading to declining product quality arose. Pressure
from TMs hurt long-term OS software development by forcing MMs to take
short cuts, and it amplified MMs’ internal fear, which made them submit over-
optimistic reports, maintaining TMs’ over-optimistic perception.
TMs’ low technological competence. Because organizations often hire
outside executives from unrelated industries (e.g., Forbes, 2009), and most cog-
nition research focuses on TMs’ perceptions of external market needs (e.g.,
Eggers and Kaplan, 2013), one might expect that deep technological knowledge
is not critical for top-level executives’ effectiveness. Our data suggest the oppo-
site: TMs’ low technological competence made them more dependent on MMs’
communication and thus amplified the effects of the fear-based communication
on TMs’ perception of organizational capability. When MMs showed them
26 Administrative Science Quarterly XX (2015)
technological demos or early prototypes, TMs were not able to directly assess
whatprogresswastrulybeingmadeandhadtorelyonwhatMMsreported:‘A
sufficiently competent guy can just look at the phone and how it works and see
what’s still to be done. Just by looking and testing it, [he knows] what’s going
on. But if you don’t have a technical background, you just can’t understand; you
can’t get it’’ (MM#18, software). TMs’ low technological competence also influ-
enced how they could assess technological limitations during goal setting, partic-
ularly in the absence of critical feedback from technological MMs:
If you consider Apple, the TMs are engineers. They tried to recruit [a senior Nokian] to
Apple.HecamebackfromhavingmetJobsandeveryoneandhesaid,‘Nokiais
business-case driven. We make everything into a business case and use figures to
prove what’s good, whereas Apple is engineer-driven. It was pure technology and the
top management was immersed in the technology.’ That’s often how it is in a product
company,youhavetounderstandhowtheproductisbuilt....Youhavetomakealot
of product decisions based on what’s possible and what’s not. Of course you must
have stretching goals, but just deciding to build a product [will not work]. (UMM#7)
Nokia’s TMs also admitted that, as one TM told us, ‘‘there was no real software
competence in TMT.’ The chairman likewise stated, ‘[the CEO] said himself
that he wasn’t a software expert. That was true, [and] I wasn’t a software
expert either’’ (Ollila and Saukkomaa, 2013: 458). Another TM said, ‘The real
lesson [from the Nokia case] is that you should not appoint someone who does
not understand the technology . . . in a deep way, but only [knows it] through
numbers, to be the CEO of a technology company.’’ Ironically, several infor-
mants also questioned the technological competence of this TM, whose back-
ground was not in technology. A third TM also admitted, ‘‘When the business
changes you have big masses of people, including people in top management,
whose competence is not quite at the core here. Then, this kind of faith is
formed that is partly based on the missing competence.’’ Thus it seems that
Nokia’s TMs had low technological competence, which amplified the effects of
the various fears on their decoupling interaction with MMs and contributed to
the development of their over-optimistic capability perception.
6
In addition, it is
possible that TMs’ low technological competence had a direct effect on Nokia’s
innovativeness. Had the TMs had higher technological competence, they would
have made more informed and thus more optimal choices regarding
technology.
6
As a counterfactual (a situation in which TMs have high technological competence), consider
Steve Jobs’ influence at Apple, which seemed to have a somewhat similar (if less extreme) fear
profile as Nokia. Jobs felt high external fear regarding Apple’s long-term success; he was very
upset when Google announced Android and worried that it would destroy the iPhone (Isaacson,
2011: 511). He was also notorious for instilling fear among MMs (Apple Insider, 2013). But Jobs
seemed reasonably in touch with the technology used in the iPhone and was closely involved in its
development. For example, he spent ‘‘part of every day for six months helping to refine the
[iPhone] display’ (Isaacson, 2011: 469) and was extremely demanding about technological quality.
