Achieving contextual ambidexterity in R&D organizations: a management control system approach

Achieving contextual
ambidexterity in R&D
organizations: a management
control system approach
Ian P. McCarthy and Brian R. Gordon
Beedie School of Business, Simon Fraser University, 500 Granville Street, Vancouver, BC, Canada
V6C 1W6. ian_mccarthy@sfu.ca; brg@sfu.ca

Research on how managers control R&D activities has tended to focus on the performance
measurement systems used to exploit existing knowledge and capabilities. This focus has been
at the expense of how broader forms of management control could be used to enable R&D
contextual ambidexterity, the capacity to attain appropriate levels of exploitation and
exploration behaviors in the same R&D organizational unit. In this paper, we develop a
conceptual framework for understanding how different types of control system, guided by
different R&D strategic goals, can be used to induce and balance both exploitation and
exploration. We illustrate the elements of this framework and their relations using data from
biotechnology firms, and then discuss how the framework provides a basis to empirically
examine a number of important control relationships and phenomena.

1. Introduction A feature of research on R&D control is that
it has largely focused on the design and impact

C ontrolling R&D is a challenge. Managers
have long struggled to develop effective
control systems for directing and adjusting
of performance measurement systems, which are
only one type of control system – the diagnostic
control system (Otley, 1980; Simons, 1994). As
R&D behaviors and outcomes. Consequently, diagnostic systems are used to evaluate and reward
researchers have been motivated to examine organizational activities, they tend to induce
R&D control from three complementary levels relatively measurable exploitation behaviors that
of analysis: the firm, the market and the inno- ensure current viability, at the expense of the more
vation system (Chiesa and Frattini, 2009). This intangible exploration behaviors needed for ensur-
has resulted in studies that explore how R&D ing future survival. This bias is counter to the
activities and outputs should be measured (e.g., notion that sustained organizational performance
Souder, 1972; Schumann et al., 1995; Werner requires an organization to effectively balance
and Souder, 1997; Kerssens-van Drongelen and exploitation with exploration (March, 1991), a
Bilderbeek, 1999; Chiesa and Frattini, 2007); how capability known as ‘organizational ambidexterity’
R&D organizations should be designed and man- (Duncan, 1976; Tushman and O’Reilly, 1996).
aged (e.g., Tymon and Lovelace, 1986; Whitting- Although sustained organizational performance
ton, 1991); and how markets and governments is associated with a firm’s ability to be ambidex-
can support R&D (e.g., Moravesik, 1973; Martin trous, it is a capability that is conceptually ambig-
and Irvine, 1983). uous and difficult to achieve. On the one hand,

240 R&D Management 41, 3, 2011. r 2011 The Authors. R&D Management r 2011 Blackwell Publishing Ltd,
9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA.

Achieving contextual ambidexterity in R&D organizations

ambidexterity is viewed as the attainment of a difficult to achieve because the organizational units
balance between exploitation and exploration, are disconnected. A second complementary ap-
whereby ‘organizations make explicit and implicit proach is ‘contextual ambidexterity’ (Birkinshaw
choices between the two’ (March, 1991, p. 71) to and Gibson, 2004; Gibson and Birkinshaw, 2004;
attain an ‘optimal mix’ (March, 1991, p. 75). On Raisch and Birkinshaw, 2008). It involves creating
the other hand, exploitation and exploration are an organizational context – the organizational
considered to be mutually enhancing, so that it is stimuli that inspire, guide and reward people to
possible for firms to attain high levels of both act in a certain way (Ghoshal and Bartlett, 1997) –
(Gupta et al., 2006; Jansen et al., 2006). In our that will allow exploitation and exploration beha-
paper we follow the ‘balanced’ view of ambidexter- viors to transpire in the same organizational unit.
ity, and assume that R&D managers face decisions We argue that contextual ambidexterity is im-
about allocating resources and attention to activ- portant and suited to R&D organizations for at
ities that can be relatively explorative or relatively least two major reasons. First, the problems of
exploitative in nature. attaining ambidexterity by structural separation
Although the classic definition of ambidexterity are compounded for R&D organizations. R&D
provided by March (1991) would seem to suggest activities are often already structurally separated
that R&D is exploration, the dilemma of balancing and operationally distinct from other organiza-
exploitation and exploration clearly exists in R&D tional activities such as legal, manufacturing or
organizations. Ahuja and Lampert (2001), for sales. Thus, any further partitioning (i.e., separat-
example, explain how R&D activities vary across ing the ‘R’ from the ‘D’) increases the problem of
the exploitation-exploration continuum. They de- integrating and utilizing R&D outputs through-
fine R&D exploration as activities and outputs out the organization. This is especially so for
that focus on novel, emerging and pioneering small- to medium-sized organizations, whose
technologies; they define R&D exploitation as R&D activities are tightly intertwined. Second,
activities and outputs concerned with mature, we suggest that contextual ambidexterity is suited
familiar and propinquituous technologies. Simi- to the ‘clan control’ typically found in R&D
larly, McNamara and Baden-Fuller (1999, 2007) organizations (Ouchi, 1979), as it involves using
argue that ambidexterity is fundamentally about ‘processes or systems that enable and encourage
different forms of learning, and that R&D orga- individuals to make their own judgments about
nizations, in common with other types of organi- how to divide their time between conflicting
zation, must maintain a balance of short-term demands for alignment and adaptability’ (Gibson
exploitation and long-term exploration to be suc- and Birkinshaw, 2004, p. 211).
cessful over time. These characterizations of R&D In this paper we focus on the problem of how
exploration and R&D exploitation follow the view to attain contextual ambidexterity in a single
that ambidexterity is the capability to balance R&D organizational unit. We present and illus-
different types of knowledge production (Levinthal trate a conceptual framework that shows how
and March, 1993). broader forms of control system, guided by R&D
While organizational ambidexterity is a relatively goals, could be used to encourage teams and
straightforward concept to understand, it is not an individuals in R&D organizations to simulta-
easy capability to attain. Exploration and exploita- neously pursue both exploitation and exploration.
tion have fundamentally different qualities. Exploi- We present our arguments in four major sections.
tation is characterized by short-term time horizons, First, we review the R&D management control
efficiency, reliability and refinement, while explora- literature, highlighting the need to move beyond
tion involves long-term time horizons, search, ex- exploitation and metric-focused performance
perimentation, innovation and adaptability. To measurement. We identify the importance of
simultaneously induce and balance these differ- linking the design and use of control systems
ences, there are two recognized approaches. One to the R&D goals of the organization. Second,
is ‘structural ambidexterity’ (Tushman and O’Re- we develop our conceptual framework by synthe-
illy, 1996), which involves splitting exploitation and sizing R&D control concepts with control the-
exploration into different organizational units (i.e., ories developed in the fields of accounting and
separate divisions, departments or teams). It is then strategic management. Specifically, we adapt
the task of senior managers to ensure that the Simons’ (1994) ‘levers of control’ framework,
respective exploitation and exploration outcomes which consists of four types of control system:
of each organizational unit are integrated to create beliefs systems, boundary systems, diagnostic
value. This integration task, however, can also be systems and interactive systems. We explain how

r 2011 The Authors R&D Management 41, 3, 2011 241
R&D Management r 2011 Blackwell Publishing Ltd


