Benner-2003-Exploitation, Exploration, and Process Management-The Productivity Dilemma Revisited. The Academy of Management Review

Exploitation, Exploration, and Process Management: The Productivity Dilemma Revisited
Author(s): Mary J. Benner and Michael L. Tushman

Reviewed work(s):
Source: The Academy of Management Review, Vol. 28, No. 2 (Apr., 2003), pp. 238-256
Published by: Academy of Management
Stable URL: http://www.jstor.org/stable/30040711 .
Accessed: 06/01/2013 05:52
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
Academy of Management is collaborating with JSTOR to digitize, preserve and extend access to The Academy
of Management Review.
http://www.jstor.org
This content downloaded on Sun, 6 Jan 2013 05:52:15 AM

All use subject to JSTOR Terms and Conditions
A
Academy of Management Review
2003, Vol. 28, No. 2, 238-256.
EXPLORATION,AND
EXPLOITATION,
PROCESS MANAGEMENT:THEPRODUCTIVITY
DILEMMAREVISITED
MARY J. BENNER
University of Pennsylvania
MICHAEL L. TUSHMAN
Harvard Business School
We develop a contingency view of process management's influence on both techno-

logical innovation and organizational adaptation. We argue that while process man-
agement activities are beneficial for organizations in stable contexts, they are fun-
damentally inconsistent with all but incremental innovation and change. But dynamic
capabilities are rooted in both exploitative and exploratory activities. We argue that
process management activities must be buffered from exploratory activities and that
ambidextrous organizational forms provide the complex contexts for these inconsis-
tent activities to coexist.
Twenty-five years ago, Abernathy (1978) sug- ance between exploitation and exploration, or
gested that a firm's focus on productivity gains between incremental and radical organization-
inhibited its flexibility and ability to innovate. al change, has been a consistent theme across
Abernathy observed that, in the automobile in- several approaches to research in organization-
dustry, a firm's economic decline was directly al adaptation (e.g., Brown & Eisenhardt, 1998;
related to its efficiency and productivity efforts. Burgelman, 1994; Gavetti & Levinthal, 2000;
He suggested that a firm's ability to compete Levinthal & March, 1993; March, 1991; Tushman
over time was rooted not only in its ability to & Romanelli, 1985). Yet this need for dual organ-
increase efficiency but also in its ability to be izational capabilities arises in the context of a
efficient and innovative simultaneously (Aber- wave of managerial activity and institutional
nathy, 1978: 173; Hayes & Abernathy, 1980). Strat-
pressures focusing on process management and
egy and organization theorists have similarly control (e.g., Adler, 1993; Cole, 1998; Hackman &
observed that dynamic capabilities are an-
Wageman, 1995; Hammer & Stanton, 1999;
chored in a firm's ability to both exploit and
Winter, 1994).
explore (Ghemawat & Costa, 1993; March, 1991;
Process management, based on a view of an
Weick, 1969). A firm's ability to compete over
time may lie in its ability both to integrate and organization as a system of interlinked pro-
build upon its current competencies while si- cesses, involves concerted efforts to map, im-
multaneously developing fundamentally new prove, and adhere to organizational processes.
capabilities (Teece, Pisano, & Shuen, 1997). Initially based on the seminal work of Ishikawa
Twenty-five years after Abernathy's observa- (1985), Deming (1986), and Juran (1989), process
tions, the pressures for organizations to meet management practices became popular as a
multiple, often inconsistent, contextual de- central element of total quality management
mands have escalated (e.g., Christensen, 1998; (TQM) programs in the 1980s (Hackman & Wage-
Tushman & O'Reilly, 1997). The notion of bal- man, 1995). Since then, these practices have con-
tinued to spread as a core element of a continu-
ing progression of quality-related initiatives,
We thank Diane Burton,Don Hambrick,Andy Henderson, including the Malcolm Baldrige National Qual-
Monica Higgins, Lee Fleming, David Garvin, Giovanni ity Award, the International Organization for
Gavetti, Peter Kolesar, Nitin Nohria, Nelson Repenning, San-
dra Sucher, Ruth Wageman, former associate editor Dev Standardization's Series 9000 program (ISO
Jennings, and three anonymous reviewers for comments on 9000), business process reengineering, and,
earlier drafts of this article. more recently, Six Sigma programs. By 1992
238

2003 Benner and Tushman 239
every Fortune 100 firm had adopted TQM prac- these studies have been equivocal. Research
tices (Nohria, 1996). aimed at resolving the debate suggests that dif-
While some suggest that interest in TQM had ferential outcomes from TQM adoption arise be-
waned by the 1990s (Powell, 1995), thousands of cause organizations implement different prac-
organizations were subsequently certified in the tices under the TQM umbrella (Westphal et al.,
ISO 9000 process management program (Quality 1997; Zbaracki, 1998). Similarly, Sitkin, Sutcliffe,
Digest, 1999), and many companies, including and Schroeder (1994) argue for an alternative set
highly visible ones like GE, Honeywell, 3M, of TQM practices when task environments are
Amazon.com, Toshiba, and Ford have recently highly uncertain.
embraced Six Sigma (e.g., Feyder, 2001; Gabor, Although these studies provide insight into
2001; Honeywell, 1998). Process management possible contingencies, their focus is specifi-
practices have been institutionally mandated cally on TQM programs. Similarly, much of the
as government agencies or powerful buying or- organizational literature on the topic of process
ganizations require adoption by their suppliers management has been focused on TQM (e.g.,
(e.g., Harrington & Mathers, 1997; Westphal, Easton & Jarrell, 1998; Hackman & Wageman,
Gulati, & Shortell, 1997). 1995; see also the July 1994 Total Quality special
Process management's contribution to im- issue of AMR and Cole & Scott, 2000). Research-
proving manufacturing efficiency has led to its ers have made no attempt to build theory that
migration beyond operations to other parts of links this TQM literature to broader concepts of
organizations-for instance, to adjacent pro- process management or with more recent re-
cesses for selecting and developing technologi- search on dynamic capabilities (e.g., Eisenhardt
cal innovations (Brown & Duguid, 2000; Sitkin & & Martin, 2000; Teece et al., 1997).
Stickel, 1996). As the variation-decreasing and We extend the process management literature
efficiency-oriented focus of process manage- by developing a model and testable proposi-
ment spreads to centers of innovation, or varia- tions about how process management activities
tion creation activity in organizations, it increas- affect both technological innovation and organ-
ingly affects an organization's dynamic izational adaptation. We explore how both tech-
capabilities. Yet there has been a lack of re- nological and organizational contexts moderate
search about how these institutionally man- the relations between process-focused activities
dated and pervasive practices affect technolog- and organizational adaptation (e.g., Sitkin et al.,
ical innovation or adaptation. 1994), arguing that process management tech-
Much existing literature is prescriptive and niques stabilize and rationalize organizational
aimed at educating managers on implementing routines while establishing a focus on easily
process management practices. Process man- available efficiency and customer satisfaction
agement's proponents have promoted process measures. Although increased efficiency results
improvement practices as universally beneficial from these dynamics in the short run, they also
for organizations, spurring continuous innova- trigger internal biases for certainty and predict-
tion that results in efficiency improvements, cost able results and ensure process management's
reductions, improved customer satisfaction, progression to more activities throughout the
and, ultimately, higher profits (Hammer & Stan- firm. This diffusion of process management
ton, 1999; Harry & Schroeder, 2000; ISO, 1999). techniques favors exploitative innovation at the
Reflecting these assumptions, empirical re- expense of exploratory innovation. We argue
search on process management's effects has that while exploitation and inertia may be func-
been limited to assessing the financial perfor- tional for organizations within a given techno-
mance implications from process management
adoption (e.g., Ittner & Larcker, 1997; Powell,
1995; Samson & Terziovski, 1999).1 The results of most research on TQMwhether and to what extent the or-
ganizations involved undertook process-focused practices,
interpreting or comparing existing research is hampered by
1 In other empirical research scholars explore implica- heterogeneity in the bundle of TQMpractices adopted. The
tions of more broadly defined TQM programs. While process three studies cited break out process-focused activities from
improvement practices may be included in an organization's other practices within TQMand, thus, are particularlyrelevant
adoption of TQM, this is not necessarily the case (Westphal to our topic. We also review the larger body of existing re-
et al., 1997,Zbaracki, 1998).Further,since it is not clear from search on process management and TQMlater in this paper.

