DOI: 10.1111/emre.12535
S P EC I A L I S SU E ART I C L E
How agency and self-efficacy moderate the effects of strategic
improvisational behaviors on sales performance: Evidence from an
emerging market
Magnus Hultman1 | Nathaniel Boso2,3 | Abena Animwaa Yeboah-Banin4 |
Ian Hodgkinson5 | Anne L. Souchon5 | Ekaterina Nemkova6 | Joao Oliveira7 |
Paul Hughes8
1Goodman School of Business, Brock Abstract
University, St. Catharines, Ontario, Canada This study develops and tests arguments that improvisation is not universal in its
2KNUST School of Business, Kwame benefits for the firm, but rather its multidimensional characteristics (action-
Nkrumah University of Science and
Technology, Kumasi, Ghana orientation, creativity, and spontaneity) hold differential performance effects. The
3Gordon Institute of Business Science, study further examines whether these relationships are contingent upon individual
University of Pretoria, Pretoria, South Africa agency and self-efficacy. Drawing on primary data from industrial sales account
4Department of Communication Studies, managers in Ghana, the study finds that an increasing level of action-orientation
University of Ghana, Legon, Ghana is associated with decreases in perceived sales performance and the decrease in
5School of Business and Economics, performance is more pronounced under conditions of stronger sense of agency
Loughborough University, Loughborough,
UK, Loughborough, UK and self-efficacy. Similarly, an increasing level of creativity is associated with
6IE
SEG School of Management, Paris, France decreases in perceived sales performance when agency is stronger. However, an
7Essex Business School, University of Essex, increasing level of spontaneity is associated with increases in performance and this
Southend on Sea, UK increase is strengthened under conditions of stronger sense of self-efficacy. The
8School of Business, University of Leicester, study concludes that the effect of strategic improvisation on sales performance
Leicester, LE2 1RQ, UK outcome within the context of an emerging economy (such as Ghana) is more
nuanced than established improvisation literature suggests.
Correspondence
Magnus Hultman, Goodman School of
Business, Brock University, St. Catharines, ON, KEYWORDS
Canada. action-orientation, agency, creativity, emerging economy, improvisation, perceived sales performance,
Email: mhultman@brocku.ca self-efficacy, spontaneity
INTRODUCTION performance (Perry, 1991), strategic improvisation is
observed to encompass “a reduced temporal gap between
There is limited fine-grained theory and evidence about the planning and implementation of unique actions … it
the occurrence of strategic improvisation and when it applies to actions and decisions that are novel, or devia-
generates value for firms operating in fast changing and tions from standard practices” (Bergh & Lim, 2008,
precarious market environments. Strategic improvisation p. 599).
encompasses improvisational activities that can shape While studies have captured the strategic improvisa-
firm-level goals and performance (Cunha et al., 2016). tion construct as an umbrella term encompassing a com-
Such improvised activities are often defined as “the spon- plex set of organizational processes (Pina e Cunha
taneous and creative process of attempting to achieve an et al., 2014), at its core is individual decision practices
objective in a new way” (Vera & Crossan, 2005, p. 733), and actions. Strategic improvisation (hereafter improvi-
highlighting the importance of realized actions (Cunha sation) has been conceptualized as a three-dimensional
et al., 2002; Pina e Cunha et al., 1999), of extemporane- construct comprising elements of creativity, spontaneity,
ousness (Moorman & Miner, 1998), and of inventiveness/ and action-orientation (Nemkova et al., 2015). Specifi-
creativity (Kamoche et al., 2003; Kamoche & cally, creativity encapsulates generation of novel and use-
Cunha, 2001; Nemkova et al., 2015). Understood as the ful solutions to emergent and neglected problems and
simultaneous enactment of strategy composition and opportunities (Evans et al., 2012; Proctor, 1991).
European Management Review. 2022;19:417–435. wileyonlinelibrary.com/journal/emre © 2022 European Academy of Management (EURAM). 417
418 HULTMAN ET AL.
Spontaneity captures ability to react to situations and solving organizational problems under unpredictable and
make decisions “in the moment” (Souchon et al., 2016). precarious business environment conditions. To test the
Action-orientation consists of a willingness to perform study’s hypotheses, therefore, primary data are obtained
intended activities to address imminent problems from sales managers involved in complex business-to-
(Hmieleski et al., 2013). As discussed by Nemkova business sales negotiations with industrial customers
et al. (2015, p. 44): “The spontaneity dimension is related where competence in finding creative, spontaneous, and
to time orientation, meaning that people react to situa- actionable solutions to imminent customer problems is
tions and make decisions in the moment rather than very important. This study examines the behavior of
anticipate what might happen […] The creativity dimen- industrial sales managers in Ghana, a turbulent emergent
sion of improvisation relates to the search for novelty economy market that is experiencing significant market
and usefulness while making decisions […] Finally, restructuring and institutional changes. The empirical set-
improvisation incorporates action orientation, which ting for this study addresses several calls on researchers
reflects managers’ ability to maintain an activity and to examine the improvisation construct in evolving and
focus their attention on imminent problems.” While pre- fast changing industrial and market settings (e.g.,
vious research has made efforts to advance knowledge on Panagopoulos et al., 2011).
the dimensionality of the improvisation construct
(e.g., Ciuchta et al., 2021; O’Toole et al., 2020), knowl-
edge is lacking on the extent to, and the conditions under THEORY AND HYPOTHESIS
which, the individual dimensions of improvisation con- DEVELOPMENT
tribute to firm-level outcomes.
Thus, the objective of this study is to explore the Improvisation as a choice-action process
microfoundation of the strategic improvisation construct
my examining the relationship between its conceptual Decision theory suggests that decision choices may be
components (i.e., spontaneity, creativity, and action- driven by either normative or descriptive approaches
orientation) and sales performance (a firm-level outcome), (Cabantous & Gond, 2011; Nemkova et al., 2012;
and the psychological processes under which these rela- Nutt, 2008). The traditional normative decision-making
tionships are more or less pronounced. Accordingly, this perspective argues that individuals make optimal and
study examines the research question: What are the rational choices based on formalized generation, evalua-
forms of the relationships between improvisational tion, and utilization of information surrounding situa-
behaviors and sales performance, and under what condi- tions (Cao et al., 2019). The dominant theme of the
tions are these relationships more or less pronounced? normative decision-making approach suggests that orga-
In answering this research question, we advance the nizations should examine their environment and then
existing improvisation literature by demonstrating how derive lists of possible decisions (Bell et al., 1988). For
each component element uniquely contribute to sales this, they need to initially develop a set of objective
performance. The study further advances the improvisa- criteria to evaluate alternatives. Therefore, decision-
tion literature by identifying individual agency and self- making represents a series of sequential and analytical
efficacy as two psychological processes to explain the processes that include collection and analysis of informa-
conditions under which improvisational behaviors are tion (e.g., surveys and focus groups), formulation of dif-
more or less related to sales performance (Shalley ferent alternatives of action, evaluation of these
et al., 2004; Singh & Koshy, 2010; Sousa & alternatives, and making a choice that will maximize the
Coelho, 2011). Agency is defined as an individual’s abil- outcomes. Here, individuals draw on mental frameworks
ity to make decisions, take purposeful action, and influ- to guide decision choices (Simon, 1960). Drawing from
ence decision-making (Corbett et al., 2018; Lawrence & cognitive theories, mental frameworks (e.g., Gaglio &
Suddaby, 2006; Wolfgramm et al., 2015). Because Katz, 2001) such as cognitive schemas (e.g., Baron &
agency enables individuals to be increasingly purposeful Ensley, 2006) determine how individual decision-makers
and persistence when making decisions, greater agency respond to new information. For instance, individuals
may minimize the extent to which improvisational may use customer problem alertness schemas to identify
behaviors contribute to sales performance. Additionally, problems and respond to this new information by system-
self-efficacy captures individuals’ belief in their own atically designing rational solutions to resolve the prob-
abilities to meet challenges before them and to effec- lem (Gaglio & Katz, 2001). Consequently, the rational-
tively complete tasks (Akhtar, 2008). Because self- design literature emphasizes a normative step-by-step
efficacy highlights individuals’ overall belief in their approach that is grounded on formal information
ability to succeed, greater self-efficacy may be associated processing as a precursor to effective selling (Anderson &
with a stronger relationship improvisational behaviors Huang, 2006; Moncrief & Marshall, 2005). The rational-
and sales performance. planning model, as has been observed in emerging East
The unique contextual setting of the study is one Asian markets, constitutes an example of such type of
where use of improvisation is salient and relevant in approach (Hughes et al., 2018).
 17404762, 2022, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/emre.12535 by University of Ghana - Accra, Wiley Online Library on [19/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
HULTMAN ET AL. 419
However, we argue that in the current era where sales challenging and vaguely structured situations such as
success is based on contextual responsiveness (Wang & under crises (e.g., Hughes et al., 2020) and often experi-
Netemeyer, 2004), the descriptive decision approach enced in the sales context (Wang & Netemeyer, 2004)
offers more insight (Locander et al., 2014; Nemkova and, thus, to contribute to enhance organizational perfor-
et al., 2012). In reality, “decision makers satisfy instead mance (e.g., Agnihotri et al., 2013; Martinaityte &
of optimize, rarely engage in comprehensive search and Sacramento, 2013; Sousa & Coelho, 2011; Wang &
discover their goals in the process of searching” Miao, 2015; Wang & Netemeyer, 2004).