these four types of control system, guided by to the values, purpose and direction of the organi-
R&D strategic goals, can work together to de- zation; (ii) boundary systems that limit strategically
velop and harness both exploitation and explora- undesirable activities and outcomes; (iii) diagnostic
tion in an individual R&D organization. Third, systems that measure activities to ensure they are in
although studies have recognized that beliefs, accordance with organizational objectives; and (iv)
boundary, diagnostic and interactive systems interactive systems that scan for and communicate
work together to create different behaviors and strategic information to employees so as to adjust
outcomes (e.g., Widener, 2007; Chiesa et al., the direction of the organization. For each study
2009a), significant ambiguity remains in the lit- we also list the type of analysis undertaken, and the
erature regarding what these systems actually are. contribution made.
That is, what actual rules, policies, procedures, Looking across these studies, we identify four
processes, technologies and incentives might themes that characterize much of the existing
R&D managers use to create the control asso- research in the area, and provide the motivation
ciated with each type of control system? In for the conceptual framework that we develop.
response, we presented our framework to man- First, existing studies have predominantly focused
agers and scientists employed by small- and on how performance measurement systems (i.e.,
medium-sized biotechnology firms. This was not diagnostic control systems) promote the efficiency
done to inductively derive the framework, nor to of behaviors central to R&D exploitation.
provide strong empirical support for it. Rather, Although there are some studies that examine
we sought examples to help describe and illustrate broader forms of control, including the effects of
the framework, to provide some preliminary face process formality (Bart, 1993), project structure
validity for our arguments, and to exemplify what (Cooper and Kleinschmidt, 1986), professional
these control systems actually are. Fourth, we rituals (Whittington, 1991) and goal setting activ-
discuss some general implications of our concep- ities (Kerssens-van Drongelen and Bilderbeek,
tual framework for R&D management, each of 1999; Yawson et al., 2006), there is only one study
which points to future areas of research. We we know of that has used Simons’ (1994) control
suggest that our framework provides a basis for framework in an R&D context (see: Chiesa et al.,
empirically studying the extent to which multiple 2009a). Furthermore, although Table 1 only lists
R&D strategic goals drive the use of different empirical studies (so as to focus and limit our
types of control system. We also contend that our review to established R&D control concepts), a
framework provides a starting point from which wider reading of the R&D management control
to examine how the use and attention of different literature reveals that diagnostic control has so far
management control systems can be altered over dominated prior work on R&D control frame-
time so that R&D managers can ‘dynamically’ works (e.g., Chiesa and Masella, 1996; Bremser
manage the exploitation–exploration balance. and Barsky, 2004), taxonomies (e.g., Tymon and
Lovelace, 1986) and reviews of R&D measures
(e.g., Werner and Souder, 1997; Geisler, 2002;
2. R&D management control: moving Garcı´ a-Valderrama and Mulero-Mendigorri, 2005).
beyond performance measurement Consequently, the motivation for our paper follows
the view that although ‘measuring performance is
When Freeman (1969, p. 11) argued ‘if we cannot helpful, it is only part of the story’ (Chiesa et al.,
measure all of the information generated by R&D 1996, p. 105). In particular, we argue that different
activities because of a variety of practical diffi- types of control system, guided by R&D strategic
culties, this does not mean that it may not be goals, can work together to balance different levels
useful to measure part of it,’ he spurred a genera- of exploitation and exploration in individual R&D
tion of scholars to understand what constitutes organizations.
effective R&D management control, in both in- Second, research on the diagnostic control of
dustrial and government settings. Table 1 presents R&D has traditionally focused on the perfor-
a selection of control system studies published in mance measures as opposed to the systems (i.e.,
leading R&D management and innovation jour- the rules, procedures and technologies) that
nals. For each study, the table lists the type of managers might use to direct and adjust R&D
control system examined, according to the control behaviors. This measure-based approach treats
systems in Simons’ (1994, 1995a) framework. R&D organizations as ‘black boxes,’ ignoring
These are (i) beliefs systems that are used to their inner workings and the relationships be-
inspire employees to engage in activities central tween goals, controls, behaviors and outcomes.

242 R&D Management 41, 3, 2011 r 2011 The Authors


Table 1. Review of empirical studies within the R&D management control literature
Article Control Empirical analysis Findings
systems
examined

Schainblatt Diagnostic Literature and company survey that Systems should differ according to
(1982) compared the use of R&D research activities and development
productivity measurement systems activities, and overall R&D goals
Martin and Diagnostic Measures for allocating funds to A system of indicators for assessing
Irvine (1983) scientific research institutes scientific research progress
Cooper and Diagnostic; How variations in the structure and A model that explains how control
Kleinschmidt boundary stages of the new product efficiency and control reliability
(1986) development process influence influence project outcomes
innovation outcomes
Cordero Diagnostic A study the links between firm level A model that combines technical and
(1990) R&D investments, productivity and commercial performance and
reward allocation specifies how measures vary
according to organizational levels
and process stages
Whittington Boundary A study of in-house and independent R&D organizations with market
(1991) R&D organizations and three types control are more productive than
of structural control: market, those with hierarchical and
hierarchical and professional professional control
Bart (1993) Diagnostic; Interviews with R&D managers in The importance of balancing formal
boundary; large companies on the tightness or and informal controls, in line with
interactive formality of their control systems R&D goals
Hauser and Diagnostic Interviews with CEOs, CTOs and Effective measures depend upon the
Zettelmeyer researchers at ten research-intensive goals and research intensity of the
(1997) organizations on the use of R&D and engineering activity
performance measures
McGrath and Diagnostic Determining R&D effectiveness in Firms with a high R&D effectiveness
Romeri (1994) electronics companies by assessing index are more productive, reliable
investment versus new product and innovative
performance
Kerssens-van Diagnostic Literature review, company survey Outlines the importance of
Drongelen and and in-depth interviews to assess use contingency factors and specifies
Cook (1997) of R&D measures and system design control system requirements and
principles design parameters
Werner and Diagnostic Survey to understand measurement Control system design is dependent
Souder (1997) philosophy and perceived usefulness on control aims, type of R&D
of measurement activity, data availability and cost
Kerssens-van Diagnostic; Survey of performance measurement Explores the importance of
Drongelen and beliefs practices and effectiveness contingency factors and highlights
Bilderbeek the importance of feedback and feed-
(1999) forward control
Godener and Diagnostic Use and impact of performance Using performance results will
Soderquist
¨ measurement on decision-making improve R&D relevance and
(2004) and operations coherence, decision-making and
employee motivation
Karlsson et al. Diagnostic Case study that examines product Systems should be designed to suit
(2004) and process development type of R&D activity, control needs
and strategic goals
Yawson et al. Diagnostic; Case study that examines balanced- Systems can align measurement with
(2006) beliefs score card use in a research institute strategic objectives and address
capability and utilization issues
Chiesa et al. Diagnostic; Case studies that examine how these Beliefs and interactive systems are
(2009a) beliefs; systems are employed in different more prominent in the early stages of
interactive; phases of the radical innovation the process, while diagnostic systems
boundary process are more prominent in the later stages
of the process

By focusing on the different types of control provide examples of these systems. This follows
system used, and collecting data from managers other studies of R&D control that focused on the
and scientists in biotechnology firms, we aim to actual systems used (e.g., Szakonyi, 1995; Chiesa



et al., 1996) and emphasized that control is about ment control ideas and advances from the fields
more than choosing a set of metrics (Kerssens-van of accounting and strategic management, and
Drongelen and Cook, 1997; Kerssens-van Dron- apply these to the domain of R&D control. In
gelen and Bilderbeek, 1999). particular, we explain how the use of the four
Third, by definition, research on the diagnostic types of control system proposed by Simons
control of R&D tends to highlight what we call a (1994), guided by different R&D strategic goals,
specific ‘control orientation.’ This is the extent to can be used to induce and balance the exploita-
which individuals and teams conceive and under- tion and exploration behaviors, and the feedback
take control in an ex post (after-the-event) or ex and feed-forward control orientations necessary
ante (before-the-event) manner. Prior research on for attaining contextual ambidexterity.
R&D control has tended to focus on the ‘feed-
back control orientation,’ which is when after-
the-event information (e.g., errors, failures and
other unsatisfactory organizational outcomes) is
3.1. Simons’ levers of control
used to direct and adjust organizational beha- In the fields of accounting and strategic manage-
viors. This feedback control orientation is central ment, researchers have argued that management
to exploitation as it promotes single-loop learning control involves using a number of different, but
and the continuous refinement of organizational inter-related types of system (Ouchi, 1979; Otley,
practices and capabilities (Argyris and Schon, ¨ 1980; Eisenhardt, 1985; Marginson, 2002; Turner
1978; Kerssens-van Drongelen and Bilderbeek, and Makhija, 2006). To explain how control
1999). In contrast, a ‘feed-forward control orien- systems can vary and function, Simons (1994)
tation’ involves seeking and receiving before-the- proposed an influential framework of manage-
event information about future trends, events and ment control built around what he termed the
their effects (e.g., changes in regulations, compe- ‘four levers of control’ – beliefs systems, bound-
tition and demand). This information is used to ary systems, diagnostic systems and interactive
adjust organizational behaviors so as to prevent systems. Collectively these four types of control
unacceptable outcomes from occurring. It is a system represent the policies, procedures and
control orientation that energizes the exploration technologies that influence the cultural norms,
and double-loop learning needed for individuals behaviors and outcomes of individuals and
and organizations to radically rethink and alter groups. Each type of control system has unique
their existing capabilities (Argyris and Schon, ¨ effects but, importantly, they also work in con-
1978; Kerssens-van Drongelen and Bilderbeek, junction with one another to manage ‘the inherent
1999). Using our conceptual framework, we ex- tensions between (1) unlimited opportunity and
plain how different types of control system work limited attention, (2) intended and emergent
together to generate both feedback and feed- strategy and (3) self-interest and the desire to
forward control orientations, which together pro- contribute’ (Simons, 1995a, p. 28). We place
vide informational stimuli to induce the behaviors Simons’ (1994) four types of control system as
necessary for contextual ambidexterity. the central element in our conceptual framework
Fourth, it is clear from the studies listed in (see Figure 1), and argue that R&D managers,
Table 1 that the effectiveness of R&D control is guided by R&D strategic objectives, can use
contingent on a number factors, one of the most Simons’ control systems to shape the organiza-
prominent of which is the R&D goals of the tional conditions necessary for contextual ambi-
organization (see: Schainblatt, 1982; Bart, 1993; dexterity. We now discuss each type of control
Chiesa et al., 2009b). Thus, in the next section of system in more detail.
our paper we describe how four R&D goals – Beliefs systems are ‘the explicit set of organiza-
growth, innovation, reliability and efficiency – tional definitions that senior managers commu-
relate to and drive the use of control systems nicate formally and reinforce systematically to
and the attainment of R&D ambidexterity. provide basic values, purpose and direction for
the organization’ (Simons, 1995a, p. 34). They
help ensure that the attitudes and behaviors of
3. A management control framework for individuals are aligned with the R&D strategic
R&D contextual ambidexterity goals and the scientific principles that underpin
the R&D organization. They provide the ‘positive
In this section of our paper we develop our energy’ necessary for exploration (Simons,
conceptual framework. We synthesize manage- 1995a), and the strategic coherence necessary for



Figure 1. Relationships between R&D strategic goals, control system type and R&D contextual ambidexterity.