Consequently, Jobs was likely able to make stretching yet realistic demands on product developers,
and MMs, however scared they were of him, could not mislead him technologically. This pattern
changed when Jobs stepped down: Jobs’ successor, Tim Cook, had lower technological compe-
tence and might thus have been less able to critically assess his subordinates’ communication. This
might help explain why Apple launched its maps application in 2012 with so many errors that the
company ultimately advised its customers to use Google’s or Nokia’s rival applications while Apple
fixed its own (Apple, 2012).
Vuori and Huy 27
Innovation Underperformance
Innovation underperformance at Nokia followed TMs’ over-optimistic capability
perception. As the TMs had an inaccurate understanding of their organization’s
capabilities, their decisions regarding resource allocation to various innovation
processes were decoupled from organizational reality. Temporal myopia
resulted, as illustrated by the additional data in table B5 of the Online Appendix.
Nokia allocated disproportionate attention and resources to the development of
new phone devices for short-term market demands at the expense of develop-
ing the OS software; TMs took little corrective action. One might infer that this
‘‘rigidity’’ occurred because threat perceptions caused TMs to think in narrow
ways (cf. Staw, Sandelands, and Dutton, 1981), but this was not the case.
Rather, rigidity in the form of temporal myopia came about because the infor-
mation that the TMs received from MMs indicated that their current actions
would lead to success. Nokia’s TMs thus made boundedly rational decisions
based on the inaccurate understanding that they had sufficient time, resources,
and capabilities. Gradually, product quality declined.
Extant research suggests that modularity in software (e.g., MacCormack,
Rusnak, and Baldwin, 2006) enables higher-quality development in the long
term, although it does take time to develop such modularity. At Nokia, pressure
to introduce new phone models in the short term deprived the Symbian unit of
the time they needed to make the OS more modular:
[The Symbian OS software] had a very antiquated architecture in many ways, which
[software developers] could never modernize and they weren’t given the time to mod-
ernize. They tried to make very different kinds of products based on that architecture,
which meant that they had to bolt on all sorts of things to make an individual product
happen. And a terrible technical complexity emerged through that process. All sorts of
product-specific things piled up in there that they could no longer maintain. . . . [For
example,] the way the user interface was done, it was really old. A totally antique sys-
tem. So doing anything with [this old system resulted in] very slow [performance].
Then instead of saying early on that we have to get rid of this [old system], it’s not
worth fixing it, they had just been patching it up. It might help in getting the next prod-
uct out, but it doesn’t solve the [core] problem. (UMM#7)
Another element of the problem was that Nokia’s structural arrangements did
not prevent product units’ immediate needs from overwhelming the long-term
needs of the Symbian software unit. In ambidextrous structures, the unit focus-
ing on long-term development should not have short-term pressures
(Christensen and Bower, 1996; Gilbert, 2005; Taylor and Helfat, 2009). But in
Nokia’s case, because of the emotional dynamics, TMs did not perceive a real
and urgent need to change the structural relationship between the software
unit and PUs. Instead:
Every PU sold the idea to top management that changes to the [OS] software were
needed, or we won’t be able to deliver in time. Or we won’t be able to make a prod-
uct that sells. And this led to a great decline in R&D productivity. The effort was frag-
mented into multiple programs and the OS software was not integrated. (UMM#6)
In Nokia’s case, where the product was integrated hardware and software, and ser-
vices are built on top of the hardware, integration must be done. The offering is inte-
grated so the organization must also be integrated. We reflected on this a lot. The
28 Administrative Science Quarterly XX (2015)
good things that come from integration are that you get complementarities and the
ability to keep things connected, but the negative side is that speed and flexibility suf-
fer....Structure must follow strategy. The strategy was to offer integrated solutions
so the organization also needed to be integrated. It required a lot of coordination in
the development. Retrospectively one can see that it was slow. In the end it was
concluded that it was not possible [to develop software and hardware in the way
Nokia was developing them]. (TM)
Nokia could have used several approaches to correct its temporal myopia, such
as taking more time to improve Symbian or improving its capabilities to speed
up long-term development. (Additional examples are described in Online
Appendix table B5.) But TMs did not take any significant corrective action until
they realized that Symbian had become unfixable.