240 Academy of Management Review April
logical trajectory or for existing customers, these tion of departments with separate functions and
variance-reducing dynamics stunt exploratory outputs to a view of them as systems of inter-
innovation and responsiveness to new customer linked processes that cross functions and link
segments (Henderson et al., 1998; Sterman, Re- organizational activities (Dean & Bowen, 1994).
penning, & Kofman, 1997). Finally, we explore Processes are collections of activities that, taken
how ambidextrous organizational forms provide together, produce outputs for customers (Garvin,
buffered contexts such that exploitation and ex- 1998; Ittner & Larcker, 1997). Customers include
ploration can coexist (Bradach, 1997; Tushman & not only external consumers of the organiza-
O'Reilly, 1997). tion's products or services but also a series of
We proceed in three sections. In the first we internal recipients at linkage points between
define process management techniques and ex- processes, as outputs from upstream processes
plore empirical evidence of their impact on organ- become the inputs for subsequent processes. Al-
izational outcomes. In the second we explore the though programs and awards like TQM, ISO
effects of process management techniques on ex- 9000, Six Sigma, the Malcolm Baldrige Award,
ploratory as well as exploitative innovation. We and business process reengineering differ in
also consider the structural features and moderat- scope and approach, they share a core focus on
ing effects of ambidextrous organizational de- measuring, improving, and rationalizing organ-
signs that isolate process control activities from izational processes (cf. Hammer & Champy,
exploratory activities. Finally, in the third section 1993; Harrington & Mathers, 1997; Harry &
we explore how process management practices, Schroeder, 2000; Ittner & Larcker, 1997; Powell,
through their influence on both technological in- 1995; Repenning, 1999; Winter, 1994).
novation and organizational inertia, affect adap- Process management entails three main prac-
tation. We also explore the organizational and tices: mapping processes, improving processes,
environmental contingencies that moderate pro- and adhering to systems of improved processes.
cess management's effects. Once underlying processes have been recorded
This paper contributes to research in several through process mapping, process improvement
ways. We extend the process management liter- involves using measures of process effective-
ature with contingent theory and testable prop- ness and statistical methods to continually
ositions about process management's effects on eliminate variation in processes and outputs
organizations. We also contribute to organiza- (Hackman & Wageman, 1995; Harry & Schroeder,
tion and strategy research considering how sta- 2000). Process improvement involves not only
ble processes influence outcomes like techno- rationalizing individual work processes but also
logical innovation, organizational adaptation, streamlining the handoffs between processes
and failure (Christensen & Bower, 1996; Hannan (Garvin, 1995; Harry & Schroeder, 2000). As or-
& Freeman, 1984; Henderson & Clark, 1990; ganizational participants are trained in effec-
Levinthal & March, 1993; Nelson & Winter, 1982). tive ways to facilitate cross-functional team
Our model helps resolve inconsistent empirical meetings and learn standard approaches for
findings in the process management literature, identifying and solving problems, they tightly
while providing a base for further research into integrate and coordinate a broad set of activi-
the relationship between process management ties throughout the organization (Hackman &
activities and dynamic capabilities. Although in Wageman, 1995; Repenning, 1999).
prior research scholars have examined how rou- Process management programs also stress
tines influence organizational outcomes, they ongoing adherence to the resulting mapped and
have not fully explored the effects on innovation improved processes. Repeatable processes al-
or adaptation of institutionally mandated qual- low organizations both to reap the benefits of
ity programs aimed at stabilizing organization- improvement efforts and to continue incremen-
al routines and processes. tal improvements (Hackman & Wageman, 1995;
Harrington & Mathers, 1997; Mukherjee, Lapre, &
Van Wassenhove, 1998). The ISO 9000 program,
BACKGROUND:THEPROMISEAND REALITY
in particular, has a strong, explicit focus on ad-
OF PROCESSMANAGEMENT
hering to documented processes. ISO 9000 certi-
The process revolution has been marked by a fication requires third-party audits to periodi-
shift from the view of organizations as a collec- cally ensure that an organization follows its

documented practices (Cole, 1998; Harrington & shown performance benefits from adoption of a
Mathers, 1997). Similarly, the final stage of Six TQM program (Easton & Jarrell, 1998) and from
Sigma implementation requires coordinating winning a quality award (Hendricks & Singhal,
standardized best practices throughout an or- 1996), and Dean and Snell (1996) found that TQM
ganization (Harry & Schroeder, 2000). as one component of integrated manufacturing
Process management's proponents cite nu- was associated with performance benefits.
merous organizational benefits from process im- In contrast, Analog Devices' adoption of a
provement activities. These include increased TQM program resulted in short-term improve-
yields and less rework and waste as stream- ments in efficiency, such as higher yields and
lined processes eliminate non-value-added ac- less waste, but this was soon followed by finan-
tivities. Tighter intraorganizational linkages cial performance far worse than the industry
also increase efficiency by streamlining the average (Sterman et al., 1997). Similarly, proac-
handoffs between activities, thus speeding tive quality efforts at Alcoa initially led to com-
product development and delivery times (Dean petitive gains (Kolesar, 1993), yet soon after this
& Snell, 1996; Garvin, 1995). Further, products quality effort was initiated, Alcoa's CEO Paul
that result from improved processes are likely to O'Neill lamented the slow pace of change and
better satisfy customers, leading to increased initiated a "quantum-leap improvement" (Kole-
revenues and, ultimately, increased profits. sar, 1993: 161). Henderson and colleagues (1998)
Yet empirical research on the effects of pro- found that despite significant returns to process
cess management practices fails to yield con- improvements at Hewlett Packard, these activi-
clusive evidence of these promised benefits. The ties were underfunded out of concerns that pro-
few empirical studies that consider the implica- cess-focused efforts would impede product inno-
tions of process-focused techniques show no ev- vation. Sitkin and Stickel (1996) found that TQM
idence that these techniques are consistently efforts in an R&D setting led to organizational
helpful. Powell (1995) and Samson and Terzi- conflicts, distrust, conformity, and, in turn, to
ovski (1999) found no relationship between pro- decreased innovation. And Staw and Epstein
cess management utilization measures and or- (2000) found that while TQM adoption was asso-
ganizational performance. Ittner and Larcker ciated with higher CEO compensation, there
(1997) found process management techniques was no association between adoption and fi-
associated with increased performance in the nancial performance.
automobile industry, but decreased perfor- In other empirical research scholars have
mance in the computer industry. Others have aimed at resolving these inconsistent findings.
noted that poor financial performance has fol- This work suggests that outcomes fall short of
lowed the process-focused efforts necessary for anticipated benefits, because while firms adopt
winning the Baldrige Award (Garvin, 1991; Hill, TQM, they fail to fully implement the associated
1993). Winners like Motorola, Cadillac, and Fed- efficiency-generating practices (Easton & Jarrell,
eral Express have suffered subsequent financial 1998; Westphal et al., 1997; Zbaracki, 1998) or fail
setbacks, leading some Wall Street analysts to to give them sufficient time or the right culture
recommended shorting the stocks of Baldrige in which to work (e.g., Cameron & Barnett, 2000;
Award winners (Garvin, 1991). Sterman et al., 1997). Further, it may be that, in
In related research scholars also have ex- implementing TQM and related process im-
plored the organizational effects of TQM pro- provement activities, firms drive in fear, which,
gram adoption. Because TQM programs include in turn, triggers resistance to these administra-
actual implementation of a different bundle of tive practices (Repenning, 2000).
associated practices within each organization It may also be that the effects of process man-
(Hackman & Wageman, 1995; Westphal et al., agement activities are contingent on the techno-
1997), it is difficult to determine whether organi- logical and organizational contexts in which
zations implement process-focused techniques. they are applied. The initial successes of pro-
Even so, this research has also produced con- cess management activities have spurred their
flicting or paradoxical results. Kearns and Nad- spread beyond manufacturing into upstream ac-
ler (1992) and Wruck and Jensen (1994) report on tivities, such as processes for selecting and de-
TQM efforts that led to substantial organization- veloping technological innovations, or down-
al improvements. Similarly, event studies have stream, into distribution, sales, and service

activities. Indeed, the 9001 version of the ISO hardt, 1997; Levinthal, 1991; Nelson & Winter,
9000 program involves processes for product de- 1982; Tushman & Nelson, 1990). Thus, to under-
sign, development, and service, while Design for stand how process management techniques af-
Six Sigma similarly promotes extending process fect organizational adaptation, we first address
control techniques into R&D, including product how they affect both exploratory and exploit-
design and development activities (Harry & ative innovation. A firm's ability to innovate pro-
Schroeder, 2000; ISO, 1999). Process manage- vides its senior team with options either to rein-
ment practices have also been applied to senior force or destabilize a technological regime
team processes (e.g., Garvin, 1995; Ghoshal & (Burgelman, 1994; McGrath, 1999; Tushman &
Bartlett, 1995; Hammer & Stanton, 1999). Murmann, 1998). Innovations generated within
Process management practices may be or absorbed by firms provide variation to proac-
adopted because of bandwagon effects, created tively shape or reactively respond to technolog-
when other high-status organizations adopt ical transitions (Brown & Eisenhardt, 1997;
them (Abrahamson, 1996; Abrahamson & Rosen- Cohen & Levinthal, 1990; Tushman & O'Reilly,
kopf, 1993; Meyer & Rowan, 1977; Staw & Epstein, 1997). An organization's dynamic capabilities
2000). As these techniques are embraced by depend on simultaneously exploiting current
highly visible organizations like IBM, GE, 3M, technologies and resources to secure efficiency
and Amazon.com, among others (Hammer & benefits and creating variation through explor-
Stanton, 1999), they become taken for granted atory innovation (Ghemawat & Costa, 1993;
and viewed as sensible approaches for operat- March, 1991; Teece et al., 1997). As process man-
ing organizations (cf. Meyer & Rowan, 1977). agement techniques focus on continuous im-
These pressures may lead to process manage- provement in routines and variation reduction
ment utilization, even in organizational or envi- (Anderson, Rungtusanatham, & Schroeder, 1994;
ronmental contexts where it is ineffective or Hackman & Wageman, 1995; Harry & Schroeder,
even harmful. 2000), their increased utilization in an organiza-
Further, once adopted, the spread of process tion affects the balance between exploratory
management techniques beyond manufacturing and exploitative innovation (Benner & Tushman,
gives rise to unintended effects on innovation 2002).
and adaptation. Activities focused on measur- Process management influences technologi-
able efficiency and variance reduction drive out cal innovation in several ways. Process-focused
variance-increasing activities and, thus, affect activities stabilize the resource allocation and
an organization's ability to innovate and adapt decision processes that determine which tech-
outside of existing trajectories (e.g., Levinthal & nological projects will be supported (cf. Chris-
March, 1993; Sutcliffe, Sitkin, & Browning, 2000; tensen & Bower, 1996). Process management
Weick, 1995). Core capabilities may become core techniques also tighten internal communication
rigidities or competency traps in rapidly chang- linkages and affect the types of technological
ing environments (Leonard-Barton, 1992; changes that are recognized and addressed
Levinthal, 1991; Levitt & March, 1988). (Henderson & Clark, 1990). Programs like Design
Thus, although process management activi- for Six Sigma or ISO 9001 also ensure that pro-
ties are employed to help organizations adapt, cess management activities extend to innova-
to the extent that these practices increase iner- tion design and development processes (e.g.,
tia and limit variation creation, they may in- Harry & Schroeder, 2000). We argue that process
stead impede adaptation in all but routine con- management activities will facilitate some in-
texts (e.g., Levinthal, 1997a). Yet there has been novation types but dampen others. We first need
no effort to build testable theory to explore these to clarify innovation types.
broader implications of process management
activities for innovation and adaptation.
Types of Innovations
AND We classify innovations along two dimen-