R&D employees to search for new knowledge and the systems that define and enforce the limits
opportunities in an autonomous, but focused beyond which employees must not stray. This
manner. demarcation role of boundary systems helps pre-
Interactive systems enable ‘top-level managers vent R&D organizations from over-exploring and
to focus on strategic uncertainties, to learn about becoming too stretched. Thus, boundary systems
threats and opportunities as competitive condi- are central to reliability-based exploitation beha-
tions change, and to respond proactively’ (Si- viors, in that they ‘transform unbounded oppor-
mons, 1995b, p. 81). These systems are used by tunity space into a focused domain that
R&D managers to scan, ‘explore’ and acquire organizational participants can be encouraged
information about events and trends in their to exploit’ (Simons, 1995a, p. 41).
organization’s external environment (Daft and Diagnostic systems are the ‘feedback systems
Weick, 1984). They also include the communica- used to monitor organizational outcomes and
tion processes that R&D managers use for insti- correct deviations from preset standards of per-
gating debate with colleagues about the future of formance’ (Simons, 1994, p. 170). If non-confor-
the organization. Intra-organizational networks mance is identified this can prompt changes in
(Swan et al., 1999), information technology organizational activities and in the other types of
(Alavi and Leidner, 2001) and group based pro- control system, to adjust what is done and how it
cesses such as brainstorming and sand-pit events is done. As highlighted by our review of the R&D
(Cummings, 2004), are all example of interactive control literature, diagnostic systems tend to
systems that managers can use to engage in focus on measuring tangible and exploitation
exploration and knowledge sharing. activities, which in turn motivate R&D employees
Boundary systems delineate ‘the acceptable to be productive and efficient. Consequently,
domain of strategic activity for organizational strong diagnostic systems, if used in conjunction
participants’ (Simons, 1995a, p. 39). These are with weak or inappropriate boundary and beliefs



systems, can promote the ‘what you measure, is consistent with Simons’ (1995a) claim that man-
what you get’ phenomenon, which can sometimes agers will use beliefs systems to inspire employees
lead to unintended and undesirable consequences. to overcome organizational inertia and grow (i.e.,
build), or alternatively to focus, be persistent and
complete existing projects (i.e., harvest). Conse-
quently, we suggest that the nature and use of
3.2. R&D strategic goals beliefs system in an R&D organization will be
One of the earliest definitions of a management linked to its growth goal.
control system describes it as a collection of Innovation goals define the kinds of outcomes
systems that managers use to ‘ensure that re- that R&D organizations seek to produce. At the
sources are obtained and used effectively and most general level, this involves specifying the
efficiently in the accomplishment of the organisa- velocity, magnitude and application range of
tion’s objectives’ (Anthony, 1965, p. 17). Simi- technological change (McCarthy et al., 2010). In
larly, R&D management scholars have argued an effort to be parsimonious, we focus only on the
that for an R&D management control system to magnitude of change in a technology’s capability,
be effective it should be aligned with the R&D and the degree of benefit it brings to the market
strategic goals of the organization (Schainblatt, (Wheelwright and Clark, 1992; Maine and Garn-
1982; Bart, 1993; Chiesa et al., 2009b). sey, 2006). As recently argued by Chiesa et al.
We define R&D strategic goals as statements (2009a, p. 419), such innovation differences sig-
that motivate R&D organizations to attain a level nificantly influence ‘the adoption of specific man-
of proficiency in a specific R&D capability. In agerial approaches, organizational solutions and
terms of what these goals might be, we follow operative instruments,’ i.e., the design and use of
studies that emphasize the importance of R&D appropriate control systems. Furthermore, it is
goals to control system design and use (Schain- argued, as interactive control systems promote
blatt, 1982; Bart, 1993; Karlsson et al., 2004; exploration and learning they are more prominent
Chiesa et al., 2008). Together these studies sug- in the early stages of the radical innovation
gest that R&D goals can vary in terms of how process (Chiesa et al., 2009a). On this basis, we
they specify different types of research activity suggest that R&D organizations in pursuit of
(i.e., basic research, applied research, and devel- radical innovation goals would benefit from
opment), innovation activity and outputs (i.e., greater use of interactive control systems; con-
incremental and radical), so as to support the versely, R&D organizations in pursuit of more
overriding business strategy and rules of competi- incremental innovation goals would require less
tion governing the R&D organization. The four use of interactive control systems.
R&D strategic goals we use – growth, innovation, Reliability goals indicate the extent to which
reliability and efficiency – are distilled from these the activities of R&D organizations should be
principles, and from the notion that each of the trustworthy and dependable (Kiella and Golhar,
control systems proposed by Simons (1994, 1997). This type of goal has become important,
1995a) is individually oriented toward one of with R&D organizations increasingly undertak-
these strategic goals. This is indicated by the ing quality improvement programs. Reliability
goal-control linkages in Figure 1: growth-beliefs, goals determine the extent to which an R&D
innovation-interactive, reliability-boundary and organization should be proficient at reaching
efficiency-diagnostic. These relationships are not project milestones on time (Cooke-Davies and
exclusive, however, and this is indicated by the Arzymanow, 2003), and the propensity of the
single arrowed line that broadly connects all of organization to engage in post-project assess-
the R&D strategic goals with all of the control ments that promote organizational learning (von
system types. Zedtwitz, 2002). As these activities involve adher-
We define R&D growth goals as the extent to ing to acceptable R&D domains and practices, we
which an R&D organization seeks to increase or suggest that the degree to which the R&D orga-
maintain its organizational size (e.g., number of nization desires high levels of reliability, will
employees, R&D capacity and R&D outputs). determine the extent to which boundary systems
They indicate the degree to which an R&D are used to ‘establish explicit limits and rules
organization is concerned with developing its which must be respected’ (Simons, 1994, p. 170).
innovative capacity by increasing its project port- Efficiency goals define the extent to which an
folios, the number of R&D employees and other R&D organization focuses on using its resources
related resources (Addison et al., 1976). This is to maximize the production of innovations. As



discussed in our review of the R&D control 4. Framework illustration: the case of
literature, this particular goal has captured the biotechnology firms
attention of managers and scholars who have
focused on performance measurement systems In this section, we present data to exemplify the
and exploitation-related criteria such as R&D elements of our framework in terms of what they
productivity. Consequently, the attainment of this are (i.e., the actual goals, systems and behaviors),
goal is closely linked to diagnostic control systems and what they do (i.e., their effect on other
that establish targets, and measure activities and elements of the framework and on an R&D
outcomes to help ensure that the other R&D organization). These data are used to illustrate
strategic goals are being achieved efficiently. the framework, and make it more connected to
R&D reality. Similar approaches have been used
to illustrate conceptual frameworks dealing with
R&D performance measurement systems (Chiesa
3.3. R&D contextual ambidexterity et al., 2008), user innovation (Berthon et al., 2007)
We define R&D contextual ambidexterity as the and external technology commercialization (Bian-
ability to attain appropriate levels of exploitation chi et al., 2009).
and exploration behaviors in the same R&D
organizational unit. The right-hand element of
our framework indicates how the four control
4.1. Setting and methodology
systems combine to produce the behaviors and
control orientations necessary for this ability. We We focused on biotechnology firms, defined
suggest that beliefs systems and interactive sys- broadly as those firms that undertake life-science
tems jointly produce exploration and a feed-for- research to develop therapeutic products, medical
ward control orientation. Beliefs systems provide devices or biotechnology related services. Bio-
‘momentum and guidance for opportunity-seek- technology firms are particularly appropriate for
ing behaviors,’ and interactive systems ‘focus illustrating our framework, for several reasons.
organizational attention on strategic uncertainties First, they are often considered to be a proto-
and thereby provoke the emergence of initiatives typical example of an R&D organization. Second,
and strategies’ (Simons, 1994, p. 172). Together even though researchers have argued that the
these two types of control system promote pro- long-term success of these firms depends on
specting, experimentation and sense-making; all continued exploration (e.g., discovery, product
of these are not only central to exploration, but formulation and preclinical trials), and effective
also promote a feed-forward control orientation exploitation (e.g., clinical trials and the new drug
for anticipating future events and their effects. application stage) (McNamara and Baden-Fuller,
Thus, beliefs systems and interactive systems 1999, 2007), we know relatively little about how
underlie the proactive scanning and planning this R&D ambidexterity can be attained. Third,
behaviors essential for determining when and the exploitation and exploration activities of these
how R&D activities should be modified. firms are significantly intertwined with each other.
Our framework also suggests that the exploitation This means that biotechnology firms are suited to
aspect of R&D contextual ambidexterity is linked to contextual ambidexterity, because it is proble-
the joint use and effects of diagnostic and boundary matic to structurally separate these intertwined
systems. Boundary systems permit discovery and activities. This is especially the case for small- and
learning, but within clearly defined limits of free- medium-sized biotechnology firms as their orga-
dom. Diagnostic systems monitor R&D activities nizational size limits any major and viable separa-
and outputs, and use this after-the-event information of resources. Lastly, by focusing only on
tion to reward conformance, or to modify processes biotechnology firms, our data are bounded, help-
and systems to correct non-conformance. Diagnostic ing to provide a focused illustration of the ele-
systems measure activities and outputs so that R&D ments of our framework.
managers know when things are going well, or are To collect the data, we took advantage of a
going wrong. This creates a context for making biotechnology management education program,
informed decisions about resource allocation and led by one of the authors of this paper. The
process redesign. Thus, together diagnostic and learning nature of this university-industry pro-
boundary systems induce exploitation as employees gram was useful for our research, as it provided
are directed and rewarded to refine and apply respondents with the opportunity and environ-
existing knowledge and competences. ment to reflect, analyze and discuss the control