The quality of Nokia’s high-end phones gradually declined—although the firm
continued to produce dozens of successful medium- to low-end phones, which
required less-advanced software. For example, in 2007, Nokia launched the
N95 smartphone, which had full music features, GPS navigation, a large screen
(not a touch screen), and full Internet browsing capability. Even though some
‘‘compromises [in software were] accepted to get the product ready on time’’
(Laukkanen, 2012: 71), it was still seen as a huge leap forward and went on to
be the most profitable Nokia phone ever. But more serious quality problems
soon emerged. In 2008, Nokia launched its first touch-screen phone, the 5800,
at a lower price point than the iPhone. It was a commercial success, but ‘‘it
was about one and a half years late’’ (TM) because of difficulties in software
development. In 2009, Nokia launched the N97 to overthrow the iPhone and,
according to an EVP, ‘‘change how people think about mobile devices’’
(Symbian-Freak, 2009), but the phone was a ‘total fiasco in terms of the quality
of the product. Not just [in terms of] user experience, but anyone who used
the product could see that it simply did not work’ (TM). Another TM admitted
that ‘‘N97 was a cold shower. It was so sudden and unexpected, that some-
thing was very wrong—that came as a surprise.’
In 2010, Nokia launched the N8, another purported ‘‘iPhone killer’’ with a touch
screen. Its original intended launch date had been a year earlier. During the devel-
opment process, the phone ‘‘was extensively tested [to ensure high quality], and
it got delayed [repeatedly], and when it was tested again, the conclusion was that
it still wasn’t good enough. These delays proved fatal [i.e., prompted the com-
pany to search for a new CEO]’’ (TM). In addition, the N8 failed to match the
competition: ‘‘Usability is where the Nokia N8 . . . falls short the most . . . if you
are coming from webOS, iOS, or Android, things are likely to feel kludgy to you’’
(Mobile Burn, 2010). Nokia had been also developing its Linux-based OS, MeeGo,
in parallel to Symbian, but it suffered from major development delays. A new
CEO hired in September 2010 decided that Nokia would be better off buying soft-
ware from external firms and thus struck a strategic alliance with Microsoft in
February 2011. Microsoft ultimately acquired Nokia’s phone businesses in 2013.
DISCUSSION
As summarized in figure 2, the structural distribution of attention at Nokia and
TMs’ past aggressive behaviors generated external and internal fear among
TMs and MMs, respectively, and these different types of fear caused
Vuori and Huy 29
Figure 2. Shared fears and innovation underperformance.
Structural
distribution
of attention
TMs’ history
of aggression
TMs’ high
external fear
MMs’ high
internal fear
and low
external fear
TMs’ over-
optimistic
capability
perception
Temporal
myopia
SHARED FEARS
TM–MM DECOUPLING
INTERACTIONS
TM–MM
ASSESSMENT GAP
STRUCTURAL,
ATTENTION-RELATED
ANTECEDENTS OF
SHARED FEARS
STRUCTURAL–BEHAVIORAL
ANTECEDENTS OF
SHARED FEARS
INNOVATION
UNDERPERFORMANCE
MMs’ dependence
on organizational
status
TMs’ and MMs’
different appraisals
TMs’ increasing pressure on
MMs for fast performance
& MMs’ over-optimistic
reporting
MMs’ intuitive fear
reactions
TMs’ low
technological
competence
30 Administrative Science Quarterly XX (2015)
decoupling interactions between groups of TMs and MMs. These interactions
produced an assessment gap of organizational capability between TMs and
MMs that contributed to Nokia’s innovation underperformance. In particular,
internal and external fear led to TM–MM interaction cycles that led TMs to
believe they could allocate resources to short-term developments without com-
promising long-term developments. The outcome was that long-term software
development suffered, which hindered Nokia’s ability to implement its strategy
successfully. We also identified how MMs’ dependence on organizational sta-
tus amplified their internal fear; how a feedback loop from different groups’ per-
ceptions had the same effect; and how TMs’ low technological competence
increased their dependence on MMs’ reporting, amplifying the effects of fear-
based interactions.