(Eisenhardt & Zbaracki, 1992, p. 22). A tenet of descrip- In spite of the expected positive outcomes of improvi-
tive decision theory is that decision choices are not con- sation for organizations, empirical evidence suggests
text free and often “flow from cognitive limitations, inconclusive and conflicting findings at the dimensional
political processes, routines, and environmental con- level (e.g., Ford & Sullivan, 2004; Gong et al., 2013;
straints” (Haley & Stumpf, 1989, p. 447). Such notion Martinaityte & Sacramento, 2013). The ambiguity in the
often manifests itself, in practical terms, through the exis- extant literature highlights that while improvisation may
tence of a discrepancy between what organizations expect contribute to firm performance, this may not always be
of their staff and what staff are themselves more likely to the case. As explained by Vera and Crossan (2005),
do (Crossan et al., 2005). This suggests that choices are improvisation in itself is neither good nor bad; its effec-
inherently bounded in rationality (Etzioni, 2014), as tiveness in terms of contributing to firm performance will
human rationality has its natural limitations due to the depend on decision-makers’ improvisational skills. There-
psychological ability to process information and draw fore, there is a strong argument to further investigate the
conclusions (Simon, 1955). To remove this limitation, form of relationship between improvisation and perfor-
individuals must focus on cues present within given situa- mance and the boundary conditions of this relationship.
tions to respond quickly and appropriately to Figure 1 displays this study’s conceptual framework.
unpredictable events (Mendonça et al., 2004) and identify
situation-relevant solutions and opportunities (Bonney &
Williams, 2009). In addition to situational cues, individ- Improvisation and sales performance
uals also rely on heuristics, experience, and intuition to
devise useful solutions to problems (Eling et al., 2014; Sparks (1994) viewed sales improvisation as the degree of
Hodgkinson & Sadler-Smith, 2018; Leybourne & Sadler- latitude and autonomy afforded to salespeople and con-
Smith, 2006). For instance, decisions following the cluded that low levels of improvisation inhibit sales suc-
descriptive school of thought can be linked to “fast and cess. Since then, improvisation in the sales function has
frugal” decision-making where a decision-maker relies on enjoyed a more direct conceptual understanding (Banin
a repertoire of simple heuristics, including the Recogni- et al., 2016) mirroring evolving work on strategic impro-
tion heuristic and the Take the Best heuristic visation (e.g., Ciuchta et al., 2021; O’Toole et al., 2020;
(Goldstein & Gigerenzer, 2002). The former allows for Parhizgar et al., 2017). That said, results remain few and
exploiting knowledge gaps whereas the latter deliberately inconclusive. Following the approach used by previous
ignores information (is akin to information disavowal). research, we address this issue by positing that improvisa-
Therefore, rather than using information to plan future tion may have nonlinear (i.e., curvilinear) relationship
choices (as prescribed by the normative approach), in the with sales performance. That is, the strength and sign of
descriptive approach, individuals are in tune with what is the relationships between sales performance and the dif-
happening in the “here and now” to inform behavior and ferent dimensions of improvisation (i.e., creativity, spon-
choice options within the given situation (Wiltbank taneity, and action-orientation) may not be constant for
et al., 2006). Such phenomenon is appropriately all levels of those dimensions. More specifically, we argue
described by Hodgkinson et al. (2009) as knowing the that those relationships may have an inverted U-shape.
“right” course of action but without knowing why or, in
other words, as improvisation in decision-making.
Because sales situations are typically ambiguous
(Wang & Netemeyer, 2004) due to the closeness of the
sales function to customers’ evolving problems (Kim
et al., 2019), the most plausible options often tend to con-
sist of descriptive choices that emphasize unplanned solu-
tions. Because sales roles are largely unstructured and
often self-governed (Agnihotri et al., 2013), sales account
managers often choose to “delight customers in uncon-
ventional ways” through their extemporaneous decision
actions (Agnihotri et al., 2013, p. 2). Moreover, behaviors
associated with improvisation (e.g., Strutton et al., 2009)
typically enable decision-makers to better cope with F I GURE 1 Conceptual framework
 17404762, 2022, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/emre.12535 by University of Ghana - Accra, Wiley Online Library on [19/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
420 HULTMAN ET AL.
We argue that improvisation enhances timely respon- outcomes through greater responsiveness (Nemkova
siveness to specific situations (Nemkova et al., 2015). et al., 2015). Yet spontaneity could lead to “chaos”
Because real responsiveness lies not merely in generating (Eisenhardt & Brown, 1998), which may “result in
solutions to customer problems, but in doing so time- actions that emerge with little consideration of their place
ously (Homburg et al., 2007; O’Toole et al., 2020; within the firm’s overarching strategy” (Souchon
Tom & Lucey, 1997), the ability to apply moderate levels et al., 2016, p. 676), thereby undermining firm success.
of improvisation in decision-making should improve sales Furthermore, similar assertions can also be made regard-
performance. However, high levels of improvisation may ing action-orientation. For instance, at low levels of
have detrimental effects on sales performance. Being a action-orientation, the sales activity will not function
descriptive choice process by nature means that improvi- effectively given a lack of responsiveness to the dyna-
sation exudes bounded rationality. Coupled with this mism of customer needs. On the other hand, at high
fact, high levels of improvisation will likely mean that levels of action-orientation, the decision-maker will pur-
managers rely excessively on fast and frugal decision- sue action over analysis to stay focused on the customer
making processes (e.g., heuristics). As such, given that problem (Nemkova et al., 2015). This may lead to an
bounded rationality implies that sales account managers escalation of commitment and unequivocal belief in what
lack situational clarity, especially in economies experienc- the “solution” can be, rather than in what it should
ing significant transitions as in Ghana, excessive reliance be. Therefore, emphasizing creativity, spontaneity, and
on fast and frugal decision-making processes action-orientation at their respective median levels
(e.g., heuristics) means that managers may fall too short (as opposed to low and high levels), is expected to boost
in terms of expending the needed effort to engage in situ- firm-level sales performance. Thus, we argue that
ations enough to boost their clarity. As such, a high
degree of improvisation in emergent markets may gener- H1. The dimensions of improvisation have
ate ineffective choices, which is detrimental for sales inverted U-shaped relationships with sales
performance. performance: such that the impacts of
Moreover, because improvisation itself is disruptive (a) creativity, (b) spontaneity, and (c) action-
of established routines with which customers are familiar orientation on sales performance are low for
and comfortable, it is also surrounded by uncertainty low levels of those dimensions of improvisa-
(Ferguson, 2009). According to Mueller et al. (2011), tion, the impacts then increase at moderate
such a situation can be unsettling for customers and may levels and then go down again for high levels
produce undesired outcomes when customers see the of those dimensions.
company’s offerings being high risk (Nemkova
et al., 2015). Nevertheless, the absence of improvisation,
or improvisation at low levels, may inhibit the develop- Moderating roles of individual agency and
ment of agility within firm decision-making, the latter self-efficacy
being a necessity for firms operating in high resource-
constrained and precarious environments (Souchon Given that improvisation is a descriptive choice behavior,
et al., 2016), such as those found in Ghana. Collectively, we identify two psychological processes that may explain
the arguments put forward in the above discussion the boundary condition of the relationship between the
(as well as extant empirical evidence) suggest that the dimensions of the improvisation and sales performance.
impact of improvisation on sales performance is low for Prior research suggests that the environment and individ-
low levels of improvisation, it then increases at moderate ual characteristics shape individual actions while the lat-
levels of improvisation and then goes down again for ter also, in turn, affects the environment and the
high levels of improvisation, indicating the existence of a individual (Bandura, 1989). Thus, conditions of the sell-
nonlinear (i.e., curvilinear) relationship between improvi- ing situation may place explain the extent to which
sation and sales performance. improvisational behavior is enacted by sales account
In line with this study’s objective of exploring the managers and the consequences of the behavior. Such
microfoundation of the relationship between improvisa- conditions interact with person-based characteristics in
tion and sales performance, therefore, we argue that cur- determining the actual enacted behaviors. In turn, the
vilinear relationship are also expected at the dimensional result of the chosen behavior affects firm-level sales out-
level. For instance, several researchers have tested the comes and future client interactions, as well as the future
creativity–performance relationship and confirmed that it success and survival of the firm (Hultman et al., 2019).
is beneficial to sales organizations (Agnihotri et al., 2013; Bandura (1982) highlights two psychological pro-
Lassk & Shepherd, 2013; Proctor, 1991; Strutton cesses as key in this bidirectional interaction between
et al., 2009), while others point to conflicting and incon- environment, individual, and behavior: individual agency
clusive findings regarding creativity’s impact (e.g., Gong and self-efficacy. Agency drives intentional behaviors
et al., 2013; Martinaityte & Sacramento, 2013). Similarly, through goal setting and behavior persistence until they
spontaneity has been observed to lead to positive firm are achieved (Bandura, 1989; Ling & Dale, 2013).
 17404762, 2022, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/emre.12535 by University of Ghana - Accra, Wiley Online Library on [19/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
HULTMAN ET AL. 421
Self-efficacy, on the other hand, refers to judgments of situations, high-agency might dispose sales account man-
one’s capability to effectively respond to given situations. agers to a selling orientation that aligns to their personal
It has significant implications for whether a given choice is goals rather than a customer orientation, at the expense
made and acted upon (Maddux, 2016). Both self-efficacy of firm performance (e.g., Boles et al., 2001). Thus,
and individual agency are likely to alter the hypothesized
quadratic relationships between creativity, spontaneity H2. The inverted U-shaped relationships
and action-orientation, and sales performance. between (a) creativity, (b) spontaneity, and
(c) action-orientation, and sales performance
is steepened (more pronounced) as individual
Individual agency agency increases.