Table 2. Companies and Respondents
Firm* Area of biotechnology Age of No. of Role of respondents
firm (years) employees

Firm A Research testing services 15 35 Chief Executive Officer,
Project Manager
Firm B Research testing services 29 175 Analytical Chemist
Firm C Research testing services 40 51 Chief Executive Officer
Firm D Research modeling services 10 38 Senior Technologist
Firm E Research modeling services 5 48 Research Scientist –
microbiology
Firm F Research modeling and testing services 18 62 Account Manager
Firm G Research modeling and testing services 7 32 Project Manager
Firm H Research modeling and testing services 23 134 Account Manager
Firm I Discovery and development of medical 17 284 Business Development
devices Manager
Firm J Discovery and development of therapeutic 28 120 Research Scientist –
drugs toxicology
Firm K Discovery and development of therapeutic 6 50 Quality Assurance Manager,
drugs Senior Chemist
Firm L Discovery and development of therapeutic 8 70 Senior Research Associate
drugs
Firm M Discovery and development of therapeutic 17 201 R&D Technologist, Project
drugs Manager
Firm N Discovery and development of therapeutic 14 59 Senior Research Associate
drugs
Firm O Discovery and development of therapeutic 14 67 Research Scientist – oncology
drugs
*Pseudonyms and basic descriptions of the area of biotechnology areas are used to protect the anonymity of firms and respondents.

systems within their firms. From 2006 to 2008 we ment of the framework was defined and explained.
collected data from over 40 senior managers and This was followed by a discussion to further clarify
scientists, from 15 different biotechnology firms the function and scope of each element of the
whose organizational size ranged from 35 to 284 framework. Next, we collected data from indivi-
employees (see Table 2). With this number of dual respondents using a semi-structured inter-
firms, we did not seek to develop rich case studies view. Respondents were first asked to confirm
for inductive theory building, or to provide strong that their firms had an active R&D capability,
empirical support for our framework. Instead, we and to provide the following background informa-
sought multiple sources of data to help describe tion about their firms: age, size in terms of
and tentatively validate the elements and logic of employee numbers and the area of biotechnology
our framework. the firm focused on. The respondents were then
All of the firms were located in Western Canada, asked to comment on the validity of the logic of
and undertook biotechnology related R&D. Ap- the framework in general terms, and to consider
proximately half of the sample, Firms A–H, were the extent to which their firms focus on and use the
research service organizations that undertook R&D different types of control system. This latter point
activities to develop their portfolio of testing and required the respondents to reflect on the number
modeling services for drug development and health- of people, rules and processes associated with
care organizations. The other half of our sample, each type of system. Next, the respondents were
Firms I–O, were involved primarily in the discovery asked to give examples of how each element of
and development of drugs or medical devices. All 15 the framework exists, and to exemplify links be-
firms were at least three years old and employed tween the different elements of the framework
more than 10 people, ensuring that they were likely (see Table 3).1 The aim was to elicit actual exam-
to employ some form of formal management con- ples of the goals, the control systems and the
trol system (Davila and Foster, 2007). associated behaviors and control orientations.
Given our illustrative aims, the data collection The final stage of data collection involved a
began by presenting our conceptual framework number of follow-up interviews, where respon-
(Figure 1) to groups of between five and 10 dents were contacted to either seek further infor-
respondents. During these presentations, each ele- mation or to clarify aspects of their answers.



Table 3. Illustration of Model Elements
Model elements Representative and abbreviated answers*

R&D strategic goals
Define what the goals Growth: We are trying to survive and grow; and so we are totally focused on securing
mean to your firm? the next round of investment. We need the money to keep existing people, to recruit new
people and buy more equipment (Firm K). When we started out we hoped to build a
company big enough to develop multiple drugs on its own, instead, we explored
alliances and are finalizing a deal to license the technology (Firm L). Unlike most
biotech start-ups, we are a service business that must produce a return on investment
now (Firm B)
Innovation: We are in the process of getting new approvals for our products, both in
terms of serving new regional markets and in terms of new applications for our existing
products. We are trying to better target the drugs we have (Firm I). We have this
technology, which if it works out, will completely transform the value chain for this
testing service (Firm C)
Reliability: In this industry we succeed and fail on the quality of our data. If we are not
scientifically valid at every stage of the process; then our products will not get approved
(Firm O). If we are not reliable we will soon be out of business (Firm G)
Efficiency: This is a major driver for us. Our testing services are up against existing rival
services that are continuously improving. If our R&D does not deliver desirable
enhancements to the tests and models we offer, then in the long run we won’t be able to
compete (Firm D). While return on investment is important, we only worry about
efficiency when the money starts to run out (Firm J)
Describe how each type of Growth and beliefs systems: Originally we were a local company servicing local biotech
goal relates to each type firms; now we are trying to access global markets and work with partners around the
of control system? world, and this is clearly reflected in our mission statement and core values. (Firm F).
We have grown so we have different types of stakeholders whose interests are
communicated to us (Firm H)
Innovation and interactive systems: A few years back the CEO started organizing
company retreats, where we think about how the organization is evolving – what it
might become (Firms J). From an R&D basis we deliver monthly presentations to the
chief technology officer so that he knows what we are coming up with (Firm E)
Reliability and boundary systems: The trustworthiness of our research is so important
that we strongly adhere to good laboratory practice (GLP) (Firm A). Archiving and
peer review are common systems in our industry for facilitating boundary-like control
in labs (Firm M)
Efficiency and diagnostic systems: Our financial officer monitors and measures our cash
burn rate (Firm E). Our project managers monitor progress each week in terms of the
number of tests, and the quantity of data produced and recorded (Firm K)
Management control systems
Provide examples of the Beliefs: Our mission statement is everywhere – it reminds everyone that we aspire to be a
rules, processes and sustainable business and not just a collection of research projects (Firm N). On the walls
technologies that your of our labs are framed posters of famous scientists along with motivational messages
firm has for each type of that encourage us to try and make a difference (Firm I). Like the university recruitment
control system? process each member of my team gets to meet with potential new hires to suss out how
well they would fit in (Firm D)
Interactive: We use crude technology road-mapping exercises to forecast and
communicate technological developments (Firm C). This involves two types of activity.
First, our senior scientists and business development managers produce reports that
detail relevant trends for our industry. Second, we have strategic planning sessions
where we report this information to employees and develop action plans and allocate
resources (Firm M). A bit like 3M and Google we are allowed to allocate a percentage
of our time to work on pet projects (Firm F)
Boundary: Code of Conduct and standard operating procedures (All Firms); employees
can call an independent whistle-blowing hotline (Firm H). We adhere to a host of
regulatory constraints regarding the disposal of waste and handling of radioactive
material (Firm M). The tests, inspections and audits conducted by our quality
assurance department, as well as adhering to industry procedures and guidelines e.g.,
International Laboratory Accreditation Cooperation (Firm B). The FDA requires us to
develop and maintain risk management systems to ensure that the benefits of our
product outweigh the risks (Firm N). All scientific laboratories must enforce laboratory
dress codes (e.g., no shorts, no skirts, no sandals and no open-toed shoes) for safety and
reliability reasons (Firm A)