PROCESSMANAGEMENT
sions: (1) their proximity to the current techno-
TECHNOLOGICAL
INNOVATION
logical trajectory and (2) their proximity to the
Technological innovation is a central engine existing customer/market segment (Abernathy &
of organizational adaptation (Brown & Eisen- Clark, 1985). On the technological dimension,

different innovation types have contrasting de- Process Management's Effects on Innovation
terminants and organizational effects (Morone,
In this section we develop propositions about
1993; Tushman & Smith, 2002). Incremental inno-
the relationship between process management
vation, characterized by small changes in a
and innovation. Our propositions and the links
technological trajectory, builds on the firm's cur-
between them are outlined in Figure 1.
rent technical capabilities, while radical inno-
Because the implementation of process man-
vation fundamentally changes the technologi-
cal trajectory and associated agement techniques initially entails an explicit
organizational
focus on innovation and change in organization-
competencies (Dosi, 1982; Green, Gavin, &
al activities, it is likely to spur innovation (Win-
Smith, 1995). Innovation can be further catego-
rized by how it affects existing subsystems ter, 1994). Tacit organizational routines are re-
vealed and made explicit (Brown & Duguid,
and/or linking technologies (Baldwin & Clark,
2000; Henderson & Clark, 1990; Tushman & Mur- 1991), leading to richer cognitive models of or-
mann, 1998). Modular innovations affect sub- ganizational activities (Repenning, 1999); firms
can learn before doing (Pisano, 1996). Team ap-
system or component technology, leaving link-
ing mechanisms intact, while architectural proaches to problem solving, decentralized de-
innovations involve changes in how subsystems cision making, and colocated information facil-
are linked together (Henderson & Clark, 1990; itate the search for solutions to help drive initial
improvements in efficiency in organizational ac-
Iansiti & Clark, 1994). For example, the addition of
a stepper motor was a modular innovation in the tivities (see also Hackman & Wageman, 1995,
and Wruck & Jensen, 1994). In underperforming
photolithography industry, while the move from
14- to 8-inch disk drives was an architectural in- firms the initial organizational changes associ-
novation in the disk drive industry (Christensen & ated with process management may well be
Bower, 1996;Henderson & Clark, 1990). radical (e.g., Kearns & Nadler, 1992), or, in the
Innovations have also been classified by case of lethargic competitors, process manage-
whether they address the needs of existing cus- ment efforts can lead quickly to substantial per-
tomers or are designed for new or emergent formance gains (e.g., Kolesar, 1993).
markets (Christensen & Bower, 1996). Products While this innovative phase of process man-
designed for new customer sets are often organ- agement entails substantive change, such
izationally disruptive and require significant change is focused on improving efficiency
departures from existing activities. For exam- within an existing technological trajectory. The
ple, the advent of digital photography not only founders of process management focused on in-
represented a technological change from chem- cremental and exploitative innovation, rather
ical-based film but also involved new distribu- than radical, architectural, or exploratory inno-
tion channels for electronic cameras. For some vation. The associated process-oriented tools
photography firms, the change in marketing and and techniques, thus, are aimed at making an
distribution presented a challenge equal or organization more efficient through incremental
greater to the technological change (Tripsas & improvements in processes and outputs (Ander-
Gavetti, 2000). Similarly, progressively smaller son et al., 1994; Harry & Schroeder, 2000). Indeed,
disk drives were innovations for emergent cus- the prescriptive literature encourages organiza-
tomer sets in the disk drive industry. These tech- tions to view improvements as controlled exper-
nologically simple innovations created organi- iments that involve repetition of practices and
zational challenges for incumbents, and they measurement prior to making small, testable
found it difficult to respond (Christensen, 1998). changes (e.g., Harry & Schroeder, 2000). More-
Incremental technological innovations and over, process improvement techniques apply to
innovations designed to meet the needs of ex- the organization's current activities; innovation
isting customers are exploitative and build and change in those activities are accomplished
upon existing organizational knowledge. In by paring from existing routines. Once changed,
contrast, radical innovations or those for emer- these improved routines become standardized
gent customers or markets are exploratory, best practices.
since they require new knowledge or depar- After this initial phase of innovation, repeti-
tures from existing skills (Levinthal & March, tion of organizational routines through adher-
1993; March, 1991). ence to best practices leads to deepened com-

306 academy of management Review April
P11
P10 P9,
Adaptation P8, Performance
P7,
Responsiveness
* *
Adaptation:
technological
Adaptation P11
technological
cycles:
and Environment
change
change
Incremental
Nonincremental
* *
Technology
Innovation
subunits
1 form subunits
P1 P2 P3 P4 P5
dual withacross
or
Technological sets
FIGURE coupling
on emergent coupling
innovation innovation Organizational
innovation for for
sets Tight
Loose
customer innovation * *
organization:
Ambidextrous
Effects
current
Incremental
Innovation customer
Innovation
Architectural
Radical
Technological * *
Exploitation: * * *
Exploration:
P6
Management's
Process
processesto processes
management process
of
processes
improved
Streamlining
Mapping Adhering
Process * *
management
activities:
Extent *

petence (e.g., Pisano, 1996). As an organization trous results. Process management techniques
learns and increases its efficiency through rep- stabilize organizational routines and tighten the
etition of a set of activities, its subsequent inno- linkages between them, yet they make cross-
vation is increasingly incremental (Levinthal & boundary, cross-community linkages more diffi-
March, 1993; Levitt & March, 1988). Thus, while cult (Sitkin & Stickel, 1996). Organizations fo-
process management activities drive innova- cused on incremental enhancements of current
tion, the focus on variation reduction, search for technology treat architectural innovation as
incremental improvements in routines, and in- merely incremental, fail to forge linkages across
creased proficiency through repetition of organ- organizational boundaries, and, in turn, under-
izational activities ensures that this innovation perform (Henderson, 1993; Lawless & Anderson,
will be in the neighborhood of existing capabil- 1996; Tripsas, 1997). Although incremental inno-
ities. As process management activities extend vation may, in some circumstances, actually ac-
in an organization over time, more and tighter commodate architectural or modular innova-
linkages between routines further constrain in- tions, adherence to standardized best practices
novation to incremental changes in existing ac- ensures repetition of practices through these
tivities and outputs. stable linkages within local domains, and an
organization's ability to actually take advan-
Proposition 1: Increases in process
tage of subsystem and linking technologies is
management practices promote incre- stunted.
mental innovation.
Process management activities also aim at
management practices decrease ar-
improvements in processes and outputs for the chitectural innovation.
organization's existing customers. Measures of
customer satisfaction are collected and used as Process management techniques, by design
the basis of improvement and coordination ef- and intent, exploit existing capabilities. Where
forts (Cole, 1998; Garvin, 1988; Hackman & short-term performance pressures, the demands
Wageman, 1995; Powell, 1995), and such efforts of existing customers, and ease of measurement
are reinforced by supportive resource allocation dominate, exploitation overwhelms exploration
systems. As the locus of decision making is (Levinthal & March, 1993; March, 1991; Sitkin et
pushed down in the firm, multiple levels of the al., 1994). Process management affects variation
hierarchy are focused on meeting customer creation through the use of statistical tech-
needs (Winter, 1994). Von Hippel (1988) and niques designed to reduce variation, and also
Christensen and Bower (1996) show how persis- through repetition of linked sets of streamlined
tent resource allocation processes among equip- organizational processes. Rapid exploitation, as
ment suppliers and disk drive manufacturers employees carry out activities in coordinated
channeled innovation away from product devel- processes, produces measurable, short-term
opment for emergent customer sets. Thus, as benefits. As an organization achieves faster new
process management creates an organizational product introductions, cost reductions, or im-
focus on better understanding and satisfying provements in customer satisfaction, these out-
existing customers, it also channels innovation comes drive the organization to persist with
into areas that benefit existing customers. greater zeal in extending process activities and
linking disparate parts of the firm (Levinthal &
March, 1993). Managers with documented suc-
management practices promote inno-
cesses from process management may be pro-
vation for existing customer sets.
moted and charged with extending such activi-
Henderson and Clark (1990) found that, in the ties to more domains. Those opposed to the
photolithography equipment industry, increas- encroachment of process management in cre-
ingly stable routines and communication link- ative or innovative domains may opt out of the
ages within firms constrained their ability to organization, further increasing the influence of
initiate and respond to changes in subsystem process-intensive activities (Brown & Duguid,
and linking technologies. Incumbent firms 2000).
treated seemingly minor architectural innova- As systems of improved routines are repeated
tions as incremental innovations, with disas- throughout an organization, the organization be-