Individual agency is defined as a drive to shape one’s life
outcomes (Bandura, 2001). It depicts individuals as not Self-efficacy
mere passive experiencers but also as driven to generate
actions that contribute to their desired life outcomes Bandura (1982) defines self-efficacy as judgments about
(Bandura, 2001; Ling & Dale, 2013). Agency manifests people’s own capability to effectively respond to situa-
as intentionality, enabling anticipatory behavioral con- tions (see also Akhtar, 2008). Far from actual skills held
trol and a commitment to making things happen. It for effective response, efficacy operates at the evaluative
enables persistence such that high-agency individuals level to prescribe to individuals a sense of what they can
stick to the predefined behaviors they believe would lead and cannot handle. The strength of the resulting convic-
to personal goal achievement. Agency also manifests as tion defines whether they will even try to cope with given
self-reflectiveness, an ongoing metacognitive evaluation situations (Bandura, 1997, 1977). Consequently, where
of one’s actions towards intended outcomes. This ensures people perceive the demands of a situation to exceed their
that individuals constantly examine the extent of fit capabilities, they resort to avoidance behaviors. Whereas,
between their choices and their desired outcomes in contrast, when judging themselves capable of compe-
(Bandura, 2001, 2006). tently handling a situation they behave decisively
While ordinarily, high-agency should make sales (Maddux, 2016).
managers’ behaviors more purposeful, in the specific con- Self-efficacy has been widely proposed as a strong
text of achieving situational responsiveness, it appears to driver of employee performance as demonstrated by
be counterproductive. Inherent in individual agency is meta-analytic evidence (e.g., Stajkovic & Luthans, 1998).
the tendency to ignore situational nuances (such as the Within sales research, similar evidence is proposed by
complexity of operating in precarious and challenging Brown et al. (1998) and Wang and Netemeyer (2002).
environments as such the emergent in Ghana), a condi- However, our interest in self-efficacy lies beyond its direct
tion that renders high-agency people to sometimes appear implications for sales performance. We posit that self-
ruthless and self-centered (Bagozzi et al., 2010). While efficacy has an attenuating effect on the hypothesized
engaging in creative, spontaneous, and action-oriented quadratic (inverted U-shaped relationship) link
behaviors, high-agency managers may be too preoccu- between the improvisation dimensions of creativity,
pied with their choice-predefined outcome fit resulting in spontaneity, and action-orientation, and firm-level sales
a neglect of situational requirements and customer needs performance.
(Bagozzi et al., 2010). Agency might reduce the task Specifically, whereas high levels of improvisation as
focus which is instrumental in boosting performance by captured across creativity, spontaneity, and action-
directing attention (Brown et al., 2005), to the extent that orientation in decision-making may be detrimental
this creates “off-task thoughts” (Martin & Tesser, 1996, (in that they are inherently disruptive and surrounded by
p. 12). In addition, where situations demand deviation uncertainty), a strong sense of efficacy will likely counter
from predefined structures, this agentic self-reflective ten- the diminishing effects of these improvisational behav-
dency may generate negative affective consequences that iors. Efficacy enables individuals to “focus their attention
interfere with successful selling. and motivation on the tasks necessary for achieving
Hence, a strong sense of agency is expected to further targeted performance levels and persevere in the face of
weaken the hypothesized diminishing effects of high- difficulties” (Brown et al., 2005, p. 974); thus, account
improvisation. Because creativity, spontaneity, and managers deploying creative solutions, spontaneous deci-
action-orientation suffer as a function of bounded ratio- sions, and real-time focus on customer needs would
nality and cognitive limitations, the tendency for high- invest the needed effort to attain situational clarity
agency managers to overlook situational demands (Wang & Netemeyer, 2002). The prevalence of high-
heightens the threat to the benefits of improvisation for efficacy would also imply that sales managers exude
the firm. In other words, high-agency managers may be confidence when applying improvisation in their
too concerned with fitting their responses to personal decision-making. This should make them appear out-
goals instead of attending to customers’ needs. In such wardly credible and reduce customer apprehensions when
 17404762, 2022, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/emre.12535 by University of Ghana - Accra, Wiley Online Library on [19/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
422 HULTMAN ET AL.
presented with untested and unchartered courses of drawn from the original 1472 to receive invitation letters.
action (Krishnan et al., 2002). Therefore, The letters were addressed to the divisional heads
requesting them to introduce the study to their respective
H3. The inverted U-shaped relationship industrial (or business-to-business) account managers/
between (a) creativity, (b) spontaneity, and directors. A trained researcher subsequently administered
(c) action-orientation and sales performance structured questionnaires in person. A total of 224 com-
is flattened (less pronounced) as self-efficacy pleted surveys were received, yielding a 56% response
increases. rate. Post hoc data screening led to the exclusion of six
responses, leaving an effective sample of 218, with an
average of 3.84 years of industrial selling experience
METHOD (SD = 2.65) at a senior capacity at their current firm.
The sample firms mirror the underlying population
Research context and data collection with regards to industry representation with wholesale
and retail being the largest group (20.2% of the sample)
The study’s hypotheses are tested using data from indus- followed by financial and insurance (18.8%), information
trial senior sales managers and directors in Ghana. and communication (18.3%), manufacturing (17.0%). A
Ghana provides a suitable and important context to test majority of the informants (59.6%) were mainly involved
the applicability of constructs previously examined in with selling services with the remainder selling tangible
developed market settings, thereby, bringing turbulent products (40.4). Most of the sampled firms were domesti-
emerging country insight to the strategic selling dis- cally oriented (63.8%) rather than internationally ori-
course. With more than three decades of experience oper- ented (36.2%), and there were more male (79%) than
ating an open market economy, GDP growth rates female (21%) key informants.
surpassing those of many developed economies, and We controlled for nonresponse bias through
being ranked as the easiest place to do business in West Armstrong and Overton’s (1977) extrapolation method.
Africa (World Bank, 2019), Ghana presents an interest- Specifically, t tests comparing study’s key constructs
ing opportunity to examine improvisation. Despite its between those who agreed to participate after the first
growth trajectory, the Ghanaian economy is still charac- contact and those who needed reminders indicated no
terized by major institutional adversities (e.g., weak law significant differences thus enhancing our confidence in
enforcement) that have the potential to further boost the unbiased data. A series of analysis of variance tests fur-
need for sales managers to exhibit improvisational behav- ther failed to detect any significant differences in neither
iors in their boundary spanning roles (Hultman the independent nor the dependent the study variables
et al., 2019). Additionally, Ghana is largely a small- based on industry, type of product sold, domestic versus
business economy with almost 90% of all economic activ- international orientation, or gender (p ≥ 0.118), thus
ity carried out by small- and medium-sized enterprises ruling out industry, organization, and gender biases.
(SMEs) (Donbesuur et al., 2020). As such, the boundary
spanning sales managers’ and directors’ improvisation
decisions also effectively become their firms’ strategic Measures
decisions (Child, 1972) as opposed to the case in more
mature and consolidated markets where individual man- Multi-item indicators were used to measure all constructs
agers’ decisions carry less weight for overall firm in the model (see Appendix A). The dependent variable,
strategy. sales performance (Cronbach’s α = 0.896), was
Following extant industrial selling literature operationalized as the level of perceived sales achieve-
(e.g., Auh & Menguc, 2013; Baldauf & Cravens, 2002), ment of sales account managers: increasing market share,
we used structured questionnaire to collect data from selling to large volume customers, and expanding share
sales account managers/directors working in industrial of business with major accounts (Miao & Evans, 2014;
firms across a multitude of sectors. We identified 4125 Theodosiou & Katsikea, 2007). Creativity (α = 0.808)
industrial firms (with at least five employees) listed in the measured the extent to which managers tried new
2014 Ghana business directory and Ghana association of approaches to sales problems, aimed at originality in gen-
industries database (Acquaah, 2012). However, given the erating sales solutions, and were inventive in overcoming
heavy concentration of Ghana’s commercial activities in barriers across sales situations (Wang &
few cities (Grant, 2001), and to balance cost of survey Netemeyer, 2004). Spontaneity (α = 0.808) focused on
administration and sample size required to achieve statis- being responsive in the moment, acting spontaneously,
tical power, we limited our sampling to four major com- and responding on the spot impulsively (Vera &
mercial cities (i.e., Accra, Tema, Takoradi, and Kumasi), Crossan, 2005). Action-orientation (α = 0.713) was
thus reducing the sampling frame to 1472 firms. The total adapted from O’Reilly et al. (1991) with four items tap-
sampling frame could not be approached due to research ping the managers’ focus in dealing with upcoming situa-
budget constraints so a random sample of 400 firms was tions and responding by taking action.
 17404762, 2022, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/emre.12535 by University of Ghana - Accra, Wiley Online Library on [19/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
HULTMAN ET AL. 423
Individual agency (α = 0.889) was captured as the similar pattern of significant and nonsignificant correla-
drive to shape life outcomes by persisting in behavior tions after adjustment, CMV is not likely to distort the
towards the achievement of their plans, being conscious study results.
of how events relate to their identity, being conscious of As an additional safeguard, we followed Carson
what they can and cannot handle, and constantly evalu- (2007) to run a combined congeneric measurement model
ating their actions (Bandura, 2001). Self-efficacy by estimating a confirmatory factor analysis (CFA)
(α = 0.882) was also captured at individual level and model for all multi-item scales in the presence of an
measured with a three-item scale capturing respondents’ unmeasured common method factor, modeled to load on
confidence in their ability to perform their job well all observed items. This enabled us to control for some of
(Wang & Netemeyer, 2002). As previous research sug- the variances and covariances that may have been intro-
gests possible confounding effects, we also included adap- duced as a result of obtaining responses from single infor-
tive selling behavior (α = 0.874), customer mants. Carson’s approach also made it possible to
demandingness (α = 0.798), sales account manager expe- recalculate the factor loadings, composite reliability
rience (α = 0.875), compensation type (percentage of sal- (CR), and average variance extracted (AVE) values in
ary versus commissions), and firm size (natural logarithm the presence of the common method factor and respective
of number of firm employees) as controls. error terms. Findings show that all adjusted factor load-
Despite being the most appropriate respondents for ings, CR, and AVE values remained qualitatively
the survey, our data were obtained from single infor- unchanged. Thus, our confidence in the absence of severe
mants in each organization, which can raise the potential CMV is further enhanced.