Table 3. (Contd.)
Model elements Representative and abbreviated answers*

Diagnostic: Individual scientists have specific objectives, which relate to project
milestones that keep the investors happy (Firm K). These are the systems that measure
how much work individuals do e.g., tests per day (Firm L). Employee awards would
constitute this form of control (Firm G). For our R&D we have budgets and budgetary
controls; these dictate who does what and what gets done (Firm F)
R&D contextual ambidexterity
Describe how the Exploration (beliefs and interactive): Some of the biggest successes in our industry have
different types of control involved companies changing direction with only 6 months of operating cash left. To do
system work together to this required beliefs and interactive systems capable of helping management to first spot
generate the exploitation the opportunity and then to steer the company in a new direction (Firm O). Our beliefs
and exploration behaviors systems direct what we do and why we do it, while our interactive systems provide the
in your firm? freedom and mechanisms to rethink what we do and what we do it (Firm B)
Exploitation (diagnostic and boundary): First and foremost we are a service company,
and we view innovation and learning as necessary but costly and hard to justify. As a
consequence our control systems focus on improving project efficiency and optimizing
revenues from existing assets (Firm H). Our boundary and diagnostic systems help us to
reassess risk and revise the go/no go guidelines for research activities (Firm O). I feel
that the whole drug development process, with all its checks and regulations, creates a
high level of consumer protection but also a risk averse product development culture
(Firm N).
Describe how the Feed-forward control orientation (beliefs and interactive): I would say these systems are
different types of control much more open and intangible in nature – they have to be – because we use them to try
system involve a feedback to make sense of all uncertainty in our industry (Firm I). What typically happens is that
and/or a feed-forward we use the industry forecasts and scenarios to see if we should maintain our current
control orientations in portfolio of projects . . . if a major change is required then that would almost certainly
your firm? require a change in aspects of our mission statement and project activities (Firm F). We
could do with formalizing these systems a lot more. This would help stop all the micro-
management we have, and free these guys to see problems coming in time so that we can
do something about them (Firm C).
Feedback control orientation (diagnostic and boundary): These systems are obviously
feedback in nature – they are all about doing checks and tests to see what has happened
or should have happened (Firm E). There is a view in my company that we have too
many of these systems and they provide so much feedback that we do not know what to
do with it (Firm L). Reviews, audits and certifications – these are all industry-
recognized ways of obtaining feedback (Firm G).
*Some of the answers have been abbreviated to protect to protect the anonymity of firms and respondents.

4.2. Analysis and findings 4.2.1. R&D strategic goals in biotechnology firms
In terms of R&D growth goals, our framework
We approached our analysis from a broadly and data indicate that this varied largely accord-
descriptive perspective, focusing on how the ing to the lifecycle stages of the firm, its markets
data illustrate the elements and relationships in and its products. This is consistent with strategic
our framework. Following guidelines for present- options which characterize the extent to which
ing qualitative case data (see Eisenhardt, 1989), firms focus on appropriating returns from stable
and the format used by Nag, Corley and Gioia and limited project portfolios (harvest), versus
(2007) for their study of strategic change in R&D building capacity to create new knowledge and
organizations, we present our questions and illus- technologies for new products and markets
trative answers in Table 3. This summary infor- (build) (Gupta and Govindarajan, 1984). Firms
mation complements our narrative in the E, G and K, for instance, focused on securing new
following sections, where we describe, using ex- sources of investment funding so as to grow R&D
amples, the conceptual logic and reality of our capacity, and to develop their early stage technol-
proposed framework. It is important to note that ogies. These firms were relatively young, small,
even though there were no major contradictory idea-rich, but resource-poor and thus concerned
comments or negative assessments of the frame- with building R&D capacity in a way that the
work, this does not constitute empirical support larger and more established firms were not. The
for the framework. The data are simply used to larger, more mature biotechnology organizations
illustrate the elements of the framework. (e.g., Firms B, H, I and M) were more concerned



with developing efficient R&D processes to cap- end-use markets, meant that they faced many
ture returns from existing resources and projects, technological, regulatory and market uncertainties.
and thus their investment intensity in new R&D This required their top-management team to de-
projects was much smaller. A respondent from velop and continuously use interactive systems:
Firm M, for example, stated that ‘once our ‘We set up committees to collect and report in-
company became public the whole nature of our formation on who was doing what in our industry.
R&D operations dramatically shifted from creat- This information was internally communicated to
ing new technologies, to marketing and selling the necessary project teams, so they could comment
our approved product line.’ Furthermore, respon- on and help us plan for the opportunities and
dents reported that this type of goal was linked to threats that we were facing.’
the use of beliefs systems which help create a Reliability goals specify the extent to which
common language, a shared understanding and R&D will be conducted in a timely fashion, and in
strategic coherence within an R&D organization. accordance with expected standards and codes of
For instance, the respondent from Firm N re- practice. Our data indicated that this goal is
ported that ‘when we were first formed we didn’t central to biotechnology firms, with all of the
have a mission statement. We were simply a respondents in our study making statements that
group of researchers who did research. Sixteen concurred with the view that ‘the success of any
years later, however, we have had four or five biotech firm is dependent on producing and
different mission statements, with the current one reporting good data, before funding runs out’
emphasizing our commitment to provide value to (Firm L). The importance of R&D reliability to
shareholders.’ Similarly, the respondent from biotechnology firms is reflected in the demands of
Firm E reported that ‘We all know that the goal their various stakeholders (e.g., patient groups,
is to build a company that will be big enough, in investors, collaborators and government agen-
terms of talent and promising intellectual prop- cies), who expect biotechnology firms to closely
erty, so as to attract partners or buyers. . . . . . . follow good scientific practice and conform to
and in terms of the control system that reminds us recognized rules and guidelines. Consequently, in
of this goal – it is communication, communica- terms of the link between reliability goals and
tion and more communication.’ control system use, our illustrative data support
Innovation goals delineate the kinds of out- the link to boundary systems. All of the respon-
comes that R&D organizations seek to produce. dents provided comments similar to those listed in
Although these goals can vary in a number of Table 3, indicating that reliability requires strong
ways, we focused solely on the magnitude of and effective boundary systems to reduce the risk
change in a firm’s technology, and the degree of of improper behaviors that might cause a project
benefit it brings to the market (e.g., incremental or service failure.
versus radical) (Wheelwright and Clark, 1992). In The fourth goal in our framework, efficiency,
terms of our data, respondents from Firms A and specifies the extent to which R&D organizations
B reported that their R&D focused on incremen- are concerned with productivity and cost effec-
tally enhancing their existing service offerings for tiveness as ways to maximize returns on invest-
existing customers in the biotechnology and phar- ment. Our data indicate that while efficiency is on
maceutical industries. In contrast, Firm O fo- the whole important to all the biotechnology
cused on adapting its existing technologies for firms in our study, it was less significant to the
human therapeutic disorders for animal care six drug development firms (Firms J, K, L, M, N
(farm and pet) markets. Our data also indicated and O), and the one medical device firm (Firm I),
that when an innovation goal specified radical than it was to the nine research services firms
innovations, then this was associated with greater (Firms A, B, C, D, E, F, G and H). The view of
use of interactive control than if the goal stipu- efficiency held by the drug and medical device
lated incremental innovations. For instance, a firms, is reflected in the statement that the ‘major
respondent from Firm A reported that his orga- time lag between R&D action and R&D outcome,
nization’s focus on refining existing technology typically limits our ability to efficiently control
for existing customers was largely driven by the our activities, and as a consequence being efficient
occasional project meeting with top managers, is not really a top priority. We just focus on doing
intended to ‘boost organizational dialogue’ about good science’ (Firm J). In contrast, the research
how to improve existing testing services. In the service organizations considered efficiency goals
case of Firm O, their desire to develop radically to be central to their R&D, which must continu-
new platform molecular technologies, for different ously produce innovations that help keep their



services cutting edge and competitive. This varia- and analyze data on changes in demand, pro-
tion in attitudes toward efficiency was also linked ducts, technology, competition and regulation.
to variations in the extent to which firms used Respondents also reported how R&D projects
diagnostic systems. For instance, the respondent were started, adjusted and stopped using infor-
from Firm B stated ‘many of our R&D projects mation from forecasting and assessment systems.
are in collaboration with customers who have Forecasting systems provide projections about
time, quality and cost expectations, . . . . . . and when things will happen (i.e., a drop in the
to meet these we have numerous budget and demand for existing products and services, or
project monitoring systems.’ the approval of a new regulation or competitive
product), while assessment systems provide esti-
4.2.2. Management control systems in biotechnol- mations of the impact of these changes on the
ogy firms organization and its R&D portfolio. Such inter-
The most common examples of beliefs systems active systems are also ‘interactive’ in the sense
identified by our respondents were the company that the managers use the information they un-
reports, mission statements and website pages cover ‘to continuously and directly involve them-
that each firm used to articulate the research selves in the decisions and behavior of their
vision and values of the organization. For drug subordinates’ (Chiesa et al., 2009a, p. 421). For
development firms, beliefs systems typically fo- instance, respondents reported the use of planning
cused on installing in employees the noble vision sessions ‘to distribute strategic information to em-
of serving patients, saving lives and eliminating ployees, and then work with them to develop action
pain and suffering, while being guided by research plans and allocate resources for existing and new
values such as respect, ethics, and team work. In projects’ (Firm I). Respondents also reported that
contrast while the research service firms we sur- project and organizational based interactive sys-
veyed had similar statements about research va- tems are used by managers and their teams to
lues, their visions focused more on being the best speculate on future R&D scenarios, and to ensure
or first choice service provider in their market. that the foci and aims of the other types of control
Other types of beliefs systems reported by our system are adjusted as R&D strategic goals shift.
respondents included the company recruitment Furthermore, all of the firms in our study had
process that ‘seeks to attract and recruit talent scientific advisory committees that provided advice
with attitudes and skills that are consistent with that could alter the strategies for planned, pipeline
our research values’ (Firm K), and the training and approved R&D projects. These committees
process, which ‘develops people in the ‘company dictated the direction of exploration and the scope
way’ by continually communicating our goals and of feed-forward control, necessary for R&D con-
achievements to all staff’ (Firm O). Respondents textual ambidexterity.
also reported the use of intra-company challenges Boundary systems act ‘like an organization’s
and socialization events to promote fun and brakes’ (Simons, 1995b, p. 84). They restrain and
creative thinking, and to build organizational guide employees so that the firm does not experi-
coherence. Also in some cases the architecture ence an accident, i.e., a failure and crisis. For
and decor of company buildings, along with the biotechnology firms the systems are typically built
pictures on the walls of corridors and labora- into a firm’s laboratory practices, project manage-
tories, were all designed to inspire employees with ment methods and resource allocation decision
creative, funky and free-thinking values. All of making processes. For instance, many respon-
these examples of beliefs systems are intended to dents reported that their firms had ‘a Code of
focus and energize employees in a way that is Conduct that all our employees must be certified
necessary for the exploration and feed-forward to’ (Firm D) and that ‘product approval is de-
control aspect of R&D contextual ambidexterity. pendent on us producing, analyzing and archiving
In terms of interactive systems, our framework risk data to ensure that the benefits of our
and data indicate that these include R&D specific planned drug outweigh the risks’ (Firm L). There
systems such as technology road-mapping (Phaal are also non-firm specific boundary systems,
et al., 2006) and real options methods (Barnett, which for biotechnology firms included regional
2005), all of which help companies understand the and national laws, and regulations from institu-
effects of emerging technologies. There are also tions such as the US Food and Drug Adminis-
more general strategic scanning and monitoring tration and the European Medicines Agency.
systems (e.g., market research, competitor analy- Furthermore, there are certification regimes such
sis and technology benchmarking) that collect as Good Laboratory Practices, which are ‘pri-