On a broader theoretical level, the model presented in figure 2 suggests
how organizational structures can influence micro-level factors such as emo-
tions, TM–MM interaction, and cognition, which influence choices made during
the innovation process and thus contribute to innovation outcomes at the orga-
nizational level. The model also includes the possible direct effect of top man-
agers’ low technological competence with a dashed line. By identifying these
macro–micro–macro linkages in the model, we provide textured multilevel the-
orizing on innovation, strategy, and organization.
Structurally Based Fear
Our findings suggest that the structural distribution of attention in an organiza-
tion (Ocasio, 1997) could lead TMs and MMs to experience external and inter-
nal fear, respectively, and that these emotions can have a substantial influence
on behaviors and communication patterns in the innovation process. Our find-
ings suggest that external fear might result when a group’s organizational role
is to focus mainly on threats in the external environment. Internal fear might
result when a group’s organizational role is to focus mainly on implementing
and responding to other members’ directives and requests. External and inter-
nal fears are primarily related to the focal group’s roles in the organizational
structure and could be more generally labeled as structurally based fear.
In contrast to prior research, which has investigated attention structures and
their effects mainly from an unemotional, information-processing perspective
(e.g., Ocasio, 2011; Gavetti et al., 2012), we have shown how attention struc-
tures could generate shared emotions among groups. Emotional reactions
occur because when groups attend to information, it is not merely processed
analytically to determine a satisfactory course of action but also triggers apprai-
sals of the person–environment relationship that generate emotional reactions
(Lazarus, 1991). During such appraisals, groups do not just process events
through a retrospective lens but may also use the information to anticipate a
future outcome (cf. Gavetti and Levinthal, 2000). Previous research has shown
how such anticipation can happen through analogical reasoning (Gavetti,
Levinthal, and Rivkin, 2005), discussions with venture capitalists (Maula, Keil,
and Zahra, 2013), or collective sensemaking among members of the groups
(Balogun and Johnson, 2004), but it has not investigated how the processes
also trigger emotional reactions and how they influence subsequent behaviors.
Although prior research has focused on the evolution of cognition during
innovation processes, our study reveals that the same processes can also elicit
Vuori and Huy 31
emotional reactions that cause the processes’ trajectory to diverge from the
path predicted by purely cognitive accounts. A purely cognitive account might
predict that distributed attention and frequent changes in MMs’ positions
would enable the organization to collect more comprehensive information (with
each MM focusing on his or her specialized segment) and integrate such infor-
mation through communication channels (cf. Joseph and Ocasio, 2012) to avoid
various types of myopia (cf. Levinthal and March, 1993). Instead, our findings
showed how the emotions that emerged as the outcomes of these practices
hindered the integration of attention, leading to temporal myopia. This suggests
that future research could shed new light on organizational structures by exam-
ining the shared emotions they create for different groups. Even if the organiza-
tion in aggregate had accurate information about the environment, and TMs
made choices accordingly, diverse groups’ shared emotions could still influence
the quality of information exchange.
Our study focused on what occurred in a traditional hierarchy, but we could
think of other structural arrangements that might generate emotional reactions
that harm organizational action and performance. For example, time-pacing and
semi-structures (Brown and Eisenhardt, 1997) might, over time, inadvertently
amplify MMs’ internal fear and reduce their external fear, thereby causing inno-
vation underperformance. Time-pacing ‘creates a relentless sense of urgency’
(Brown and Eisenhardt, 1997: 24–25) that, if left unmanaged at the emotional
level, might translate into internally focused fear and short-termism. Likewise,
the inherent ambiguity of semi-structures may exacerbate MMs’ focus on
intra-organizational matters, as they strive to make sense of how and with
whom they should exchange information. This excessive internal focus risks
reducing the attention devoted to processing external threats, generating high
internal fear and low external fear, which influence subsequent behavior.