comes more competent and efficient, but var- (e.g., Brown & Duguid, 2000; Sitkin & Stickel,
iation in the outcomes of those activities 1996; Tripsas & Gavetti, 2000).
decreases (March, 1991). The objective of stabi- Incremental innovation associated with pro-
lizing organizational routines and "attacking cess management reduces significant explor-
variation" (Harry & Schroeder, 2000) contrasts atory activity and learning outside the existing
with the variation-increasing roles played by technological trajectory (Levinthal & March,
internal corporate venturing units (Burgelman, 1993; March, 1991). The path-dependent nature of
1983), as well as with redundant and overlap- innovation suggests an even longer-lasting ef-
ping information designed to increase organiza- fect of process management practices. Past in-
tional innovation and variety (Nonaka & Takeu- novative activities play a role in future innova-
chi, 1995). This explicit focus on incremental tion by providing a firm with a knowledge base
innovation and increased proficiency with local that allows it to absorb technological compe-
search makes it unlikely that process manage- tence from external sources (Cohen & Levinthal,
ment activities will produce innovations that 1990; Levitt & March, 1988). An organization that
significantly depart from the neighborhood of lacks exploration in one period may be excluded
the organization's existing technological or mar- from areas of future exploratory activity be-
ket competencies. cause it has no relevant knowledge base (e.g.,
Cohen & Levinthal, 1990; Teece et al., 1997).
Further, over time, process management influ-
ences variation creation by affecting the selec- Since process management techniques reduce a
tion of innovations. Short-term, easy-to-measure firm's exploratory activities, its absorptive ca-
pacity will be stunted. Therefore, the firm is less
efficiency improvements make vague, uncer-
tain, difficult-to-quantify exploratory activities likely to produce subsequent innovations that
less attractive (Levinthal & March, 1993; March, incorporate new technologies.
1991). As process management practices spread Proposition 4: Increases in process
in an organization, the predominant measures management practices decrease radi-
of effectiveness are increasingly focused on cacl innovation.
speed, efficiency, and reductions in costs or Measures of customer satisfaction drive im-
waste. These dynamics lead to selecting inno-
provement efforts through an increasingly coor-
vations that leverage efficient, streamlined
dinated system of processes. As improved man-
manufacturing or distribution processes or that create
ufacturing or distribution processes
utilize materials that are cost-effectively ob-
measurable improvements for existing custom-
tained from streamlined purchasing processes. ers, process management will drive innovations
Such innovations build on existing firm capabil- that continuously improve products for existing
ities and tend to be closer to existing products. customers. Emergent customer sets or market
Because innovation development and selec- segments often do not lend themselves to mea-
tion processes are linked with other improved surement or are highly uncertain (Sitkin et al.,
processes, the focus shifts away from creating 1994; von Hippel, 1988). Thus, exploratory inno-
variation. A focus on refining project selection vation into new markets becomes increasingly
processes to yield continuous improvement in unattractive compared to the short-term, tangi-
the speed or success rates of new products tips ble, measurable successes from further improv-
project selection toward those with greater pre- ing existing capabilities (Christensen, 1998;
dictability, lower variation, and increased cer- Leonard-Barton, 1992).
tainty. Thus, as the reach of process manage- Through process management practices, an
ment activities extends further into research, organization becomes increasingly skilled at
R&D project selection activities, or product de- producing outputs that leverage existing knowl-
velopment, radical innovation projects increas- edge about inputs, technologies, manufacturing
ingly give way to more certain, incremental ac- techniques, or distribution channels. New inno-
tivities (e.g., Henderson et al., 1998). As process vations that further utilize these capabilities
intensity increases, even structures designed to will benefit from these efficiencies and lend
produce radical innovations (e.g., heavyweight themselves to even more measurable successes.
teams or independent units) increasingly will As these codified and tightly linked processes
produce innovations close to past innovations begin to further influence the development and

delivery of technological innovations, innova- units. In these loosely coupled organizational

tions for emergent or new markets will be forms it is unclear where the integration required
squeezed out. This is likely to further exacerbate to drive streams of innovation is accomplished.
the tendency to innovate for existing customers Ambidextrous or dual organizational forms
that Christensen and Bower (1996) observed in are organizational architectures that build in
the disk drive industry. both tight and loose coupling simultaneously
(Bradach, 1997; Sutcliffe et al., 2000; Tushman &
Proposition 5: Increases in process O'Reilly, 1997). These organizational forms are
management practices decrease inno- not loosely coupled, nor do they switch between
vation for new customer sets.
contrasting structures. Rather, ambidextrous or-
ganizations are composed of multiple tightly
coupled subunits that are themselves loosely
Exploration, Exploitation, and Ambidextrous
coupled with each other. Within subunits the
Organizations
tasks, culture, individuals, and organizational
Abernathy (1978) highlights the inconsisten- arrangements are consistent, but across sub-
cies between activities focused on productivity units tasks and cultures are inconsistent and
improvements and cost reductions and those fo- loosely coupled. Strategic integration-the abil-
cused on innovation and flexibility. He ques- ity to drive innovation streams and take advan-
tions whether it is possible for organizations to tage of contrasting organizational capabili-
pursue both types of activities simultaneously. ties-occurs at the senior team level of analysis.
Although both types of activities are important In contrast to Lawrence and Lorsch's (1967)
for organizational survival, exploration and ex- design, ambidextrous organizational designs
ploitation are contradictory organizational pro- are composed of highly differentiated but
cesses (Adler, Goldoftas, & Levine, 1999; March, weakly integrated subunits. While the explor-
1991; Teece et al., 1997). How can organizations atory units are small and decentralized, with
balance these potentially conflicting activities? loose cultures and processes, the exploitation
The organizational literature has stressed the units are larger and more centralized, with tight
importance of a balance between efficiency and cultures and processes. Exploratory units suc-
exploration, but there are multiple points of ceed by experimenting-by frequently creating
view on how organizations actually strike this small wins and losses (Sitkin, 1992). Because
balance. process management tends to drive out experi-
Hedberg, Nystrom, and Starbuck (1976) sug- mentation, it must be prevented from migrating
gest that organizations engage in multiple into exploratory units and processes. In contrast,
forms of learning by alternating between differ- exploitation units that succeed by reducing vari-
ent organizational designs and by being both ability and maximizing efficiency and control
consistent and inconsistent in their actions. Sim- are an ideal location for the tight coordination
ilarly, Brown and Eisenhardt (1998) suggest that associated with process management efforts.
dynamic capabilities are rooted in an organiza- These contrasting, inconsistent units must be
tion's ability to rhythmically switch between physically and culturally separated from one
more organic and more mechanistic structures. another, have different measurement and incen-
In contrast to alternating between organizing tives, and have distinct managerial teams (Bra-
modes, others argue for creating loosely coupled dach, 1997; Nonaka, 1988, 1991; Sutcliffe et al.,
organizations, where the experimenting units 2000; Tushman & O'Reilly, 1997). For example, in
are highly buffered from the exploiting units Hewlett Packard's (HP's) Scanner Division, the
(Levinthal, 1997b; Weick, 1976). For example, more routine flatbed scanners had a completely
Burgelman (1991) describes buffered-induced different organizational architecture from the
and autonomous processes at Intel, whereas emerging consumer/knitting technology scan-
Leonard-Barton (1995) describes experimenting ners. These distinct units were physically sepa-
units distinct from units focused on more incre- rated from one another and had their own man-
mental innovation. At the extreme, Christensen agement teams.
(1998) suggests that because of the disruptive Because dynamic capabilities are rooted in
nature of the technology, experimenting units driving streams of innovation in a product class
must be completely separated from exploiting (Morone, 1993; Teece et al., 1997), these highly

differentiated but loosely coupled subsystems term effectiveness (McGrath, 1999). These inter-
must be strategically integrated by the senior nally inconsistent operating modes must be
team. Such strategic linkage is anchored by strategically linked by the senior team through
common aspiration levels and a senior team their aspirations and actions and through a lim-
that both provides slack to the experimental ited set of core values (Hambrick, Nadler, &
subunits and holds the differentiated units to Tushman, 1998).
fundamentally different selection and search Thus, the impact of process management ac-
constraints (Levinthal & March, 1993; Rotemberg tivities on innovation outcomes will be contin-
& Saloner, 2000). To be effective, ambidextrous gent on organizational context. In ambidextrous
senior teams must develop processes for estab- organizational forms, process management ac-
lishing new, forward-looking cognitive models tivities drive exploitative innovation, even as
for exploration units, while allowing backward- they are buffered from intruding on exploratory
looking experiential learning to rapidly unfold innovation.
for exploitation units (Gavetti & Levinthal, 2000;
Louis & Sutton, 1989). Proposition 6: In the context of an ambi-
To create dual organizational structures, se- dextrous organizational form, increases
nior teams must develop techniques that permit in process management practices in-
them to be consistently inconsistent as they crease exploitative innovation but do
steer a balance between the need to be small not dampen exploratory innovation.
and large, centralized and decentralized, and
focused both on the short term and long term, PROCESSMANAGEMENTAND
simultaneously (Gavetti & Levinthal, 2000; Hed- ORGANIZATIONAL
ADAPTATION
berg et al., 1976; Weick, 1995). If the locus of
strategic integration is low in the firm, experi- As technological capabilities affect organiza-
mentation will be stunted, as the short-term suc- tional fate, so, too, does process management's
cesses of rapid exploitation will drive out explo- influence on innovation have implications for
ration. If this integration is at too high a level in organizational adaptation. Process manage-
a multidivisional firm, or done across indepen- ment activities, through their impact on organi-
dent firms (e.g., Christensen, 1998), the underly- zational processes, accentuate organizational
ing understanding of an innovation stream's dy- inertia and slow responsiveness to technologi-
namic will be dampened, and the ability to drive cal transitions. We now explore process man-
disruptive or radical change restricted. agement's effects on organizational adaptation
For example, at HP, when the Scanner Divi- in contrasting technological contexts.
sion's general manager and his senior team pro-
vided the integration between the unit's two
Technology Cycles
highly differentiated subunits, the division was
able to effectively both explore and exploit. Organizational environments are character-
However, when this general manager was pro- ized by cycles of technological variation, alter-
moted, the consumer unit was spun out to a nating between periods of incremental change
sector executive. Because this corporate execu- and periods of rapid innovation (Abernathy &
tive did not deeply understand the competitive Utterback, 1978; Sanderson & Uzumeri, 1995;
issues in the scanning industry, he cancelled Tushman & Anderson, 1986). Organizations out-
the more exploratory consumer unit when it did side an industry often introduce discontinuous
not show quick profits. technological advances, and this triggers peri-
While complex and politically difficult, ambi- ods of rapid innovation and change (Foster,
dextrous organizational forms permit a firm 1986; Sull, Tedlow, & Rosenbloom, 1997). This
with highly differentiated units to drive process period of rapid innovation, or era of ferment,
management, with its associated variation re- ends when a dominant technology design
duction and control, as well as exploration and emerges (Abernathy, 1978; Anderson & Tush-
option creation. Experimental units provide man, 1990; Tushman & Rosenkopf, 1992). With
variation from which the senior team can learn the emergence of a dominant technological re-
about and, in turn, bet on the future, even as the gime, the nature of technical change shifts from
exploitation units build capabilities for short- product innovation to a relatively long period of