marily intended to guarantee safe animal and situation in which his biotechnology firm was
toxicology testing, yet also help to ensure that unable to make sufficient progress toward devel-
laboratory results are internally peer-reviewed. oping its sole technology for a specific therapeutic
This helps to limit the risk of producing results disorder, and thus was struggling to attract
that are wrong, fabricated or massaged’ (Firm K). further investment. With only a few months of
In summary, as R&D researchers enjoy relatively funding remaining, the company identified a
high levels of job autonomy, boundary systems completely different disease, with a much greater
are used to avoid or mitigate unsafe or unethical market value, which could be treated using their
behaviors, or actions that constitute scientific core platform technology. The discovery of this
misconduct. These systems counter the effects of opportunity is credited to the Chief Scientific
diagnostic systems, and are most likely to produce Officer ‘The fact is – his job was to continuously
the risk averse behaviors associated with exploi- search for, evaluate and share new technological
tation and the checking of conformance that goes opportunities with the board and project teams.
with feedback control. He was our interactive control system – charged
In terms of diagnostic systems, our frame- with worrying about our future.’
work and data indicate that these include project Similarly, in terms of control orientation – the
planning systems for target drug approval dates, extent to which control is ex post (after the event)
budget systems for project costs, laboratory man- or ex ante (before the event) in nature – our
agement information systems for recording and framework and data suggest that beliefs systems
analyzing sample tests, and clinical trial systems and interactive systems work together to promote
for measuring the performance or efficacy of a a feed-forward control orientation. This follows
drug. Such systems are progress-focused. Man- the original thinking of Simons (1995a, p. 108)
agers, boards and regulatory agencies identify who argued that interactive systems promote
research project goals, review progress and arrange ‘reforecasting of future states based on revised
post-project reviews to identify lessons and correc- current information’; however, for this to occur,
tive actions. In terms of R&D output, both the the projections must be focused and appropriate
drug development firms and the modeling and to the needs of the organization. Thus, beliefs
testing service firms in our study used a number systems work with interactive systems to ensure
of corporate level diagnostic systems. These mea- that information is sought and used in a way that
sured the production of scientific papers, returns promotes prediction and change that are relevant
from R&D collaborations, the creation and ap- to the organization. These systems are feed-for-
proval of patents and the revenue from new service ward in nature because they help managers to
technologies, licensed patents and approved pro- anticipate or forecast events and trends before
ducts. Thus, by their very nature, diagnostic con- they occur, allowing them to proactively redirect
trol systems are central to exploitation behaviors in organizational values and activities. For instance,
that they provide feedback, or after-the-event in- one respondent reported ‘we know that the pa-
formation, that is used to adjust and improve the tents on our products and our competitors’ pro-
performance of existing processes. ducts will expire someday, and even though we
cannot be certain what the impact will be when
4.2.3. Contextual ambidexterity in biotechnology these expiries occur, we still monitor, forecast and
firms develop scenarios so that we are prepared for
In terms of exploration, our framework and data these events’ (Firm I).
indicate that beliefs systems and interactive sys- In terms of exploitation, our framework and
tems work together to generate search and dis- data suggest this is attained by boundary systems
covery that are relevant and adaptable. Beliefs and diagnostic systems working in tandem.
systems such as the mission statement, the recruit- Boundary systems moderate exploration beha-
ment process, employee training and the company viors, induced by beliefs systems and interactive
rhetoric and symbols, are all used to focus and systems, by defining and restricting the search
guide employees to search for and create new space and activities that can be undertaken (Si-
competences and knowledge. In combination, mons, 1995a; Widener, 2007). They reign in
interactive systems such as technology road-map- employee freedoms to counter the autonomy
ping, market forecasting and impact assessments and inspiration that drive the exploration for
are used to maintain or adjust the specific direc- new knowledge. Diagnostic systems also counter
tion of this exploration activity over time. For exploration, by using relatively short-term effi-
example, a respondent from Firm K described a ciency measures to refine and extend existing



competencies. These systems typically focus on tegic goals influence the attention placed on
and require exploitation related outcomes that are different types of control system. As the use of
‘positive, proximate and predictable’ in nature multiple control systems can require considerable
(March, 1991, p. 85). These combined effects of managerial attention, management should prior-
boundary systems and diagnostic systems were itize where to focus their attention and resources
exemplified by the statement that ‘we use bound- (Marginson, 2002; Widener, 2007). Also, as dif-
ary control in conjunction with our project per- ferent goals can have different and sometimes
formance systems to examine and check that conflicting implications for organizational beha-
everything is going according to plan. We don’t vior, the use of different control systems could
like surprises and nor do our customers’ (Firm E). help to manage any potential conflicts presented
This type of control counters the instability, by multiple goals. This can be examined by survey-
uncertainty and serendipity often associated ing employees, to assess the extent to which their
with R&D exploration. R&D organization is guided by different goals, and
In terms of control orientation, our framework the number of people hours associated with each
and data suggest that boundary and diagnostic type of control system (Chiesa and Frattini, 2007).
systems in biotechnology firms jointly promote a Second, with our framework and illustrative
feedback control orientation. Diagnostic systems data, we claim that beliefs and interactive systems
are used to monitor, review and test, so as to provide a feed-forward control orientation that
ensure things are on track in terms of ‘what’ is generates or enhances exploration, whereas
being done. If they are not, then other control boundary and diagnostic controls provide a feed-
systems and activities can be adjusted accordingly. back control orientation that generates or en-
Managers also use this after-the-event information hances exploitation. While we consider this to
to motivate and reward the behaviors of indivi- be true, all frameworks are simplified representa-
duals, teams and organizations. In contrast, tions of reality (Box, 1979). Consequently, we
boundary systems specify and check ‘how’ things suggest that our framework provides a starting
are done, in accordance with pre-defined standards point for unpacking the complexities of the con-
and regulations. These systems also provide error- trol-behavior relationship. For example, it is not
based feedback control, i.e., any deviation from just the combination of control systems used that
specified practices leads to corrective action and if matters, but also the substantive content of the
these deviations persist, then stronger more influ- controls – what is dictated, discussed, projected,
ential boundary systems are installed. For in- measured, monitored and evaluated. Thus, know-
stance, one of the respondents from Firm M ing that an organization is using a certain type of
reported that ‘if biotechnology firms experience control system provides a first-order level insight
one major incident or repeated minor incidents of into the control-behavior relationship. The next
scientific misconduct, then this typically leads to a level is to understand the effectiveness of the
corrective action where guidelines and checks for different rules, procedures, technologies and in-
laboratory practice are tightened up.’ centives that R&D managers use in conjunction
with each type of control system.
Third, researchers could explore how control
5. Implications and future research systems could be used to attain low or high levels
opportunities of balanced ambidexterity (Cao et al., 2009). Low
balanced ambidexterity is when a firm’s level of
The central contribution of our paper is the exploitation is significantly lower than that of
development and illustration of an R&D manage- exploration, and vice versa; while high balanced
ment control framework for attaining contextual ambidexterity is when a firm has similar moderate
ambidexterity in R&D organizations. We now levels of both exploration and exploitation. By
discuss several implications of the framework, of emphasizing different control system combina-
relevance to both management practice and fu- tions, it could be possible for R&D managers to
ture empirical research. attain, at specific periods in time, different mixes
First, our framework posits that there is a of this exploitation-exploration balance.
strong (though not exclusive) relationship be- A fourth implication of our framework con-
tween individual R&D goals and the use of cerns the influence of other contingency factors
different types of control system. An important such as the size, age and life-cycle of the R&D
implication of this is the need to empirically organization, and its industry conditions, on the
examine to what extent the different R&D stra- use of different management control systems.