Our data also suggest that employees such as MMs may seek to anticipate
how future structural changes will affect their personal status and privileges,
and this can trigger further emotional reactions that are not necessarily related
to general business conditions. This suggests that some earlier ideas on the
benefits of frequent structural change (e.g., Ethiraj and Levinthal, 2004; Teece,
2007) might need to be nuanced. For example, although simulation-based stud-
ies have suggested that frequent structural changes help avoid myopia in orga-
nizational attention (e.g., Siggelkow and Levinthal, 2005), our study reveals that
these changes might also elicit high internal fear among MMs and cause harm-
ful deceptive behavior.
Organizational structures could, moreover, influence MMs’ fears through for-
mal hierarchy. Existing research has shown that people are particularly sensi-
tive to the behavior of individuals of higher power and status (Kish-Gephart
et al., 2009). Some of Nokia’s TMs expressed aggression toward MMs, and
MMs shared stories about such instances widely. The emotional story sharing
increased MMs’ internal fear toward all TMs as a powerful group. This sug-
gests that one reason less-hierarchical structures might foster innovation
(Burns and Stalker, 1961) is that more organic, egalitarian structures dampen
the effects of hierarchy-based fear and the potentially harmful effects of more-
powerful groups. When the power difference between TMs and MMs is per-
ceived as modest, TMs’ aggression is less likely to amplify MMs’ internal fear
and thus less likely to cause extreme protective behavior.
32 Administrative Science Quarterly XX (2015)
MMs’ dependence on organizational status could also affect how strongly
the aforementioned structural factors influence MMs’ internal fear. In Nokia’s
case, MMs’ dependence on the status of their company was particularly high
because Nokia was one of the few globally successful high-tech companies in
small-population Finland. When MMs’ status is highly dependent on their orga-
nization, they are likely to be more sensitive to threats inside the organization,
because those threats exert a more immediate effect on MMs perceived
person–environment relationship; this increased sensitivity evokes emotional
reactions (Lazarus, 1991). Central to such dependence is the relative weight
given to organization-independent and organization-dependent factors in one’s
status: greater personal achievements should increase a person’s independent
status, whereas the high status of the organization should increase the effect
of organizational membership on personal status. For example, a junior soft-
ware engineer joining Google would enhance his or her personal status
because of the company’s high profile, whereas a Nobel Prize winner making
the same move would receive less of a reputational boost, as he or she already
had significant personal achievements. Hence, once hired, the engineer would
probably experience higher internal fear than the Nobel Prize winner. This sug-
gests the intriguing hypothesis that organizations might consider retaining their
longer-tenured, successful MMs not just because of their rich tacit knowledge
(e.g., Droege and Hoobler, 2003) but also because they are less likely to experi-
ence high internal fear and are therefore more likely to communicate honestly
about organizational reality to people in higher formal positions—in particular, to
be the bearers of bad news.
Though our data suggest how the structural distribution of attention and hier-
archy can lead to different types of shared fear among diverse groups, organi-
zational structures could also induce fear in other ways. For example,
organizational structures that encourage groups to compete with one another
likely generate different shared emotions than structures that promote colla-
boration (cf. Ouchi, 1977). Likewise, functional and divisional structures expose
groups to diverse groups of actors and interests inside and outside the organi-
zation and could thus generate different emotional reactions. Future research
could investigate more comprehensively how various structural factors influ-
ence diverse groups’ emotions—both the discrete emotions that are experi-
enced and the diverse targets of those emotions—and how these shared
emotions subsequently influence organizational outcomes.