process innovation and incremental refine- Proposition 7: Increases in process

ments of the selected technology (Abernathy & management practices will improve
Utterback, 1978; Ettlie & Reza, 1992; Utterback, performance in eras of incremental
1994). This period of incremental technical technological change.
change, in turn, is punctuated by the subse-
Tighter coordination and repetition of activi-
quent technological discontinuity.
ties embedded in standardized best practices
Thus, in the videocassette recorder (VCR) mar-
increase an organization's speed and efficiency.
ket, after a period of variation between alterna-
For example, streamlined linkages between pro-
tive formats, VHS became the industry standard.
cesses for innovation and manufacturing allow
Once this standard emerged, subsequent inno-
for rapidly screening and developing innova-
vation focused on incremental improvements
tions that best leverage downstream manufac-
and cost reductions in VHS technology (Cu-
turing or distribution capabilities. Further, as
sumano, Mylonadis, & Rosenbloom, 1992). Simi-
lar innovation dynamics have been found in a processes for identifying and addressing prob-
lems and opportunities in the environment are
range of industries, including typewriters, tele-
further refined and routinized by process man-
visions, computers, disk drives, operating sys-
agement's influence, decision making and prob-
tems, browsers, and microprocessors (see Tush-
lem solving also become more efficient (Miller,
man & Murmann, 1998, and Van de Ven, Angle, &
1993). Increasing organizational proficiency in
Poole, 1989).
recognizing and addressing recurring chal-
lenges leads to stable and increasingly efficient
communication channels and information filters
Adaptation in Stable Environments (Henderson & Clark, 1990; Tyre & Orlikowski,
1994). These stable patterns of communication
Because of the shifting nature of innovation
and interaction lead to the development of
requirements embedded in technology cycles,
norms, rules, and roles, which further channel
organizations must develop capabilities to
individual and group behavior into streamlined
move with, if not shape, these cycles (Hender-
activities that carry out an organization's mis-
son, 1993; Tushman & O'Reilly, 1997). During
sion more efficiently (Hackman, 1992; Nadler &
eras of incremental change, organizations that
sustain incremental innovation will be more ef- Tushman, 1998; Repenning, 1999). Over time, sta-
ble procedures and norms also drive increased
fective than those that initiate variance-increas-
demographic homogeneity within the organiza-
ing innovation. As process management activi-
tion, further speeding decision making and
ties stimulate incremental innovation, these
problem solving (Keck & Tushman, 1993; Wil-
activities will benefit organizations, when tech-
liams & O'Reilly, 1998). These factors drive an
nological environments are characterized by
incremental refinements of an existing techno- increasingly tightly integrated organization fo-
cused on fast response to existing customer
logical design. Greater consistency and effi-
requirements.
ciency will be beneficial in periods of incremen-
tal change. Rapid improvements in customer Proposition 8: Increases in process
satisfaction, combined with reduced waste and management practices will speed or-
costs, will likely further enhance organizational ganizational responsiveness during
effectiveness. eras of incremental technological
Organizations with routines and procedures change.
stabilized through process management activi-
ties, thus, are likely to do well in stable or pre-
dictable contexts (Donaldson, 1995; Hannan & Adaptation in Turbulent Environments
Freeman, 1984). Indeed, Ittner and Larcker (1997) Conversely, the ability to develop new tech-
found that process management was positively nological capabilities rapidly is especially crit-
associated with performance in the stable auto ical in environments characterized by rapid in-
industry, but not in the more dynamic computer novation and change (Brown & Eisenhardt, 1997;
industry. These environmental contingencies Teece et al., 1997; Tushman & Anderson, 1986).
are also supported by Sitkin et al. (1994) and Periods of technological ferment are character-
Sterman et al. (1997). ized by substantial product variation and mar-

ket uncertainty. Responding to environmental Case studies of process management efforts

uncertainty and variation requires similar vari- and simulation studies of problem-solving dy-
ety within the firm (Daft & Weick, 1984). A reduc- namics provide evidence of this tendency for
tion in variance-increasing activity within the increased resistance to change resulting from
firm prevents it from registering and/or respond- process management activities. Following suc-
ing to environmental uncertainty (Burgelman, cessful implementation of a process improve-
1994). Process management's variation-reducing ment program, Analog Devices' financial perfor-
focus restricts the development of alternatives mance fell far below the industry average. The
for responding to environmental changes. For company subsequently underwent a major reor-
example, Sitkin and Stickel (1996) found that ganization, including a reorientation of its prod-
TQM efforts in R&D settings not only bred dis- uct development efforts away from its core busi-
trust and dissension across the firms but also ness and into emerging markets (Sterman et al.,
drove out variability in R&D. 1997: 505). While process improvement efforts
Organizational outcomes are affected by de- were underway, internally generated innova-
layed or inadequate responses to environmental tions necessary for successful performance in a
turbulence (Henderson, 1993; Meyer, Brooks, & changed environment were not executed.
Goes, 1990). For example, IBM's relatively slow Extending the Analog Devices study, Repen-
response to personal computers resulted in the ning (1999) argues that, over time, process im-
successful entry of other less inertial competi- provement activities trigger cycles of increased
tors that fundamentally changed the nature of managerial control and bureaucratic proce-
the computer industry (Mitchell, 1989). Slow or dures, which undermine initial gains. Similarly,
incompetent responses to environmental shifts Alcoa CEO's frustration with the slow pace of
prevented incumbents from retaining their lead- continuous improvement might also be ex-
ership positions in the disk drive (Christensen & plained by the increased inertia associated with
Bower, 1996), photolithographic equipment (Hen- practices focused on incremental change and
derson & Clark, 1990), and watch (Glasmeier, improvement (Kolesar, 1993). Finally, Levinthal's
1991) industries, among others (Tushman & (1997b) modeling of adaptation dynamics shows
O'Reilly, 1997). Similarly, TQM practices have that tightly coupled organizations are subjected
been associated with lower performance in the to heightened inertia and high failure rates in
dynamic computer industry environment (Ittner changing environments. Because organizational
& Larcker, 1997; Sitkin et al., 1994). inertia is so strong, successful firms move
through environmental change by initiating re-
Proposition 9: Increases in process orientations (Tushman & Romanelli, 1985).
management practices will decrease Thus, the influence of process management
performance in eras of technological techniques on integrating and coordinating pro-
ferment. cesses can drive rapid exploitation and effi-
ciency, but also longer-term momentum and
The same practices that help an organization resistance to change. Although tightly coordi-
learn and achieve efficiency more quickly can nated and streamlined processes in product de-
also impede an organization's adaptation to ma- velopment and manufacturing may allow for
jor technological transitions (Levinthal, 1991, rapid response with incremental extensions or
1997a). Over time, as process management ac- enhancements of current capabilities, the asso-
tivities permeate an organization, the increas- ciated inertia is likely to make such an or-
ingly stable, tightly linked, and efficient rou- ganization slower to respond to subsystem, link-
tines that span an organization make anything ing, or radical technological change. Indeed,
more than incremental organizational change empirical results suggest that, in the digital
difficult (Hannan & Freeman, 1984). Organiza- camera industry, more extensive process man-
tions that have honed their skills in making in- agement activities dampened a firm's ability to
cremental changes in processes and products keep up with rapid technological change
develop momentum that works to impede major through new product introductions (Benner,
change (Miller & Friesen, 1980) or that trans- 2002). These results were particularly strong for
forms core competencies into rigidities (Leo- photography industry incumbents entering the
nard-Barton, 1992). new digital camera industry, suggesting that