Although the focus of our paper was on under- call ‘capability toggling’, and Thomas et al. (2005)
standing how different types of control system, call ‘irregular oscillation,’ where the balancing of
guided by different R&D strategic goals, can be exploitation-exploration tensions is much like
used to induce contextual ambidexterity, the riding a bike – it requires a continuous and
illustrative empirical evidence we present hints irregular shifting of control system use over time.
at the role of these factors. For instance, some
organizations (e.g., Firms E, G and K) were
relatively young and small, and concerned with 6. Conclusion
goals and controls that sought to build R&D
capacity in a way that the larger and more In this paper, we introduced a framework that
established firms were not (e.g., Firms B, H, I shows how four types of control system, each
and M). Thus, the framework we present could be guided by an R&D goal, combine to induce the
modified by other researchers to explore how such behaviors, outcomes and control orientations
antecedents drive the use and outcomes of R&D (feedback versus feed-forward) necessary for con-
control systems. textual ambidexterity. We illustrated these frame-
A fifth implication of our framework concerns work elements and their linkages, using data from
how R&D managers might use control systems to biotechnology firms, so as to clarify what these
‘dynamically’ shift, over time, the exploitation- elements are and what they do. While this helps to
exploration balance. The balanced viewed of make the framework more useful and meaningful
ambidexterity (March, 1991; Cao et al., 2009) is to scholars and practitioners, this illustration might
largely ‘static’ in nature, in that it refers to an also suggest that R&D control is a straightforward
optimal mix of exploitation and exploration at a task. However, this is not the case. An inherent
point in time. However, as discussed above, R&D challenge to understanding and practicing effective
control is driven and constrained by environmen- R&D management control is the fact that it is
tal antecedents such as the size, age and life cycle concerned with controlling the production of
of the R&D organization, as well as by changes in knowledge, something that is inherently unobser-
industry conditions such as demand, competitors, vable. As knowledge increasingly redefines the
technologies, products and regulation. As each of wealth of nations, firms and individuals, the chal-
these conditions has a distinct velocity (a rate and lenges and benefits of effective R&D control will
direction of change (McCarthy et al., 2010)), – an continue to capture the attention of scholars and
optimum mix of exploitation-exploration at one managers. Thus, we hope that our framework will
point in time is likely to become unsuitable as motivate researchers to further examine how
industry conditions change over time. This makes broader forms of control, guided by R&D objec-
the balancing of exploitation and exploration a tives and other environmental factors, act as orga-
dynamic problem. This is tentatively supported nizational levers for balancing different forms of
by our data where respondents explained how the knowledge production over time.
balance of exploitation and exploration altered
over time as a firm advanced through its life cycle,
or changed the focus of its R&D activities (see: Acknowledgements
Table 3, the R&D contextual ambidexterity sec-
tion). For instance, the respondent from Firm O We are grateful to two anonymous reviewers for
explained how different beliefs and interactive their helpful and constructive comments. The
systems were required to shift the balance toward development of this paper also benefited from
exploration so as to help take the company in a comments from Danny Breznitz, Elicia Maine,
new direction. The importance of this implication Jane McCarthy, Karen Ruckman, Christos Tsi-
has been highlighted by scholars who suggest that nopoulos and participants at the R&D Manage-
‘given the dynamism of markets and organiza- ment Conference 2008. We also acknowledge
tions, it is important to develop theories that research funding from Canada’s Social Science
combine static elements with more dynamic per- and Humanities Research Council.
ceptions of ambidexterity’ (Raisch et al., 2009, p.
686). Thus, we suggest that our framework can be
used to explore how R&D managers use different References
control systems to dynamically shift the balance
or mix between exploitation and exploration over Addison, L.E., Litchfield, J.W. and Hansen, J.V. (1976)
time. This follows what McCarthy et al. (2006) Managing growth and change in an R&D organiza-



tion: the role of dynamic modeling. R&D Manage- innovation projects. European Journal of Innovation
ment, 6, 2, 77–80. Management, 12, 4, 416–443.
Ahuja, G. and Lampert, C.M. (2001) Entrepreneurship Chiesa, V., Frattini, F., Lazzarotti, V. and Manzini, R.
in the large corporation: a longitudinal study of how (2009b) Performance measurement of research and
established firms create breakthrough inventions. development activities. European Journal of Innova-
Strategic Management Journal, 22, 521–543. tion Management, 12, 1, 25–61.
Alavi, M. and Leidner, D.E. (2001) Review: knowledge Chiesa, V. and Masella, C. (1996) Searching for an
management and knowledge management systems: effective measure of R&D performance. Manage-
conceptual foundations and research issues. MIS ment Decision, 34, 7, 49–57.
Quarterly, 25, 1, 107–136. Cooke-Davies, T.J. and Arzymanow, A. (2003) The
Anthony, R. (1965) Planning and Control Systems: A maturity of project management in different indus-
Framework for Analysis. Boston: Harvard University tries: an investigation into variations between project
Press. management models. International Journal of Project
Argyris, C. and Schon, D.A. (1978) Organizational
¨ Management, 21, 6, 471–478.
Learning: A Theory of Action Perspective. Amster- Cooper, R.G. and Kleinschmidt, E.J. (1986) An inves-
dam: Addison-Wesley. tigation into the new product development process:
Barnett, M.L. (2005) Paying attention to real options. steps, deficiencies, and impact. Journal of Product
R&D Management, 35, 1, 61–72. Innovation Management, 3, 2, 71–85.
Bart, C.K. (1993) Controlling new product R&D Cordero, R. (1990) The measurement of innovation
projects. R&D Management, 23, 3, 187–197. performance in the firm: an overview. Research
Berthon, P.R., Pitt, L.F., McCarthy, I.P. and Kates, Policy, 19, 2, 185–192.
S.M. (2007) When customers get clever: managerial Cummings, J.N. (2004) Work groups, structural diver-
approaches to dealing with creative consumers. Busi- sity, and knowledge sharing in a global organization.
ness Horizons, 50, 1, 39–47. Management Science, 50, 3, 352–364.
Bianchi, M., Chiesa, V. and Frattini, F. (2009) Explor- Daft, R.L. and Weick, K.E. (1984) Toward a model of
ing the microfoundations of external technology organizations as interpretation systems. Academy of
commercialization. European Journal of Innovation Management Review, 9, 2, 284–295.
Management, 12, 4, 444–469. Davila, A. and Foster, G. (2007) Management control
Birkinshaw, J. and Gibson, C.B. (2004) Building ambi- systems in early-stage startup companies. The Ac-
dexterity into an organization. Sloan Management counting Review, 82, 4, 907–937.
Review, 45, 4, 47–55. Duncan, R.B. (1976) The ambidextrous organization:
Box, G.E.P. (1979) Some problems of statistics and designing dual structures for innovation. In: Kil-
everyday life. Journal of the American Statistical mann, R.H., Pondy, L.R. and Slevin, D. (eds), The
Association, 74, 365, 1–4. Management of Organization, Vol. 1. New York:
Bremser, W.G. and Barsky, N.P. (2004) Utilizing the North-Holland, pp. 167–188.
balanced scorecard for R&D performance measure- Eisenhardt, K.M. (1985) Control: organizational
ment. R&D Management, 34, 3, 229–238. and economic approaches. Management Science,
Cao, Q., Gedajlovic, E. and Zhang, H. (2009) Unpack- 31, 134–149.
ing organizational ambidexterity: dimensions, con- Eisenhardt, K.M. (1989) Building theories from case
tingencies and synergistic effects. Organization study research. Academy of Management Review, 14,
Science, 20, 781–796. 532–550.
Chiesa, V., Coughlan, P. and Voss, C.A. (1996) Devel- Freeman, C. (1969) Measurement of Output of Research
opment of a technical innovation audit. Journal of and Experimental Development. Statistical Reports
Product Innovation Management, 13, 2, 105–136. and Studies. Paris: UNESCO.
Chiesa, V. and Frattini, F. (2007) Exploring the differ- Garcı´ a-Valderrama, T. and Mulero-Mendigorri, E.
ences in performance measurement between research (2005) Content validation of a measure of
and development: evidence from a multiple case R&D effectiveness. R&D Management, 35, 3, 311–
study. R&D Management, 37, 4, 283–301. 331.
Chiesa, V. and Frattini, F. (2009) Evaluation and Geisler, E. (2002) The metrics of technology evaluation:
Performance Measurement of Research and Develop- where we stand and where we should go from here.
ment: Techniques and Perspectives for Multi-Level International Journal of Technology Management, 24,
Analysis. Cheltenham, U.K: Edward Elgar. 4, 341–374.
Chiesa, V., Frattini, F., Lazzarotti, V. and Manzini, R. Ghoshal, S. and Bartlett, C.A. (1997) The Individualized
(2008) Defining a performance measurement system Corporation: A Fundamentally New Approach to
for the research activities: a reference framework and Management. New York: Harper Business.
an empirical study. Journal of Engineering and Tech- Gibson, C.B. and Birkinshaw, J. (2004) The antece-
nology Management, 25, 3, 213–226. dents, consequences, and mediating role of organiza-
Chiesa, V., Frattini, F., Lamberti, L. and Noci, G. tional ambidexterity. Academy of Management
(2009a) Exploring management control in radical Journal, 47, 2, 209–226.