threat of common method variance (CMV) (Podsakoff
et al., 2003). However, we argue that our conceptual
model is complex enough to minimize any risk of CMV ANALYSIS AND RESULTS
influencing our results (Siemsen et al., 2010). In fact,
Siemsen et al. (2010, p. 456) demonstrate that “quadratic Measure validation
and interaction effects cannot be artefacts of CMV.” Our
study contains both quadratic, interaction, and Confirmatory factor analyses using the elliptical
quadratic-interaction variables, thus, making it extremely reweighted least squares (ERLS) procedure was used in
difficult for the estimated parameters to be inflated by EQS 6.2 to examine reliability and validity of the scales
CMV as it, by nature, causes temporal, proximal, and in the model. The ERLS procedure was specifically cho-
psychological separation of our measurements compared sen due to its ability to produce unbiased parameter esti-
to if the model was based on direct and linear effects mates for both multivariate normal and nonnormal data
only. The complex model specifications also serve to cre- (Sharma et al., 1989). Model fit was assessed using tradi-
ate counterbalanced question ordering which helps neu- tional chi-square (χ 2) difference tests together with sev-
tralize priming effects (Podsakoff et al., 2003). eral approximate fit heuristics. Results show that the
Nonetheless, we followed two statistical procedures to measurement model fit the data well (χ 2/degrees of free-
formally rule out any major CMV influences on our dom [df] = 570.298/507, p = 0.026; root mean squared
results. error of approximation [RMSEA] = 0.024; normed fit
We followed Lindell and Whitney’s (2001) recom- index [NFI] = 0.998; nonnormed fit index [NNFI] =
mendation for accounting for CMV in cross-sectional 0.999; comparative fit index [CFI] = 0.999; standardized
research designs to include a marker variable in the root mean square residual [SRMR] = 0.061). As proof of
study: “There are too many demands on my time,” which convergent validity, all item loadings are significant at
measures role overload as a single item (Banin 1% level, with loadings ranging from 0.63 to 0.94. In
et al., 2016). We initially found that this marker variable addition, CR values for each construct exceed the mini-
is not significantly correlated with any of the study’s vari- mum cut-off point of 0.60 (see Appendix A). We assessed
ables. A further extension of Lindell and Whitney’s discriminant validity using interconstruct correlations,
(2001) involved using the average correlation for the AVE in each construct, as well as the highest shared vari-
marker variable with the other model variables (0.052) to ance (HSV) between construct pairs. All the inter-
compute a CMV-adjusted correlation matrix using the construct correlations fall below the AVE square root in
following equation: evidence of adequate discriminant validity. Table 1 shows
the measurement descriptive statistics and correlations.
rA ¼ðru – rMÞ=ð1 rMÞ,
where rA is the CMV-adjusted correlation, ru is the origi- Hypothesis testing
nal correlation, and rM is the marker variable (Malhotra
et al., 2006). Because the results showed from this exer- To ensure robust and stable results, the approach for test-
cise reveal only small and insignificant differences ing the hypotheses rely on a dual methodology consisting
between the two correlation matrices (Δr ≤ 0.063) and of both structural equation modeling (SEM) and on a
 17404762, 2022, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/emre.12535 by University of Ghana - Accra, Wiley Online Library on [19/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
424 HULTMAN ET AL.
TABLE 1 Interconstruct correlations and descriptive statistics
1 2 3 4 5 6 7 8 9 10 11
1. Firm size -
2. Creativity 0.409 0.719
3. Spontaneity 0.157 0.323 0.823
4. Action-orientation 0.374 0.478 0.581 0.709
5. Adaptability 0.247 0.365 0.106 0.316 0.849
6. Customer demandingness 0.164 0.170 0.260 0.335 0.395 0.757
7. Agency 0.159 0.378 0.110 0.266 0.479 0.325 0.820
8. Self-Efficacy 0.171 0.278 0.106 0.181 0.515 0.236 0.619 0.817
9. Experience 0.067 0.036 0.206 0.040 0.071 0.108 0.208 0.205 0.855
10. Compensation type 0.051 0.039 0.000 0.014 0.059 0.019 0.140 0.008 0.161 -
11. Performance 0.061 0.055 0.244 0.218 0.212 0.171 0.065 0.203 0.035 0.008 0.829
Mean 1.605 5.168 4.776 5.213 5.562 6.155 5.543 5.678 4.540 3.406 5.118
SD 0.186 0.676 0.964 0.683 0.718 0.759 0.683 0.670 2.714 2.120 0.805
Note: Correlations larger than 0.13 are significant at p = 0.05 (two tailed). The square root of AVE is reported on the diagonal in bold.
series of ordinary least squares (OLS) regressions with reveal that the obtained sample of 218 is appropriate and
the expectation that analogous results would enhance the suitable for the tested model (power = 0.999; α err
confidence in our findings. We controlled for multivariate prob = 0.05; df = 192; critical F = 1.564). A
outliers by calculating Mahalanobis distances without multicollinearity test further reveals that the average vari-
detecting any severe concerns (largest Mahalanobis dis- ance inflation factor (VIF) across all estimated models is
tance was 48.282(25 df ); p > 0.001). Thereafter, following 3.236 with the highest VIF being 9.492 (for creativity
2 
established procedures for testing moderated relation- agency) followed by 8.614 (spontaneity2  efficacy) and
ships, a multiplicative approach to structural equation 8.373 (action-orientation2  efficacy). Although these
modeling was adopted (Ping, 1995). Actual estimation of values are in the higher ranges, they do not appear to
the structural model was undertaken using the ERLS show a systematically problematic variable because they
estimation method. We mean-centered the constructs that are spread across different measures and are still below
were used for multiplicative interactive analysis to the generally accepted threshold value of 10.00. Larger
minimize any multicollinearity problem prior to calculat- VIFs are also to be expected given the large number of
ing the loading and error variances of the interaction quadratic and interactive effects estimated
terms. The results imply a reasonably close-fitting (O’Brien, 2007), and the multicollinearity analysis indeed
structural model (χ 2/df = 2138.196/1088; RMSEA = showed considerably lower VIFs among the non-
0.074; NFI = 0.994; NNFI = 0.995; CFI = 0.996; quadratic variables and interaction terms where the
SRMR = 0.075). highest recorded VIF was only 3.619 (for agency in
Table 2 shows the standardized parameter estimates Model 5). An investigation of the residuals from Model
and the directional significance levels for the investigated 5 further reveal that the skewness and kurtosis measures
paths. Table 2 also shows the OLS regression analyses are within the standard error range while the
that were carried out hierarchically in five steps: Model Kolmogorov–Smirnov (KS) test returned nonsignificant
1 included the four control variables. Model 2 added the results (KS = 0.109, p = 0.200) in support of approxi-
direct effects of the three improvisation dimensions and mate data normality. Finally, as the estimates in the
two moderators. Model 3 introduced the squared terms SEM and OLS models display highly similar patterns of
of the improvisation dimensions to test the nonlinear significant and nonsignificant results (Table 2), we are
effects while Model 4 saw the inclusion of all linear inter- able to interpret the findings with a heightened degree of
action terms for control purposes. Finally, Model confidence.
5 included the interactions with nonlinear terms to com-
plete the model and test the remaining hypotheses.
We compared the models by computing R2 difference Results
tests which, apart from Model 4 (which was only added
for control purposes), confirmed that the added effects The study hypothesized a curvilinear (inverted U-shape)
enhanced explanatory power to the original model relationship between the dimensions of improvisation
(p < 0.05). A post hoc power analysis based on the and sales performance. Contrary to expectations, the
achieved effect size for regression Model 5 (f2 = 0.410) results from the first hypothesis show that only
 17404762, 2022, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/emre.12535 by University of Ghana - Accra, Wiley Online Library on [19/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
HULTMAN ET AL. 425
T A B LE 2 Results
Model 1 Model 2 Model 3 Model 4 Model 5 SEM
Beta t Beta t Beta t Beta t Beta t Beta t
Firm Size 0.004 0.056 0.017 0.227 0.023 0.313 0.019 0.301 0.019 0.260
Adaptability 0.175 2.354* 0.121 1.450 0.090 1.095 0.102 1.217 0.096 1.132 0.008 0.110
Customer demandingness 0.108 1.483 0.037 0.483 0.022 0.299 0.008 0.108 0.025 0.325 0.124 1.378
Experience 0.061 0.889 0.021 0.296 0.027 0.398 0.032 0.464 0.031 0.455 0.061 0.826
Compensation type 0.010 0.144 0.015 0.228 0.021 0.320 0.004 0.065 0.011 0.168 0.022 0.319
Creativity 0.135 1.752† 0.253 2.916** 0.163 1.695† 0.120 1.211 0.134 1.121
Spontaneity 0.228 2.532* 0.411 3.895*** 0.327 2.867** 0.351 3.107** 0.332 2.105*
Action-orientation 0.087 0.936 0.009 0.096 0.022 0.226 0.071 0.732 0.077 0.384
Agency 0.099 1.042 0.080 0.858 0.078 0.840 0.050 0.446 0.023 0.181
Self-Efficacy 0.246 2.793** 0.291 3.374** 0.280 3.186** 0.247 2.225* 0.213 2.112*
Creativity2 0.078 1.016 0.061 0.710 0.097 0.921 0.103 1.592
Spontaneity2 0.310 3.276** 0.348 3.211** 0.282 2.544* 0.214 2.913**
Action-orientation2 0.239 2.937** 0.225 2.317* 0.217 2.053* 0.325 4.300***
Creativity  Agency 0.037 0.276 0.124 0.883 0.082 0.838
Spontaneity  Agency 0.001 0.007 0.039 0.236 0.046 0.624
Action-orientation  Agency 0.072 0.495 0.117 0.747 0.113 1.184
Creativity  Efficacy 0.218 1.602 0.147 0.911 0.027 0.313
Spontaneity  Efficacy 0.141 1.124 0.091 0.640 0.031 0.399
Action-orientation  Efficacy 0.132 0.928 0.131 0.908 0.206 1.629
Creativity2  Agency 0.316 1.711† 0.221 3.083**
Spontaneity2  Agency 0.444 2.045* 0.183 2.824**
Action-orientation2  Agency 0.434 2.516* 0.314 4.266***
Creativity2  Efficacy 0.033 0.163 0.106 1.282
Spontaneity2  Efficacy 0.153 0.879 0.049 0.742
Action-orientation2  Efficacy 0.394 2.268* 0.307 3.911***
F 2.585* 3.432*** 4.114*** 3.306*** 3.158***
ΔF(df ) 2.585(5)* 4.087(5)** 5.641(3)** 1.422(6) 2.294(6)*
R 2 (adjusted R 2) 0.057 (0.035) 0.142 (0.101) 0.208 (0.157) 0.240 (0.168) 0.291 (0.199) 0.513
ΔR 2 0.057 0.085 0.066 0.032 0.051
Notes: Fit indices for SEM: χ 2(1121) = 1908.002; NFI = 0.995; NNFI = 0.996; CFI = 0.996; SRMR = 0.047; RMSEA (90% confidence interval) = 0.038 (0.031–0.045).