Integration of Attention
Diverse types of structurally based fear beg a deeper discussion of how such
fears influence the organizational processes that are presumed to make any
particular structural arrangement effective. The attention-based view of the firm
emphasizes the centrality of integration of attention. This view assumes that
different groups first distribute organizational attention (Ocasio, 1997) and then
bring their perspectives together to facilitate high-quality decisions. Joseph and
Ocasio (2012: 644) noted that this can happen through ‘open and frank dialo-
gue’’ when appropriate channels are present (see also Henderson and Clark,
1990). But our study suggests that TMs’ external and MMs’ internal fear can
foster interaction patterns that amplify rather than integrate differences in per-
spective, even with regular formal and informal meetings to integrate attention.
Vuori and Huy 33
The communication channels within Nokia’s formal structure were supposed
to highlight key information; however, both MMs’ and TMs shared emotions
reduced the accuracy of the information exchanged and biased subsequent
strategic decision making. TMs’ external fear led them to exert pressure on
MMs without fully revealing the severity of the external threats and to interpret
MMs’ messages in biased ways. MMs’ internal fear made them mislead TMs
during their interactions.
The attention-based view of the firm has emphasized that organization mem-
bers’ attention is influenced by various contextual and situational factors
(Ocasio, 1997, 2011). This reveals the sequential and cyclical nature of atten-
tion, in that what people attend to first likely shapes what they attend to later
(cf. Weick, 1979). Our study reveals how various types of fear could shape this
cyclical process in TM–MM interactions and extends our understanding of
situational attention in organizations. Shared emotions felt in one group could
influence the attention of another group: TMs’ external fear can make them
exert heavier pressure on MMs, which increases MMs’ attention to intra-
organizational threats; and MMs high internal fear makes them communicate
potential problems less openly to TMs, which reduces the attention TMs pay
to those issues. Likewise, MMs’ internal fear can reduce their disposition to
publicly criticize their peers, which can reduce TMs’ attention to potential
weaknesses in open strategic debates. At Nokia, the outcome was that TMs’
understanding of what the organization was capable of increasingly diverged
from MMs’ understanding, and this assessment gap contributed to Nokia’s
temporal myopia and innovation underperformance. Hence a deeper under-
standing of how the integration of distributed attention occurs in organizations
would require scholars to consider the shared emotions that diverse groups
experience as a result of their distributed attention and how those emotions
influence their motivation and ability to share their views.
Our study tentatively suggests that MMs fearing external competition
directly—as opposed to fearing TMs as a proxy of the external environment—
might lead to more functional communication patterns (see also Grove, 1996).
Though TMs might still exert excessive pressure on MMs, MMs’ external fear
could function as a corrective, allowing them to pay sufficient attention to
threats in the external environment. With moderate internal fear and moder-
ately high external fear, MMs might resist mounting pressure from TMs more
assertively and be honest about what they could realistically deliver in a given
timeframe, forcing TMs to revise their expectations and take timely adjusting
actions. Future research could investigate this hypothesis, explore whether it is
possible to elicit salutary, adaptive external fear among MMs, and investigate
the associated side effects on other stakeholder groups such as shareholders
and customers.
One could also speculate that TMs’ high ability to regulate (e.g., Gross, 1998)
their external fear might foster higher-quality interactions with MMs. In the
Nokia case, TMs’ high external fear caused them to put too much pressure on
MMs and dismiss MMs’ warnings. We wonder what might have happened if
Nokia’s TMs had tempered their high external fear somewhat, freeing their cog-
nitive resources to consider the wider consequences of their actions. Such regu-
lation of fear might have enabled them to stay attuned to the external threat
while forming a better understanding of Nokia’s internal context and to flexibly
co-develop a more appropriate and timely corrective action plan with MMs.
34 Administrative Science Quarterly XX (2015)
Our research on the effect of structurally based fear also contributes to
research on managerial cognition by identifying social–emotional processes that