the incremental improvements in routines that atory behaviors stunts an organization's ability
enhanced their efficiency in film technology to adapt in environments characterized by tech-
may have been inappropriate for responding to nological ferment and uncertainty. Ambidex-
the rapid change in digital technology. trous organizational forms isolate process man-
agement activities in subunits where reducing
Proposition 10: Increases in process variation and increasing control are strategi-
management practices will slow or- cally vital. An ambidextrous organizational de-
ganizational responsiveness in eras of sign allows for uncoupling the variance-
technological ferment. decreasing units and activities from those units
where variation is critical.
For example, at CibaVision, a single senior
Moderating Effects of Ambidextrous team created highly differentiated subunits,
Organizational Forms
where one subunit pursued low-cost lenses even
To maximize short-term performance and sur- as another subunit developed a substitute (dis-
vive in periods of incremental technology posable) lens. Similarly, at Ciba's Crop Protec-
change, firms need to accentuate incremental tion Division, a single general manager built
change, momentum, and inertia associated with multiple internally inconsistent subunits in or-
process activities. Multiple functions and activ- der to drive an innovation stream to keep plants
ities must be linked seamlessly throughout the healthy. One unit based in Switzerland focused
organization to efficiently deliver to and satisfy on the efficient development of chemical-based
existing customers. Without this concerted re- fungicides. Anchored in molecular biology, an-
finement of capabilities, firms may not survive other unit, based in the United States, focused
long enough to face or initiate technological on developing a seed that would grow plants
change. not needing fungicides. If this latter unit were
But process intensity also stunts incumbents' successful, it would cannibalize the chemical-
ability to take advantage of internally gener- based unit in Switzerland (Tushman & O'Reilly,
ated opportunities for discontinuous change 1997).
(Cooper & Smith, 1992; Foster, 1986). For exam- These experimental, variance-creating units
ple, even though Swiss producers invented the provide the options from which a firm's senior
quartz movement, it was American and Japa- managers can select in order to shape an evolv-
nese firms that actually introduced this techno- ing innovation stream, and they must be buff-
logical discontinuity (Glasmeier, 1991). Thus, ered from the enticing spread of efficiency im-
process management promotes only one dimen- provement in the existing technology. These
sion of a firm's dynamic capabilities. ideas further suggest that organizational form
As firms achieve short-term measurable suc- moderates the impact of process management.
cess with process management, they are likely
to increase their commitment to process inten- Proposition 11: In the context of an am-
sity and broaden its influence to ever more pro- bidextrous organizational form, in-
cesses. Yet firms must be capable of forgetting creases in process management prac-
their past, breaking rules and traditions, and tices will enhance responsiveness and
increasing variation in the service of architec- performance during eras of incremen-
tural and/or radical innovation or in meeting the tal technological change but will
needs of new customer segments (Hedberg et have no effect on responsiveness or
al., 1976;Weick, 1995). Because competencies are performance during eras of technolog-
hard to develop and the rates of environmental ical ferment.
change are substantial, we argue that dynamic
capabilities are not rooted in sequential atten- DISCUSSION
tion or rhythmic pacing (e.g., Brown & Eisen-
hardt, 1998) but, rather, in exploiting and explor- In the twenty-five years since Abernathy's ad-
ing simultaneously (Sutcliffe et al., 2000; monition about the productivity dilemma, there
Tushman & O'Reilly, 1997). has been an ever-increasing influence of the
Over time, process management crowds out ideology of process management on organiza-
exploratory innovation. A decrease in explor- tions. Over the past two and a half decades,

firms have had substantial institutional pres- complex organizational architectures are at the
sures, irrespective of their organizational or en- root of dynamic organizational capabilities.
vironmental contexts, to adopt TQM, Six Sigma,
or BPR;to get ISO 9000 certified; or to compete for
Implications for Future Research
the Baldrige or Deming awards. While the sym-
bolic importance of process management may Our model provides a starting point for future
be important, its substantive benefits are much research on how process management practices
less clear. affect firm dynamic capabilities. We advance
Our review suggests that inconsistent out- testable propositions about the effects of these
comes of process management practices can be institutionally mandated practices on both tech-
reconciled with attention to the context in which nological innovation and adaptation. Moreover,
these practices are employed. Process manage- our propositions provide a basis for understand-
ment activities are positively associated with ing how the effects of process management
organizational effectiveness in a limited set of practices unfold over time in an organization.
conditions: during periods of stability or incre- Future empirical work to test and modify these
mental change and for incremental innovation propositions must incorporate a longitudinal
or existing customers. In contrast, in a much perspective and assess how increases in such
wider set of conditions-during eras of ferment; activity over time affect an organization, rather
in turbulent environments; for new customer than compare differences associated with nom-
inal adoption of quality programs across firms.
segments; and for architectural, modular, and
radical innovation-process While our ideas may be relevant for all types
management activ-
and sizes of firms, they apply most readily to
ities are less conducive to organizational effec-
firms whose strategies include both exploitative
tiveness. Under these frequently occurring con-
and exploratory innovation. Our discussion con-
ditions, process management activities build
cerns the effects of increased process manage-
resistance to change and momentum and, in
ment intensity for those firms already chal-
turn, inhibit organizational variability. These in-
lenged with balancing such activities. Our
ertial outcomes of process management activi-
ties stunt a firm's ability to adapt. propositions are therefore less relevant for firms
whose strategies focus solely either on exploi-
But organizations must outcompete rivals
tation or exploratory innovation. In addition,
both in the short and long run. Organizations
these ideas may not apply to small start-ups in
that must meet current customer requirements
their initial phases-not yet challenged with
and new customer demands do not have the
balancing exploitation and improvements of an
luxury of choice-they must deal simulta- initial project with the exploratory development
neously with the inconsistent demands of ex- of subsequent products. Future research should
ploitation and exploration. Process manage- further test the boundaries of our propositions.
ment capabilities speed exploitation and Our review also suggests that, to understand
efficiency, and while they may allow organiza- the relations between process activities and or-
tions to survive in the short run, they simulta- ganizational outcomes more fully, we need to
neously dampen the exploration required for more explicitly couple research and theory in
longer-term adaptation. Ambidextrous organiza- process management, organizations, and strat-
tional forms reconcile these paradoxical de- egy. There is much to be gained by integrating
mands by building internally inconsistent archi- these perspectives on the phenomena and by
tectures within a single organization-- taking a contingency approach to process activ-
contrasting architectures that retain the benefits ities and organizational outcomes. Technologi-
of experimentation and variability, along with cal, environmental, and structural conditions
the benefits of exploitation and process control. clearly moderate the relations between process
These tightly coupled, internally inconsistent activities and organizational outcomes. If dy-
architectures must be tactically uncoupled. namic capabilities require exploitation as well
However, to drive streams of innovation, these as exploration, more complex organizational
inconsistent units must be strategically inte- forms are required, which, in turn, demand more
grated by the senior team. It may be that heter- complex senior team capabilities. This review,
ogeneous senior team capabilities coupled with then, indicates the potential in exploring more

deeply the relationships among organizational Toyota production system. Organization Science, 10:
43-68.
architectures, senior team behaviors, innovation
streams, and process management. Anderson, J.C., Rungtusanatham, M.,&Schroeder,R. G. 1994.
A theory of quality management underlying the Deming
management method. Academy of Management Re-
view, 19:473-509.
Implications for Managerial Practice Anderson, P., & Tushman, M. L. 1990.Technological discon-
tinuities and dominant designs: A cyclical model of
Finally, our review has implications for prac-
technological change. Administrative Science Quar-
tice. The powerful institutional pressures to
terly, 35: 604-633.
adopt process management practices have cut
Baldwin, C., & Clark, K. 2000. Design rules: The power of
across industries and firms, regardless of age or
modularity. Working paper, Harvard Business School,
size. Despite the coercive pressure and promise Boston.
of legitimacy, managers need to exercise great
Benner, M.2002.Dynamic or static capabilities? Process man-
care in when and where to adopt these prac- agement and adaptation to technological change. Work-
tices. Whereas in stable, technologically certain ing paper, The Wharton School, University of Pennsyl-
settings these practices may be productive, in vania, Philadelphia.
uncertain or technologically complex contexts Benner, M., & Tushman, M. 2002. Process management and
these practices may be quite counterproductive. technological innovation: A longitudinal study of the
Indeed, the usefulness of process manage- photography and paint industries. Working paper, The
ment practices may be much more constrained WhartonSchool,Universityof Pennsylvania,Philadelphia.
than the popular literature suggests. Process Bradach,J. 1997.Using the plural form in the management of
restaurant chains. Administrative Science Quarterly, 42:
management and its associated technologies 276-303.
and philosophies are conservative and resistant
to anything but incremental or competence- Brown,J. S., & Duguid, P. 1991.Organizational learning and
communities of practice. Organization Science, 2: 40-47.
enhancing innovation. This variance-hostile fo-
cus on incremental change and existing custom- Brown,J. S., & Duguid, P. 2000.The social life of information.
Boston: Harvard Business School Press.
ers, from the senior team to lower levels of the
firm, severely stunts a firm's dynamic capabili- Brown, S. L., & Eisenhardt, K. M. 1997.The art of continuous
change: Linking complexity theory and time-paced evo-
ties. It is the promise of ambidextrous organiza-
lution in relentlessly shifting organizations. Administra-
tional forms and heterogeneous senior teams tive Science Quarterly, 42: 1-34.
that provides the possibility of building organi-
Brown,S. L.,&Eisenhardt,K.M. 1998.Competing on the edge:
zations capable of both celebrating process ac- Strategy as structured chaos. Boston: Harvard Business
tivities as well as limiting their damage. School Press.
Burgelman, R. 1983.A process model of internal corporate
venturing in a diversified major firm. Administrative
REFERENCES Science Quarterly, 28: 223-244.
Burgelman, R. 1991.Intra-organizationalecology of strategy
Abernathy, W. J. 1978.The productivity dilemma. Baltimore:
Johns Hopkins University Press. making and organizational adaptation. Organization
Science, 2: 239-262.
Abernathy, W. J., & Clark, K. 1985. Mapping the winds of
creative destruction. Research Policy, 14:3-22. Burgelman, R. 1994.Fading memories: A process theory of
strategic business exit. Administrative Science Quar-
Abernathy, W. J., & Utterback, J. M. 1978.Patterns of indus- terly, 39: 24-56.
trial innovation. Technology Review, 80(7):40-47.
Cameron, K.,& Barnett, C. 2000.Organizational quality as a
Abrahamson, E. 1996. Management fashion. Academy of cultural variable: An empirical investigation of quality
Management Review, 21: 254-285. culture, processes, and outcomes. In R. E. Cole & W. R.
Abrahamson, E., & Rosenkopf, L. 1993.Institutional and com- Scott (Eds.), The quality movement & organization the-
petitive bandwagons: Using mathematical modeling as ory: 271-294. Thousand Oaks, CA: Sage.
a tool to explore innovation diffusion. Academy of Man- Christensen, C. M. 1998.The innovator's dilemma: When new
agement Review, 18:487-517. technologies cause great firms to fail. Boston: Harvard
Academy of Management Review. 1994. 19(3). Business School Press.
Adler, P. S. 1993.Time-and-motion regained. Harvard Busi- Christensen, C. M., & Bower, J. L. 1996. Customer power,
ness Review, 71(1):97-108. strategic investment, and the failure of leading firms.
Adler, P. S., Goldoftas, B., &Levine, D. 1999.Flexibility versus Strategic Management Journal, 17: 197-218.
efficiency? A case study of model changeovers in the Cohen, W. M.,& Levinthal, D. A. 1990.Absorptive capacity: A