*p < 0.05. **p < 0.01. ***p < 0.001. †p < 0.10.
 17404762, 2022, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/emre.12535 by University of Ghana - Accra, Wiley Online Library on [19/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
426 HULTMAN ET AL.
action-orientation displays the expected inverted
U-shaped relationship (β 2action-orientation =  325;
p < 0.001). The squared term for creativity is nonsignifi-
cant (β 2creativity = 0.103; p > 0.05), and spontaneity
shows an opposite relationship to what was expected by
displaying a positive and significant coefficient
(β 2spontaneity = 0.214; p < 0.01). The plotted relationships
confirm the statistical results by showing how sales per-
formance is highest at moderate levels of action-
orientation (Figure 2) and lowest at moderate levels of
spontaneity respectively (Figure 3). Therefore, the results F I GURE 4 The moderating effect of sales account manager agencyon creativity
support only H1c while H1a and H1b are rejected.
We further argue in H2 that the inverted U-shaped
relationship between improvisation and performance will
become even more accentuated in cases when managerial
agency is high. The data favors this assumption in the
case of creativity (β 2creativity agency = 0.221; p < 0.01)
and action-orientation (β 2action-orientation agency = 0.314;
p < 0.001) in support of H2a and H2c. Surprisingly, H2b
is not supported as evidenced by the inverse relationship
for the squared spontaneity  agency term on perfor-
mance (β 2spontaneity agency = 0.183; p < 0.01), indicating
that the relationship is U-shaped rather than the other
way around. As plotted in Figures 4 and 5, with rising F I GURE 5 The moderating effect of sales account manager agency
levels of agency, the initially positive relationship on action orientation
between the squared terms of creativity and agency,
respectively, and sales performance turns negative, while
Figure 6 shows the inverse.
Finally, the argument in H3 that a high sense of effi-
cacy positively alters the relationship between creativity
and sales performance is only supported in the case of
F I GURE 6 The moderating effect of sales account manager agency
on spontaneity
F I GURE 2 Quadratic effects of sales account manager action-
orientation
F I GURE 7 The moderating effect of sales account manager self-
efficacy on action orientation
action-orientation (H3c) as evidenced by the significantly
positive coefficient for squared action-orientation  effi-
cacy (β 2action-orientation efficacy = 0.307; p < 0.001) and the
F I GURE 3 Quadratic effects of sales account manager spontaneity surface plot in Figure 7 that shows how the decreasing
 17404762, 2022, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/emre.12535 by University of Ghana - Accra, Wiley Online Library on [19/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
HULTMAN ET AL. 427
effect of sales account manager creativity on sales perfor- Δχ 2(1) = 3.991, p < 0.05). We further estimated an alter-
mance is attenuated when self-efficacy increase in magni- native second-order construct based on the three first-
tude. Conversely, H3a and H3b are not supported as order dimensions and compared it with the original con-
neither of the coefficients are significant ceptualization comprising creativity, spontaneity, and
(β 2creativity efficacy = 0.106; βspontaneityefficacy = 0.049; action-orientation modeled as individual constructs. A
p > 0.05). Taken together, the hypotheses are therefore comparison of fit indices between the two estimations
only fully supported for the action-orientation dimension reveal the superiority of the original conceptualization
of improvisation, while creativity yield mixed results by (second-order construct: χ 2/df = 93.618/51; RMSEA =
finding support in H2a but not H1a and 0.105; NFI = 0.806; NNFI = 0.852; CFI = 0.859;
H1c. Interestingly, spontaneity behaves contrary to SRMR = 0.109; first-order constructs: χ 2/df =
expectations by displaying U-shapes rather than inverse 71.006/51; RMSEA = 0.043; NFI = 0.937; NNFI =
U-shapes when testing H1b and H2b and nonsignificant 0.976; CFI = 0.981; SRMR = 0.068).
results in the case of H3b. To control the robustness of our model and further
ascertain the result pattern obtained in the hypothesis
testing, we employed an alternative analysis method that
Additional analysis involved splitting the three independent variables into
thirds based on percentile distribution, thus creating
To ensure the underlying appropriateness of treating groups of “low,” “medium,” and “high” levels of creativ-
improvisation as a construct comprising of three indepen- ity, spontaneity, and action-orientation, respectively.1
dent dimensions, we conducted an additional discrimi- The direct effects OLS regressions described in Table 2
nant validity control on the basis of a chi-square (Model 2) were then repeated for each level of each inde-
difference test in which the correlations between all possi- pendent variable. The results for the focal variables
ble combinations of creativity, spontaneity, and action- (i.e., creativity, spontaneity, and action-orientation) from
orientation are first freely estimated and thereafter set to this analysis are reported in the first panel of Table 3 and
unity (Gerbing & Anderson, 1988). All chi-square differ- shows how the split group analysis confirm the quadratic
ence tests between the three first-order constructs were regressions by returning analogous results. For instance,
significant, suggesting absence of collinearity and distinc- the effect of creativity on performance is nonsignificant
tiveness of individual study constructs (the smallest dif- at low, medium, and high levels, while action-orientation
ference was between spontaneity and action-orientation; is positively related to performance at low levels and
TABLE 3 Relationship between improvisation dimensions and sales performance at different levels
Dimensions Low Medium High
β (t) Trend β (t) Trend β (t) Trend
A: Main effect relationships between low, medium, and high levels of improvisation dimensions and sales performance
Creativity 0.024 (0.253) ! 0.082 (0.552) ! 0.218 (0.150) !
Spontaneity 0.014 (0.101) ! 0.029 (0.120) ! 0.193 (1.667†) ↗
Action-orientation 0.212 (1.878†) ↗ 0.035 (0.217) ! 0.246 (1.704†) &
B: Relationships at high versus low levels of agency
Creativity Low 0.035 (0.244) ! 0.197 (0.943) ! 0.375 (1.654†) ↗
High 0.070 (0.427) ! 0.251 (1.1450) ! 0.236 (1.385) !
Spontaneity Low 0.080 (0.369) ! 0.052 (0.138) ! 0.087 (0.755) !
High 0.062 (0.330) ! 0.109 (0.267) ! 0.262 (1.896†) ↗
Action-orientation Low 0.236 (1.335) ! 0.087 (0.384) ! 0.042 (0.217) !
High 0.315 (1.781†) ↗ 0.216 (1.488) ! 0.464 (2.360*) &
C: Relationships at high versus low levels of efficacy
Creativity Low 0.164 (1.363) ! 0.166 (1.183) ! 0.023 (0.063) !
High 0.028 (0.177) ! 0.035 (0.101) ! 0.263 (1.531) !
Spontaneity Low 0.059 (0.271) ! 0.158 (0.703) ! 0.042 (0.360) !
High 0.064 (0.328) ! 0.030 (0.084) ! 0.163 (1.169) !
Action-orientation Low 0.308 (1.881†) ↗ 0.106 (0.677) ! 0.549 (2.407*) &
High 0.171 (1.022) ! 0.207 (1.155) ! 0.053 (0.219) !
*p < 0.05. †p < 0.10.
 17404762, 2022, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/emre.12535 by University of Ghana - Accra, Wiley Online Library on [19/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
428 HULTMAN ET AL.
negatively related at high levels (pointing towards a Figure 8 illustrates that sales performance levels are
reverse U-shaped relationship). This corresponds well the highest when the improvisation profile is character-
with both the quadratic regression and SEM coefficients ized by low creativity in conjunction with high
and the plotted quadratic effects in Figures 2 and 3, spontaneity and low action-orientation (Profile 2 in
respectively. Figure 8) followed by low creativity together with low
Following a similar procedure, we also created “low” spontaneity and high action-orientation (Profile 3). The
and “high” groups of the two proposed moderators nonsignificant difference between these two profiles indi-
(agency and efficacy) using a median-split approach and cates that there is no one-size-fits-all improvisation style
then repeated the previously described series of subgroup for maximum performance and that there appears to be
analyses. Again, the pattern of results is in line with those an element of equifinality with regard to performance
previously obtained in the main analysis, providing fur- outcomes for different improvisation profiles. Figure 8
ther confidence in the robustness and stability of the also shows that there is no significant difference between
study findings. This alternative analysis may also facili- high and low creativity for the remaining two improvi-
tate the interpretation of the results by, for instance, sation profiles (Profiles 1 and 4, respectively).