Godener, A. and Soderquist, K.E. (2004) Use and
¨ McGrath, M.E. and Romeri, M.N (1994) The R&D
impact of performance measurement results in effectiveness index: a for product development per-
R&D and NPD: an exploratory study. R&D Man- formance. Journal of Product Innovation Manage-
agement, 34, 2, 191–219. ment, 11, 213–220.
Gupta, A.K. and Govindarajan, V. (1984) Knowledge McNamara, P. and Baden-Fuller, C. (1999) Lessons
flows and the structure of control within multina- from the Celltech case: balancing knowledge explora-
tional corporations. The Academy of Management tion and exploitation in organizational renewal.
Journal, 16, 4, 768–792. British Journal of Management, 10, 4, 291–307.
Gupta, A.K., Smith, K.G. and Shalley, C.E. (2006) The McNamara, P. and Baden-Fuller, C. (2007) Share-
interplay between exploration and exploitation. holder returns and the exploration-exploitation di-
Academy of Management Journal, 49, 693–708. lemma: R&D announcements by biotechnology
Hauser, J.R. and Zettelmeyer, F. (1997) Metrics to firms. Research Policy, 36, 4, 548–565.
evaluate R, D&E. Research Technology Manage- Moravesik, M.J. (1973) Measures of scientific growth.
ment, 40, 4, 32–38. Research Policy, 2, 3, 266–275.
Jansen, J., van den Bosch, F. and Volberda, H. (2006) Nag, R., Corley, K.G. and Gioia, D.A. (2007) The
Exploratory innovation, exploitative innovation, and intersection of organizational identity, knowledge,
performance: effects of organizational antecedents and practice: attempting strategic change via knowl-
and environmental moderators. Management edge grafting. Academy of Management Journal, 50,
Science, 52, 1661–1674. 4, 821–847.
Karlsson, M., Trygg, L. and Elfstrom, B. (2004) Otley, D.T. (1980) The contingency theory of manage-
Measuring R&D productivity: complementing the ment accounting: achievements and prognosis. Ac-
picture by focusing on research activities. Technova- counting, Organizations and Society, 5, 413–428.
tion, 24, 179–186. Ouchi, W.G. (1979) A conceptual framework for the
Kerssens-van Drongelen, I.C. and Bilderbeek, J. (1999) design of organizational control mechanisms. Man-
R&D performance measurement: more than choos- agement Science, 25, 833–848.
ing a set of metrics. R&D Management, 29, 1, 35–46. Phaal, R., Farrukh, C.J.P. and Probert, D.R. (2006)
Kerssens-van Drongelen, I.C. and Cook, A. (1997) Technology management tools: concept, develop-
Design principles for the development of measurement and application. Technovation, 26, 3, 336–344.
ment systems for research and development pro- Raisch, S. and Birkinshaw, J. (2008) Organizational
cesses. R&D Management, 27, 4, 345–357. ambidexterity: antecedents, outcomes, and modera-
Kiella, M.L. and Golhar, D.Y. (1997) Total quality tors. Journal of Management, 34, 3, 375–409.
management in an R&D environment. International Raisch, S., Birkinshaw, J., Probst, G. and Tushman,
Journal of Operations and Production Management, M.L. (2009) Organizational ambidexterity: balancing
17, 2, 184–198. exploitation and exploration for sustained perfor-
Levinthal, D. and March, J. (1993) The myopia of mance. Organization Science, 20, 4, 685–695.
learning. Strategic Management Journal, 14, 95–112. Schainblatt, A.H. (1982) How companies measure the
Maine, E. and Garnsey, E. (2006) Commercializing productivity of engineers and scientists. Research
generic technology: the case of advanced materials Management, 25, 3, 10–18.
ventures. Research Policy, 35, 3, 375–393. Schumann, P.A., Ransley, D.L. and Prestwood, D.C.L.
March, J.G. (1991) Exploration and exploitation in (1995) Measuring R&D performance. Research-
organizational learning. Organization Science, 2, 1, Technology Management, 38, 3, 45–54.
71–87. Simons, R. (1994) How new top managers use control
Marginson, D.E. (2002) Management control systems systems as levers of strategic renewal. Strategic
and their effects on strategy formation at middle- Management Journal, 15, 169–189.
management levels: evidence from a UK organiza- Simons, R. (1995a) Levers of Control. Boston, MA:
tion. Strategic Management Journal, 23, 1019–1031. Harvard Business School Press.
Martin, B.R. and Irvine, J. (1983) Assessing basic Simons, R. (1995b) Control in an age of empowerment.
research: some partial indicators of scientific Harvard Business Review, 80–88.
progress in radio astronomy. Research Policy, 12, Souder, W.E. (1972) An R&D planning and control
2, 61–90. servosystem: a case study. R&D Management, 3, 1,
McCarthy, I.P., Lawrence, T.B., Wixted, B. and Gor- 13–21.
don, B.R. (2010) A multidimensional conceptualiza- Swan, J., Newell, S., Scarbrough, H. and Hislop, D.
tion of environmental velocity. Academy of (1999) Knowledge management and innovation: net-
Management Review, 35, 4, 604–626. works and networking. Journal of Knowledge man-
McCarthy, I.P., Tsinopoulos, C., Allen, P. and Rose- agement, 3, 4, 262–275.
Anderssen, C. (2006) New product develop- Szakonyi, R. (1995) Measuring R&D effectiveness I.
ment as a complex adaptive system of decisions. Research-Technology Management, 37, 27–32.
Journal of Product Innovation Management, 23, 5, ´
Thomas, C., Kaminska-Labbe, R. and McKelvey, B.
437–456. (2005) Managing the MNC and exploitation/



exploration dilemma: from static balance to dynamic Notes
oscillation. In: Szulanski, G., Porac, J.P. and Doz, Y.
(eds), Advances in Strategic Management: Strategy 1. Some of the answers have been abbreviated to protect
Process, Greenwich, CT: JAI Press. 22, pp. 213–247. the anonymity of firms and respondents.
Turner, K.L. and Makhija, M.V. (2006) The role of
organizational controls in managing knowledge.
Academy of Management Review, 31, 1, 197–217.
Tushman, M.L. and O’Reilly, C.A (1996) Ambidex- Ian McCarthy is the Canada Research Chair in
trous organization: managing evolutionary and re- Technology and Operations Management in the
volutionary change. California Management Review, Faculty of Business Administration at Simon
38, 4, 8–30. Fraser University. He received his MSc and
Tymon, W.G. Jr and Lovelace, R.F. (1986) A taxon- PhD degrees from the University of Sheffield,
omy of R&D control models and variables affecting and before joining Simon Fraser University he
their use. R&D Management, 16, 3, 233–241.
was a faculty member at the Universities of
Werner, B.M. and Souder, W.E. (1997) Measuring
Warwick and Sheffield. His research focuses on
R&D performance-state of the art. Research Tech-
nology Management, 40, 2, 34–43. how firms differ in their innovation processes,
Wheelwright, S.C. and Clark, K.B. (1992) Revolutioniz- product and service customization approaches,
ing Product Development: Quantum Leaps in Speed, R&D management control systems, outsourcing
Efficiency, and Quality. New York: Free Press. practices and collaborative networks. His re-
Whittington, R. (1991) Changing control strategies in search has been published in engineering and
industrial R&D. R&D Management, 21, 1, 43–53. business journals, including The Academy of
Widener, S.K. (2007) An empirical analysis of the levers Management Review, International Journal of
of control framework. Accounting, Organizations and Production Economics, Journal of Production In-
Society, 32, 7–8, 757–788. novation Management, Technovation and
Yawson, R.M., Amoa-Awua, W.K., Sutherland, A.J.,
OMEGA.
Smith, D.R. and Noamesi, S.K. (2006) Developing a
performance measurement framework to enhance
the impact orientation of the Food Research Insti- Brian R. Gordon is a Doctoral Candidate in
tute, Ghana. R&D Management, 36, 2, 161–172. business administration at Simon Fraser Univer-
von Zedtwitz, M. (2002) Organizational learning sity. His research interests include knowledge and
through post–project reviews in R&D. R&D Man- knowledge creation processes in strategy, entre-
agement, 32, 3, 255–268. preneurship and innovation.


Achieving contextual ambidexterity in R&D organizations: a management control system approach

More Related Content

What's hot (18)

Viewers also liked (20)

Similar to Achieving contextual ambidexterity in R&D organizations: a management control system approach (20)

More from Ian McCarthy (20)

Recently uploaded (20)

Achieving contextual ambidexterity in R&D organizations: a management control system approach