new perspective on learning and innovation. Adminis- Ghemawat, P., & Costa, J. 1993.The organizational tension
trative Science Quarterly, 35: 128-152. between static and dynamic efficiency. Strategic Man-
Cole, R. E. 1998. Learning from the quality movement: What agement Journal, 14:59-73.
did and didn't happen and why? California Manage- Ghoshal, S., & Bartlett, C. A. 1995.Changing the role of top
ment Review, 41(1): 43-73. management: Beyond structure to processes. Harvard
Cole, R. E., & Scott, W. R. (Eds.). 2000. The quality movement
Business Review, 73(1):86-96.
& organization theory. Thousand Oaks, CA: Sage. Glasmeier, A. 1991.Technological discontinuities and flexi-
Cooper, A., & Smith, C. 1992. How established firms respond
ble production networks: The case of Switzerland and
to threatening technologies. Academy of Management the world watch industry. Research Policy, 20: 469-485.
Executive, 6(2): 55-70. Green, S., Gavin, M., & Smith, L. 1995.Assessing a multidi-
Cusumano, M., Mylonadis, Y., & Rosenbloom, R. 1992. Stra-
mensional measure of radical innovation. IEEETransac-
tegic maneuvering and mass market dynamics: The tri- tions on Engineering Management, 42(3):203-214.
umph of VHS over Beta. Business History Review, 66: Hackman, J. R. 1992. Group influences on individuals. In
51-95. M. D. Dunnette (Ed.),Handbook of industrial and organ-
Daft, R. L., & Weick, K. E. 1984. Toward a model of organiza- izational psychology: 1455-1525.Palo Alto, CA: Consult-
tions as interpretation systems. Academy of Manage- ing Psychologists Press.
ment Review, 9: 284-295. Hackman, J. R., & Wageman, R. 1995.Total quality manage-
Dean, J. W., Jr., & Bowen, D. E. 1994. Management theory and ment: Empirical, conceptual, and practical issues. Ad-
total quality: Improving research and practice through ministrative Science Quarterly, 40: 309-342.
theory development. Academy of Management Review, Hambrick, D., Nadler, D., & Tushman, M. 1998. Navigating
19: 392-418. change: How CEOs,top teams and boards steer transfor-
Dean, J. W., Jr., & Snell, S. 1996. The strategic use of inte- mation. Boston: HarvardBusiness School Press.
grated manufacturing: An empirical example. Strategic Hammer, M., & Champy, J. 1993.Reengineering the corpora-
Management Journal, 17: 459-480. tion: A manifesto for business revolution. New York:
Deming, E. W. 1986. Out of crisis. Cambridge, MA: MIT Press. Harper Business.
Donaldson, L. 1995. American anti-management theories of Hammer, M., & Stanton, S. 1999. How process enterprises
management. Cambridge: Cambridge University Press. really work. HarvardBusiness Review, 77(6):108-118.
Dosi, G. 1982. Technological paradigms and technological Hannan, M. T., & Freeman, J. 1984. Structural inertia and
trajectories. Research Policy, 11: 147-162. organizational change. American Sociological Review,
49: 149-164.
Easton, G. S., & Jarrell, S. L. 1998. The effects of total quality
management on corporate performance: An empirical Harrington,H. J.,& Mathers,D. D. 1997.ISO 9000and beyond:
investigation. Journal of Business, 71: 253-307. From compliance to performance improvement. New
York:McGraw-Hill.
Eisenhardt, K., & Martin, J. 2000. Dynamic capabilities: What
are they? Strategic Management Journal, 21: 1105-1121. Harry, M. J., & Schroeder, R. 2000. Six Sigma: The break-
Ettlie, J. E., & Reza, E. M. 1992. Organizational integration and through management strategy revolutionizing the
world's top corporations. New York:Currency.
process innovation. Academy of Management Journal,
35: 795-827. Hayes, R. H., & Abernathy, W. J. 1980.Managing our way to
economic decline. In M.L.Tushman &W. L.Moore(Eds.),
Feyder, S. 2001. 3M to use favorite GE quality program; Six
Sigma adoption to be companywide. Minneapolis Star Readings in the management of innovation: 11-25.
Tribune, January 25: 1D.
Marshfield, MA:Pitman.
Foster, R. 1986. Innovation: The attacker's advantage. New Hedberg, B., Nystrom, P., & Starbuck, W. 1976.Camping on
York: Summit Books. seasaws: Prescriptions for self-designing organizations.
Administrative Science Quarterly, 21: 41-65.
Gabor, A. 2001. Quality revival, Part 2: Ford embraces Six
Sigma. New York Times, June 13: C5. Henderson, R. 1993.Underinvestment and incompetence as
responses to radical innovation. Rand Journal of Eco-
Garvin, D. A. 1988. Managing quality. New York: Free Press. nomics, 24: 248-269.
Garvin, D. A. 1991. How the Baldrige Award really works. Henderson, R. M., & Clark, K. B. 1990.Architectural innova-
Harvard Business Review, 69(6): 80-93. tion: The reconfiguration of existing product technolo-
Garvin, D. A. 1995. Leveraging processes for strategic advan- gies and the failure of established firms. Administrative
tage. Harvard Business Review, 73(5): 77-90. Science Quarterly, 35: 9-30.
Garvin, D. A. 1998. The processes of organization and man- Henderson, R., Del Alamo, J.,Becker, T., Lawton, J.,Moran,P.,
agement. Sloan Management Review, 39(4): 33-50. & Shapiro, S. 1998.The perils of excellence: Barriers to
Gavetti, G., & Levinthal, D. 2000. Looking forward and look-
effective process improvement in product-driven firms.
Production and Operations Quarterly, 7(1):2-18.
ing backward: Cognitive and experiential search. Ad-
ministrative Science Quarterly, 45: 113-137. Hendricks, K. B., & Singhal, V. R. 1996.Quality awards and

the market value of the firm:An empirical investigation. McGrath,R. 1999.A real options logic for initiating technol-
Management Science, 42: 415-436. ogy positioning investments. Academy of Management
Hill, R. 1993.When the going gets tough: A Baldrige Award Review, 22: 974-996.
winner on the line. The Executive, 7(3):75-79. Meyer,A., Brooks,G., &Goes, J. 1990.Environmentaljolts and
Honeywell. 1998.Annual report.Minneapolis: Honeywell Inc. industry revolutions. Strategic Management Journal, 11:
93-110.
Iansiti, M., & Clark, K. 1994.Integration and dynamic capa-
bility. Industrial and Corporate Change, 3: 557-606. Meyer, J. W., & Rowan, B. 1977. Institutionalized organiza-
tions: Formal structure as myth and ceremony. Ameri-
International Organization for Standardization (ISO). 1999. can Journal of Sociology, 83: 364-385.
ISO 9000 "one two three." http://www.iso.ch/isolen/
accessed December 12,2002.
iso9000-14000/tour/123.html, Miller, D. 1993. The architecture of simplicity. Academy of
Management Review, 18: 116-138.
Ishikawa, K. 1985.Whatis total quality control?TheJapanese
way. Englewood Cliffs, NJ:Prentice-Hall. Miller, D., & Friesen, P. H. 1980.Momentumand revolution in
Ittner, C. D., & Larcker,D. 1997.The performance effects of organizational adaptation. Academy of Management
Journal, 23: 591-614.
process management techniques. Management Science,
43: 522-534. Mitchell, W. 1989.Whether and when? Probability and tim-
Juran,J. 1989.Juranon leadership for quality. New York:Free ing of incumbents' entry into emerging industrial sub-
Press. fields. Administrative Science Quarterly, 34: 208-234.
Kearns, D., & Nadler, D. 1992.Prophets in the dark. New York: Morone, J. 1993.Winning in high tech markets. Boston: Har-
vard Business School Press.
Harper.
Keck, S., & Tushman, M. 1993.Environmental and organiza- Mukherjee,A., Lapre,M.,& Van Wassenhove, L. 1998.Knowl-
tional context and executive team structure.Academy of edge driven quality improvement. Management Sci-
Management Journal, 36: 1314-1344. ence, 44(Supplement):S35-S49.
Kolesar, P. J. 1993. Vision values milestones: Paul O'Neill Nadl'r, D. A., & Tushman, M. L. 1998.Competing by design.
starts total quality at Alcoa. California Management New York:Oxford University Press.
Review, 35(3):133-165. Nelson, R., & Winter, S. 1982.An evolutionary theory of eco-
Lawless, M. W., & Anderson, P. C. 1996.Generational tech- nomic change. Cambridge, MA: Harvard University
nological change: Effects of innovation and local rivalry Press.
on performance. Academy of Management Journal, 39: Nohria, N. 1996. From the M form to the N form. Working
1185-1217.
Paper, HarvardBusiness School, Boston.
Lawrence, P., & Lorsch, J. 1967. Organizations and environ- Nonaka, I., 1988. Toward middle-up-down management.
ments. Boston: HarvardBusiness School Press.
Sloan Management Review, 29(3):9-18.
Leonard-Barton,D. 1992. Core capabilities and core rigidi- Nonaka, I. 1991.The knowledge-creating company. Harvard
ties: A paradox in managing new product development. Business Review, 69(6):96-104.
Strategic Management Journal, 13: 111-125.
Nonaka, I., & Takeuchi, H. 1995. The knowledge-creating
Leonard-Barton,D. 1995. Wellsprings of knowledge. Boston:
HarvardBusiness School Press. company. New York:Oxford University Press.
Pisano, G. 1996.Learning before doing in the development of
Levinthal, D. 1991.Organizational adaptation and environ-
mental selection-interrelated processes of change. Or- new process technology. Research Policy, 25: 1097-1119.
ganization Science: 140-145. Powell, T. C. 1995.Total quality management as competitive
Levinthal, D. 1997a.Three faces of organizational learning: advantage: A review and empirical study. Strategic
Wisdom, inertia and discovery. In R. Garud, P. Nayyar, & Management Journal, 16: 15-37.
Z. Shapira (Eds.), Technological innovation: Oversights Quality Digest. 1999.Database of ISO 9000-certifiedcompa-
and foresights: 167-180.Cambridge: Cambridge Univer- nies. http://www.qualitydigest.com,accessed December
sity Press. 12, 2002.
Levinthal, D. 1997b.Adaptation on rugged landscapes. Man- Repenning, N. 1999. You measure what you get: Toward a
agement Science, 43: 934-950. theory of process improvement and change. Working
Levinthal, D., & March, J. G. 1993.The myopia of learning. paper, MIT,Cambridge, MA.
Strategic Management Journal, 14:95-112. Repenning, N. 2000.Drive out fear (unless you can drive it in):
Levitt, B., & March, J. G. 1988.Organizational learning. An- The role of agency and job security in process improve-
nual Review of Sociology, 14:319-340. ment. Management Science, 46: 1385-1396.
Louis, M., & Sutton, R. 1989.Switching cognitive gears: From Rotemberg,J.,& Saloner, G. 2000.Visionaries, managers and
habits of mind to active thinking. Human Relations, 44: strategic direction. Rand Journal of Economics, 31: 693-
55-75. 716.
March,J. 1991.Explorationand exploitation in organization- Samson, D., & Terziovski,M. 1999.The relationship between
al learning. Organization Science, 2: 71-87. total quality management practices and operational