clearly indicating that the action-orientation aspect of To capture the moderating effects of agency and self-
strategic improvisation appears more detrimental for pos- efficacy, the overall sample was divided into high and
itive performance outcomes under conditions of high low levels of the two moderator variables following a
agency (β = 0.464, p < 0.05; Table 3, Panel B) and low median-split approach, and the three-way regression
efficacy (β = 0.549, p < 0.05; Table 3, Panel C), respec- models were repeated on each of the subsamples. The
tively, as evidenced by the negative and significant regres- analysis produced significant results only at high levels of
sion coefficients. agency (adjusted R2 = 0.084, F = 2.492, p < 0.05) and
Although the general study results clearly evidence high levels of self-efficacy (adjusted R2 = 0.111,
the importance of decomposing the strategic improvisa- F = 2.993, p < 0.05), respectively. Both the significant
tion constructs into its dimensions to discern their indi- subgroup analyses also displayed a similar pattern of
vidual effect, it may be of interest to investigate how results as the full dataset with nonsignificant simple
the dimensions interact and determine what combina- interactions but significant three-way coefficients
tion of dimensions (i.e., strategic improvisation profiles) (β = 0.330, p < 0.05 for the high-agency group and
are associated with the highest sales performance β = 0.317, p < 0.1 for the high self-efficacy group).
levels.2 To this end, interaction terms for all possible Figure 9 shows the plot for the significant interaction in
combinations of dimensions were computed and there- the high-agency condition, and it appears to enhance
after regressed against the sales performance construct. the effects for Profiles 2 and 3, respectively, at both high
The results reveal a significant model (adjusted and low levels of creativity, whereas the significant plot
R2 = 0.070, F = 3.317, p < 0.05), and although all the for the high self-efficacy condition in Figure 10 seems to
possible two-way interactions were nonsignificant, the further enhance the positive performance of Profile
three-way interaction between creativity, spontaneity, 2 during low levels of creativity while further dampen-
and action-orientation produced a positive and signifi- ing the performance from Profile 3 during high levels of
cant coefficient (β = 0.234, p < 0.05). To visualize the creativity. Taken together, these results further commu-
three-way interaction, we plotted (see Figure 8) the rela- nicate complexity of the strategic improvisation con-
tionship between creativity and sales performance dur- struct and the perils of studying improvisation in an
ing high and low levels of spontaneity and action- overly simplified manner without decomposing it to
orientation. individual constituents.
F I GURE 8 Three-way interaction effect of strategic improvisation F I GURE 9 Three-way interaction effect of strategic improvisation
dimensions on sales performance dimensions on sales performance at high levels of agency
 17404762, 2022, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/emre.12535 by University of Ghana - Accra, Wiley Online Library on [19/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
HULTMAN ET AL. 429
problems enables the sales account manager to address
the problems rapidly to the delight of the customer and
as a result generate sales performance returns. Finally,
there is no apparent direct relationship between creativity
and sales performance, regardless of its intensity among
sales managers.
Following decision theory logic and in viewing impro-
visation as descriptive and emergent choice behavior, we
tested the moderating effects of individual agency and
self-efficacy. First, the findings for agency suggest that
the hypothesized relationship between the squared term
of action-orientation and sales performance is strength-
F I GURE 1 0 Three-way interaction effect of strategic
improvisation dimensions on sales performance at high levels of self- ened under moderate levels of individual agency. How-
efficacy ever, the relationship between creativity and sales
performance is contingent on the prevalence of individual
agency, going from nonsignificant to significant, while
DISCUSSION the U-shaped relationship between the squared term of
spontaneity and sales performance remains and is, in
Contributions to theory fact, strengthened by agency. Thus, for action-orientation
and creativity, sales performance will benefit when sales
Given the mixed empirical evidence of the benefits of account managers are confident in their ability to deliver
improvisation and repeated assertions that improvisation to the demands of a given emergent sales situation (high-
may be fraught with challenges, this study set out to efficacy) and when they have a stronger sense of personal
examine whether dimensions of improvisation display agency.
curvilinear relationships with firm-level perceived sales While the agentic psychological tendency to consider
performance (thereafter referred to as sales performance). predefined behavioral scripts and personal goals can
We specifically sought to examine the occurrence of crea- strengthen the relationships with sales performance for
tivity, spontaneity, and action-orientation and whether both action-orientation and creativity, at high levels and
the prevalence of these behaviors have unintended conse- without recourse to contextual requirements, the impact
quences for firm sales performance. on sales performance becomes negative. As suggested by
Concerning the occurrence of improvisation at the Bagozzi et al. (2010), agency tendencies may predispose
dimensional level and the relationships therein with sales individuals to selfishness and ruthlessness, both of which
performance, dramatic differences are observed. The may be harmful to their customer relationships. Yet indi-
findings confirm that action-orientation on behalf of sales vidual agency does increase the positive spontaneity–sales
managers has an inverted U-shaped relationship with performance relationship, given the U-shaped quadratic
sales performance. Thus, low and high levels of action- relationship uncovered.
orientation are negatively related to sales performance, Second, concerning self-efficacy, the decision-making
while average levels are positively related to sales perfor- perspective and cognition literature in general suggest
mance; consistent with the logic presented in hypothesis that the ability to rely on context and intuitive frames in
development. Contrary to the hypothesized negative generating solutions should enhance sales performance
quadratic effect, the findings point towards a regular (Gaglio & Katz, 2001). According to Bandura (1997,
U-shaped association between spontaneity and sales per- 1977), self-efficacy is both inhibitory and facilitating of
formance implying that both low and high degrees of choice and action: it drives action in any given situation
spontaneity are more strongly related to performance as and provides the necessary impetus to persist in the face
opposed to the hypothesized medium level. A possible of challenges. At the extreme, high-efficacy implies that
explanation to this finding is that the inefficiencies and people become imbued with confidence to persist in exe-
aligned costs that mid-levels of spontaneity might incur cuting their intuition (Brown et al., 2005). To explain
relative to low-spontaneity may decrease performance how efficacy might strengthen the relationship between
levels. Similarly, sales performance may go up in action-orientation and sales performance, we draw on
response to low-spontaneity because it involves incursion insights from Wang and Netemeyer (2002) and reason
of risk minimization in approaches used to solving sales that because efficacious account managers would invest
problems. In other words, it is possible that under low the needed effort to attain situational clarity and apply
levels of spontaneity, sales account managers resort to their action-oriented solutions with confidence and per-
established procedures when addressing customer prob- sistence, they are likely to appear credible in customer
lems. On the other hand, a high degree of spontaneity interactions, thus reducing any customer apprehensions
may produce increases in performance because the appli- with untested action-driven decisions (Krishnan
cation of highly spontaneous solutions to customer et al., 2002). Yet at very high levels of efficacy, the
 17404762, 2022, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/emre.12535 by University of Ghana - Accra, Wiley Online Library on [19/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
430 HULTMAN ET AL.
positive relationship between the squared terms for LIMITATIONS AND DIRECTIONS FOR
action-orientation and sales performance assume a nega- FUTURE RESEARCH
tive relationship. Curiously, this significant effect is not
observed for the other two improvisation dimensions. This study suffers limitations as a result of its cross-
sectional design and as such, it is impossible to make
causal inferences from the findings. Given this limitation,
Implications for improvisation practice in we suggest that future research should extend this study’s
emerging markets findings by tracking the relationships explored over time
to establish the extent of their stability. Second, the use of
The occurrence of improvisation and its ability to influ- perceptual sales performance measures, which although
ence and shape strategic outcomes remains neglected did not pose a method bias concern in the study, is a chal-
within emerging market contexts, despite such settings lenge. In particular, it may be argued that a more objec-
offering opportunities to uncover a new understanding of tive means of assessing sales performance using actual
the origins and outcomes of strategic improvisation under sales data may help explain the actual sales value the
typically highly turbulent and fast-paced business organizations derived from their sales account managers’
conditions. improvisation. While acknowledging this limitation, it is
The study findings accomplish just this and point to important to note that such figures were not readily avail-
an unusual optimal combination of improvisation dimen- able from the organizations from which the sample was
sions for firm-level sales performance returns. Specifi- drawn. We therefore suggest that, where possible, future
cally, effective improvisation occurrence for sales research should make efforts to obtain such objective
performance gains takes the form of very low or very sales figures on individual sales account managers to fur-
high action-orientation coupled with moderate spontane- ther validate the findings attained. Third, although we
ity; irrespective of the degree of creativity which contra- established that Ghana-based industrial sales managers
dicts suggestions that creativity is a universally beneficial constitute a suitable respondent base for studying strate-
behavior for sales organizations (e.g., Agnihotri gic improvisation in sales, we recognize the limitation of
et al., 2013). This nuanced depiction of improvisation a single country emerging economy context. It would
sheds a whole new light on the capacity for firms to oper- therefore be interesting to expand the scope of the current
ate and adapt to fast changing and unpredictable envi- investigation across different markets with different levels
ronments to ensure economic survival. By disaggregating of economic development and institutional environments.
the relevance of improvisation at the dimensional level, A cross-country study of this study’s investigated rela-
the findings demonstrate the form improvisation should tionships would therefore be a natural future extension.
take if it is to become an impactful tool in emerging Fourth, a future study would benefit from examining the
markets. Added complexity is uncovered however when moderating effect of initiative (Eissa, 2020) alongside
considering the role of individual agency. Based on the individual agency. Initiative is a voluntary behavior that
findings, high-level account managers may need to tone exceeds prescribed duties. As such, it may strengthen the
down their “in-behavior” self-evaluation and self- effects of improvisation on performance in salespeople.
critique when engaging in real-time action-orientation Fifth, while considered as moderating variables in the
and creative behaviors to satisfy customers. That said, improvisation–performance relationship, and therefore
should sales executive display a greater propensity for tested for linear effects (following hierarchical procedures
spontaneity, individual agency increases the positive for moderation testing), it may be the case that individual
spontaneity–sales performance relationship given the U- agency and self-efficacy themselves affect performance
shaped quadratic relationship uncovered. Thus, prescrip- quadratically. Specifically, both the drive for achieving
tions for SMEs in emerging markets such as Ghana desired outcomes (agency) and confidence in one’s own
need to pay close attention to the multidimensionality ability (self-efficacy) are likely to lead to objective fulfill-
of improvisation and build organizational profiles to ment only up to a point of diminishing returns. Excessive
determine what the optimal combinations for superior agency and self-efficacy may be manifested in salesperson
firm performance. preoccupation with personal success at the expense of
In conclusion, unlike previous research that portrays focus on the task at hand (e.g., listening to customer
improvisation as a positive behavioral manifestation, and needs, being responsive to customer requests). A future
in extending prior research that suggests that improvisa- study may therefore consider the quadratic effects of
tion might not always be a good or a bad behavior, this these variables on performance. Finally, one cannot rule
study provides empirical evidence to demonstrate that out alternative explanations for the findings attained, as
the dimensions of improvisation contribute uniquely to there may be other variables not included in this research
sales performance and their effects are inhibited or facili- which contribute to the findings attained. As such, future
tated at varying levels of sales managers’ agency and self- research should build on the current study and to seek to
efficacy. explore alternative explanations for the findings of this
 17404762, 2022, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/emre.12535 by University of Ghana - Accra, Wiley Online Library on [19/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
HULTMAN ET AL. 431
research, for instance, via collecting data on additional Banin, A.Y., Boso, N., Hultman, M., Souchon, A.L., Hughes, P. &
performance predictors. Nemkova, E. (2016) Salesperson improvisation: Antecedents, per-
formance outcomes, and boundary conditions. Industrial Market-
ing Management, 58, 12–130.
ORCID Baron, R.A. & Ensley, M.D. (2006) Opportunity recognition as the
Magnus Hultman https://orcid.org/0000-0003-1771- detection of meaningful patterns: Evidence from comparisons of
8898 novice and experienced entrepreneurs. Management Science, 52(9),
Paul Hughes https://orcid.org/0000-0002-7164-0406 1331–1344. https://doi.org/10.1287/mnsc.1060.0538
Bell, D.E., Raiffa, H. & Tversky, A. (1988) Decision-making: Descrip-
tive, normative and prescriptive interactions. Cambridge: Cam-
ENDNOTES bridge University Press.
1 We thank an anonymous reviewer for this helpful suggestion. Bergh, D.D. & Lim, E.N.K. (2008) Learning how to restructure:
2 We thank an anonymous reviewer for this helpful suggestion. absorptive capacity and improvisational views of restructuring
actions and performance. Strategic Management Journal, 29(6),
593–616. https://doi.org/10.1002/smj.676
Boles, J.S., Babin, B.J., Brashear, T.G. & Brooks, C. (2001) An exami-
nation of the relationships between retail work environments, sales
REFERENCES account manager selling orientation-customer orientation and job
Acquaah, M. (2012) Social networking relationships, firm-specific man- performance. Journal of Marketing Theory and Practice, 9(3),
agerial experience and firm performance in a transition economy: 1–13. https://doi.org/10.1080/10696679.2001.11501893
A comparative analysis of family owned and nonfamily firms. Bonney, F.L. & Williams, B.C. (2009) From products to solutions: the
Strategic Management Journal, 33(10), 1215–1228. https://doi.org/ role of sales account manager opportunity recognition. European
10.1002/smj.1973 Journal of Marketing, 43(7/8), 1032–1052. https://doi.org/10.1108/
Agnihotri, R., Rapp, A.A., Andzulis, J.M. & Gabler, C.B. (2013) 03090560910961506
Examining the drivers and performance implications of boundary Brown, S.P., Cron, W.L. & Slocum, J.W., Jr. (1998) Effects of trait
spanner creativity. Journal of Service Research, 17, 164–181. competitiveness and perceived intraorganizational competition on
https://doi.org/10.1177/1094670513506494 sales account manager goal setting and performance. The Journal
Akhtar, M. (2008) What is self-efficacy? Banduras 4 sources of efficacy of Marketing, 62(4), 88–98. https://doi.org/10.1177/
beliefs. UK: Positive Psychology. 002224299806200407
Anderson, R.E. & Huang, W.Y.R. (2006) Empowering salespeople: Brown, S.P., Eli, J. & Thomas, W.L. (2005) The attenuating effect of
Personal, managerial, and organizational perspectives. role overload on relationships linking self-efficacy and goal level
Psychology and Marketing, 23, 139–159. https://doi.org/10.1002/ to work performance. Journal of Applied Psychology, 90(5), 972–
mar.20104 979. https://doi.org/10.1037/0021-9010.90.5.972
Armstrong, J.S. & Overton, T.S. (1977) Estimating nonresponse bias in Cabantous, L. & Gond, J.P. (2011) Rational decision making as perfor-
mail surveys. Journal of Marketing Research, 14, 396–402. https:// mative praxis: Explaining rationality’s E
ternel Retour. Organiza-
doi.org/10.1177/002224377701400320 tion Science, 22(3), 573–586. https://doi.org/10.1287/orsc.1100.
Auh, S. & Menguc, B. (2013) Knowledge sharing behaviors of 0534
industrial salespeople: An integration of economic, social Cao, G., Duan, Y. & Cadden, T. (2019) The link between information
psychological, and sociological perspectives. European Journal of processing capability and competitive advantage mediated through
Marketing, 47(8), 1333–1355. https://doi.org/10.1108/ decision-making effectiveness. International Journal of Information
03090561311324354 Management, 44, 121–131. https://doi.org/10.1016/j.ijinfomgt.
Bagozzi, R.P., Belschak, F. & Verbeke, W. (2010) The role of emotional 2018.10.003
wisdom in sales account managers’ relationships with colleagues Carson, S.J. (2007) When to give up control of outsourced new product
and customers. Psychology and Marketing, 27(11), 1001–1031. development. Journal of Marketing, 71(1), 49–66. https://doi.org/
https://doi.org/10.1002/mar.20370 10.1509/jmkg.71.1.49
Baldauf, A. & Cravens, D.W. (2002) The effect of moderators on the Child, J. (1972) Organizational structure, environment and perfor-
sales account manager behavior performance and sales account mance: The role of strategic choice. Sociology, 6(1), 1–22. https://
manager outcome performance and sales organization effective- doi.org/10.1177/003803857200600101
ness relationships. European Journal of Marketing, 36(11/12), Ciuchta, M.P., O’Toole, J. & Miner, A.S. (2021) The organizational
1367–1388. https://doi.org/10.1108/03090560210445227 improvisation landscape: Taking stock and looking forward. Jour-
Bandura, A. (1977) Self-efficacy: Toward a unifying theory of behav- nal of Management, 47(1), 288–316. https://doi.org/10.1177/
ioural change. Psychological Review, 84, 191–215. https://doi.org/ 0149206320966987
10.1037/0033-295X.84.2.191 Corbett, J., Webster, J. & Jenkin, T.A. (2018) Unmasking corporate
Bandura, A. (1982) Self-efficacy mechanism in human agency. sustainability at the project level: Exploring the influence of insti-
American Psychologist, 37(2), 122–147. https://doi.org/10.1037/ tutional logics and individual agency. Journal of Business Ethics,
0003-066X.37.2.122 147(2), 261–286. https://doi.org/10.1007/s10551-015-2945-1
Bandura, A. (1989) Human agency in social cognitive theory. American Crossan, M., Cunha, M.P.E., Vera, D. & Cunha, J. (2005) Time and
Psychologist, 44(9), 1175–1184. https://doi.org/10.1037/0003-066X. organizational improvisation. Academy of Management Review,
44.9.1175 30, 129–145. https://doi.org/10.5465/amr.2005.15281441
Bandura, A. (1997) Self-efficacy: The exercise of control. New York: W. Cunha, M.P., Cunha, J.V. & Kamoche, K.N. (2002) Organizational
H. Freeman, pp. 1–606. improvisation: What, when, how and why. In: Cunha, M.P.,
Bandura, A. (2001) Social cognitive theory: An agentic perspective. Cunha, J.V. & Kamoche, K.N. (Eds.) Organizational Improvisa-
Annual Review of Psychology, 52, 1–26. https://doi.org/10.1146/ tion. London: Routledge, pp. 96–137.
annurev.psych.52.1.1 Cunha, M.P., Miner, A.S. & Antonacopoulou, E. (2016) Improvisation
Bandura, A. (2006) Towards a psychology of human agency. Perspec- processes in organizations. In: Langley, A. & Tsoukas, H. (Eds.)
tives on Psychological Science, 1, 164–180. https://doi.org/10.1111/ The Sage Handbook of Process Organization Studies. London:
j.1745-6916.2006.00011.x Sage.
 17404762, 2022, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/emre.12535 by University of Ghana - Accra, Wiley Online Library on [19/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
432 HULTMAN ET AL.
Donbesuur, F., Boso, N. & Hultman, M. (2020) The effect of entrepre- improvisation Readiness Index Score (IRIS) to survive, adapt, and
neurial orientation on new venture performance: Contingency thrive in a crisis. Industrial Marketing Management, 88, 485–499.
roles of entrepreneurial actions. Journal of Business Research, 118, https://doi.org/10.1016/j.indmarman.2020.05.020
150–161. https://doi.org/10.1016/j.jbusres.2020.06.042 Hughes, P., Souchon, A.L., Nemkova, E., Hodgkinson, I.R.,
Eisenhardt, K. & Brown, S.L. (1998) Competing on the edge: Strategy Oliveira, J.S., Boso, N., et al. (2018) Quadratic effects of dynamic
as structured chaos. Long Range Planning, 31, 786–789. https:// decision-making capability on innovation orientation and perfor-
doi.org/10.1016/S0024-6301(98)00092-2 mance: Evidence from Chinese exporters. Industrial Marketing
Eissa, G. (2020) Individual initiative and burnout as antecedents of Management, 59, 120–130. https://doi.org/10.1016/j.indmarman.
employee expediency and the moderating role of conscientious- 2018.11.005
ness. Journal of Business Research, 110, 202–212. https://doi.org/ Hultman, M., Yeboah-Banin, A.A. & Boso, N. (2019) Linking improvi-
10.1016/j.jbusres.2019.12.047 sational behavior to customer satisfaction: the roles of relationship
Eling, K., Griffin, A. & Langerak, F. (2014) Using intuition in fuzzy cordiality and customer exercised coercive power. The Journal of
front-end decision-making: A conceptual framework. Journal of Business and Industrial Marketing, 34(6), 1183–1193. https://doi.