performance. Journal of Operations Management, 17: and innovation: An introduction.Administrative Science

393-409. Quarterly, 35: 1-8.
Sanderson, S., & Uzumeri, M. 1995. Product platforms and Tushman, M. L., & O'Reilly, C. 1997.Winning through inno-
dominant designs: The case of Sony's Walkman. Re- vation. Boston: HarvardBusiness School Press.
search Policy, 24: 583-607. Tushman, M.,&Romanelli, E. 1985.Organizational evolution:
Sitkin, S. 1992. Learning through failure: The strategy of A metamorphosis model of convergence and reorienta-
small losses. Research in Organizational Behavior, 14: tion. Research in Organizational Behavior, 7: 171-222.
232-266.
Tushman, M. L., & Rosenkopf, L. 1992.Organizational deter-
Sitkin, S., & Stickel, D. 1996.The road to hell: The dynamics minants of technological change: Toward a sociology of
of distrust in an era of quality. In R. Kramer& T. Tyler technological evolution. Research in Organizational Be-
(Eds.),Trustin organizations: 196-215. London:Sage. havior, 14:311-347.
Sitkin, S. B., Sutcliffe, K. M., & Schroeder, R. G. 1994.Distin- Tushman, M., & Smith, W. 2002. Organizational technology.
guishing control from learning in total quality manage- In J. Baum (Ed.),Companion to organizations: 386-414.
ment: A contingency perspective. Academy of Manage- London:Blackwell.
ment Review, 19:537-564.
Tyre, M., & Orlikowski, W. 1994. Windows of opportunity:
Staw, B., & Epstein, L. 2000.What bandwagons bring: Effects Temporal patterns of technological adaptation in or-
of popular management techniques on corporate perfor- ganizations. Organization Science, 5: 98-109.
mance, reputation and CEOpay. AdministrativeScience
Utterback, J. 1994. Mastering the dynamics of innovation.
Quarterly, 45: 523-556. Boston: HarvardBusiness School Press.
Sterman, J. D., Repenning, N. P., & Kofman,F. 1997.Unantic-
Van de Ven, A., Angle, H., & Poole, M. 1989.Research in the
ipated side effects of successful quality programs: Ex-
management of innovation. New York:Harper & Row.
ploring a paradox of organizational improvement.Man-
agement Science, 43: 503-521. von Hippel, E. 1988. The sources of innovation. New York:
Oxford University Press.
Sull, D., Tedlow, R., & Rosenbloom, R. 1997.Manageriar com-
mitments and technology change in the U.S. tire indus- Weick, K. E. 1969.The social psychology of organizing. New
try. Industrial and Corporate Change, 6: 461-500. York:McGraw-Hill.
Sutcliffe, K.,Sitkin, S., & Browning, L. 2000.Tailoring process Weick, K. 1976.Educational organizations as loosely coupled
management to situational requirements. In R. Cole & systems. Administrative Science Quarterly, 21: 1-19.
W. Scott (Eds.), The quality movement & organization Weick, K. 1995. Sensemaking in organizations. Thousand
theory: 315-330. London:Sage. Oaks, CA: Sage.
Teece, D. J., Pisano, G., & Shuen, A. 1997.Dynamic capabili- Westphal, J. D., Gulati, R., & Shortell, S. M. 1997.Customiza-
ties and strategic management. Strategic Management tion or conformity? An institutional and network per-
Journal, 18:509-533.
spective on the content and consequences of TQMadop-
Tripsas, M. 1997. Surviving radical technological change tion. Administrative Science Quarterly, 42: 366-394.
through dynamic capability. Industrial and Corporate Williams, K., & O'Reilly, C. 1998.Demography and diversity
Change, 6: 341-377. in organizations: A review of forty years of research.
Tripsas, M., & Gavetti, G. 2000. Capabilities, cognition, and Research in Organizational Behavior, 20: 77-140.
inertia: Evidence from Digital Imaging. Strategic Man- Winter, S. 1994. Organizing for continuous improvement:
agement Journal, 21: 1147-1161. Evolutionary theory meets the quality revolution. In
Tushman, M. L., & Anderson, P. 1986.Technological discon- J.Baum &J.Singh (Eds.),Evolutionarydynamics of organ-
tinuities and organizational environments. Administra- izations: 90-108. New York:Oxford University Press.
tive Science Quarterly, 31: 439-465. Wruck, K., & Jensen, M. 1994. Science, specific knowledge,
Tushman, M. L., & Murmann, J. P. 1998. Dominant designs, and total quality management. Journal of Accounting
technology cycles, and organizational outcomes. Re- and Economics, 18:247-287.
search in Organizational Behavior, 20: 213-266.
Zbaracki,M.J.1998.The rhetoricand realityof total qualityman-
Tushman, M., & Nelson, R. 1990.Technology, organizations agement.AdministrativeScience Quarterly,43:602-636.
MaryJ.Benner is an assistant professor in management at The WhartonSchool of the

University of Pennsylvania. She received her Ph.D. from the Graduate School of
Business, Columbia University. Her research focuses on technological innovation,
organizational adaptation to technological change, and the effects of process man-
agement on innovation and adaptation.
Michael L. Tushman is the Paul R. Lawrence MBSClass of 1942Professor of Business
Administration at Harvard Business School. He received his Ph.D. at MITand was
previously on the faculty of the Graduate School of Business, Columbia University. His
research explores the relations among technological change, executive leadership,
and organizational adaptation.


Benner-2003-Exploitation, Exploration, and Process Management-The Productivity Dilemma Revisited. The Academy of Management Review

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Benner-2003-Exploitation, Exploration, and Process Management-The Productivity Dilemma Revisited. The Academy of Management Review

Uploaded by

Copyright:

Available Formats

Exploitation, Exploration, and Process Management: The Productivity Dilemma Revisited

Author(s): Mary J. Benner and Michael L. Tushman

This content downloaded on Sun, 6 Jan 2013 05:52:15 AM

We develop a contingency view of process management's influence on both techno-

This content downloaded on Sun, 6 Jan 2013 05:52:15 AM

This content downloaded on Sun, 6 Jan 2013 05:52:15 AM

This content downloaded on Sun, 6 Jan 2013 05:52:15 AM

This content downloaded on Sun, 6 Jan 2013 05:52:15 AM

AND We classify innovations along two dimen-

This content downloaded on Sun, 6 Jan 2013 05:52:15 AM

This content downloaded on Sun, 6 Jan 2013 05:52:15 AM

This content downloaded on Sun, 6 Jan 2013 05:52:15 AM

This content downloaded on Sun, 6 Jan 2013 05:52:15 AM

This content downloaded on Sun, 6 Jan 2013 05:52:15 AM

delivery of technological innovations, innova- units. In these loosely coupled organizational

This content downloaded on Sun, 6 Jan 2013 05:52:15 AM

This content downloaded on Sun, 6 Jan 2013 05:52:15 AM

process innovation and incremental refine- Proposition 7: Increases in process

This content downloaded on Sun, 6 Jan 2013 05:52:15 AM

ket uncertainty. Responding to environmental Case studies of process management efforts

This content downloaded on Sun, 6 Jan 2013 05:52:15 AM

This content downloaded on Sun, 6 Jan 2013 05:52:15 AM

This content downloaded on Sun, 6 Jan 2013 05:52:15 AM

This content downloaded on Sun, 6 Jan 2013 05:52:15 AM

This content downloaded on Sun, 6 Jan 2013 05:52:15 AM

This content downloaded on Sun, 6 Jan 2013 05:52:15 AM

performance. Journal of Operations Management, 17: and innovation: An introduction.Administrative Science

MaryJ.Benner is an assistant professor in management at The WhartonSchool of the

This content downloaded on Sun, 6 Jan 2013 05:52:15 AM

You might also like