Product Innovation Management, 31(5), 956–972. https://doi.org/ org/10.1108/JBIM-11-2017-0298
10.1111/jpim.12136 Kamoche, K. & Cunha, M.P.E. (2001) Minimal structures: From jazz
Etzioni, A. (2014) Humble decision-making theory. Public Management improvisation to product innovation. Organization Studies, 22(5),
Review, 16(5), 611–619. https://doi.org/10.1080/14719037.2013. 733–764. https://doi.org/10.1177/0170840601225001
875392 Kamoche, K., Cunha, M.P.E. & Cunha, J.V.D. (2003) Towards a the-
Evans, K.R., McFarland, R.G., Dietz, B. & Jaramillo, F. (2012) ory of organizational improvisation: Looking beyond the jazz met-
Advancing sales performance research: A focus on five under aphor. Journal of Management Studies, 40(8), 2023–2051. https://
researched topic areas. Journal of Personal Selling and Sales Man- doi.org/10.1046/j.1467-6486.2003.00410.x
agement, 32, 89–105. https://doi.org/10.2753/PSS0885-3134320108 Kim, M., Sudhir, K., Uetake, K. & Canales, R. (2019) When salespeo-
Ferguson, B. (2009) Creativity and integrity: Marketing the “in develop- ple manage customer relationships: Multidimensional incentives
ment” screenplay. Psychology and Marketing, 26(5), 421–444. and private information. Journal of Marketing Research, 56,
https://doi.org/10.1002/mar.20281 749–766. https://doi.org/10.1177/0022243719847661
Ford, C. & Sullivan, D.M. (2004) A time for everything: How the Krishnan, B.C., Netemeyer, R.G. & Boles, J.S. (2002) Self-efficacy,
timing of novel contributions influences project team outcomes. competitiveness, and effort as antecedents of sales account man-
Journal of Organizational Behavior, 25, 279–292. https://doi.org/ ager performance. Journal of Personal Selling and Sales Manage-
10.1002/job.241 ment, 22, 285–295.
Gaglio, C.M. & Katz, J.A. (2001) The psychological basis of opportu- Lassk, F.G. & Shepherd, C.D. (2013) Exploring the relationship
nity identification: Entrepreneurial alertness. Small Business Eco- between emotional intelligence and sales account manager creativ-
nomics, 16, 95–111. https://doi.org/10.1023/A:1011132102464 ity. Journal of Personal Selling and Sales Management, 33(1),
Gerbing, D.W. & Anderson, J.C. (1988) An updated paradigm for scale 25–37. https://doi.org/10.2753/PSS0885-3134330103
development incorporating unidimensionality and its assessment. Lawrence, T.B. & Suddaby, R. (2006) Institutions and institutional
Journal of Marketing Research, 25, 186–192. https://doi.org/10. work. In: Clegg, S.R., Hardy, C., Lawrence, T.B. & Nord, W.R.
1177/002224378802500207 (Eds.) Handbook of organization studies, 2nd edition. London:
Goldstein, D.G. & Gigerenzer, G. (2002) Models of ecological rational- Sage.
ity: the recognition heuristic. Psychological Review, 109(1), 75–90. Leybourne, S. & Sadler-Smith, E. (2006) The role of intuition and
https://doi.org/10.1037/0033-295X.109.1.75 improvisation in project management. International Journal of
Gong, Y., Zhou, J. & Chang, S. (2013) Core knowledge employee crea- Project Management, 24(6), 483–492. https://doi.org/10.1016/j.
tivity and firm performance: The moderating role of riskiness ori- ijproman.2006.03.007
entation, firm size, and realized absorptive capacity. Personnel Lindell, M.K. & Whitney, D.J. (2001) Accounting for common method
Psychology, 66, 443–482. https://doi.org/10.1111/peps.12024 variance in cross-sectional research designs. Journal of Applied
Grant, R. (2001) Liberalization policies and foreign companies in Psychology, 86, 114–121. https://doi.org/10.1037/0021-9010.86.
Accra, Ghana. Environment and Planning, 33(6), 997–1014. https:// 1.114
doi.org/10.1068/a33144 Ling, C. & Dale, A. (2013) Agency and social capital: characteristics
Haley, U.C. & Stumpf, S.A. (1989) Cognitive trails in strategic decision- and dynamics. Community Development Journal, 49(1), 4–20.
making: linking theories of personalities and cognitions. Journal of https://doi.org/10.1093/cdj/bss069
Management Studies, 26(5), 477–497. https://doi.org/10.1111/j. Locander, D.A., Mulki, J.P. & Weinberg, F.J. (2014) How do salespeo-
1467-6486.1989.tb00740.x ple make decisions? The role of emotions and deliberation on
Hmieleski, K.M., Corbett, A.C. & Baron, R.A. (2013) Entrepreneurs adaptive selling, and the moderating role of intuition. Psychology
improvisational Behavior and firm performance: A study of dispo- and Marketing, 31(6), 387–403. https://doi.org/10.1002/mar.20702
sitional and environmental moderators. Strategic Entrepreneurship Maddux, J.E. (2016) In: Trusz, S. & Babel, P. (Eds.) Self efficacy. Inter-
Journal, 7, 138–150. https://doi.org/10.1002/sej.1143 personal and intrapersonal expectancies. London: Routledge.
Hodgkinson, G.P. & Sadler-Smith, E. (2018) The dynamics of intuition Malhotra, N.K., Kim, S.S. & Patil, A. (2006) Common method vari-
and analysis in managerial and organizational decision making. ance in IS research: A comparison of alternative approaches and a
Academy of Management Perspectives, 32, 473–492. https://doi. reanalysis of past research. Management Science, 52(12),
org/10.5465/amp.2016.0140 1865–1883. https://doi.org/10.1287/mnsc.1060.0597
Hodgkinson, G.P., Sadler-Smith, E., Burke, L.A., Claxton, G. & Martin, L. & Tesser, A. (1996) Some ruminative thoughts. In: Wyer, R.
Sparrow, P.R. (2009) Intuition in organizations: Implications for S., Jr. (Ed.) Ruminative thoughts: Advances in social cognition.
strategic management. Long Range Planning, 42(3), 277–297. Hillsdale: Erlbaum, pp. 1–48.
https://doi.org/10.1016/j.lrp.2009.05.003 Martinaityte, I. & Sacramento, C.A. (2013) When creativity enhances
Homburg, C., Grozdanovic, M. & Klarmann, M. (2007) Responsive- sales effectiveness: The moderating role of leader–member
ness to customers and competitors: the role of affective and cogni- exchange. Journal of Organizational Behavior, 34, 974–994. https://
tive organizational systems. Journal of Marketing, 71, 18–38. doi.org/10.1002/job.1835
https://doi.org/10.1509/jmkg.71.3.018 Mendonça, S., Pina e Cunha, M., Kaivo-oja, J. & Ruff, F. (2004) Wild
Hughes, P., Morgan, R.E., Hodgkinson, I.R., Kouropalatis, Y. & cards, weak signals and organisational improvisation. Futures,
Lindgreen, A. (2020) A diagnostic tool to determine a strategic 36(2), 201–218. https://doi.org/10.1016/S0016-3287(03)00148-4
 17404762, 2022, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/emre.12535 by University of Ghana - Accra, Wiley Online Library on [19/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
HULTMAN ET AL. 433
Miao, C.F. & Evans, K.R. (2014) Motivating industrial salesforce with Sharma, S., Durvasula, S. & Dillon, W.R. (1989) Some results on the
sales control systems: An interactive perspective. Journal of Busi- behavior of alternate covariance structure estimation
ness Research, 67, 1233–1242. https://doi.org/10.1016/j.jbusres. procedures in the presence of non-normal data. Journal of
2013.04.007 Marketing Research, 26, 214–221. https://doi.org/10.1177/
Moncrief, W.C. & Marshall, G.W. (2005) The evolution of the seven 002224378902600207
steps of selling. Industrial Marketing Management, 34, 13–22. Siemsen, E., Roth, A. & Oliveira, P. (2010) Common method bias in
https://doi.org/10.1016/j.indmarman.2004.06.001 regression models with linear, quadratic, and interaction effects.
Mueller, J.S., Melwani, S. & Goncalo, J.A. (2011) The bias against Organizational Research Methods, 13(3), 456–476. https://doi.org/
creativity why people desire but reject creative ideas. 10.1177/1094428109351241
Psychological Science, 23, 13–17. https://doi.org/10.1177/ Simon, H.A. (1955) A behavioural model of rational choice. The
0956797611421018 Quarterly Journal of Economics, 69(1), 99–118. https://doi.org/10.
Nemkova, E., Souchon, A.L. & Hughes, P. (2012) Export decision- 2307/1884852
making orientation: an exploratory study. International Marketing Simon, H.A. (1960) The new science of management decision. New
Review, 29(4), 349–378. https://doi.org/10.1108/ York: Harper and Brothers.
02651331211242610 Singh, R. & Koshy, A. (2010) Determinants of B2B sales account
Nemkova, E., Souchon, A.L., Hughes, P. & Micevski, M. (2015) managers’ performance and effectiveness: a review and
Does improvisation help or hinder planning in determining export synthesis of literature. The Journal of Business and Industrial
success? Decision theory applied to exporting. Journal of Interna- Marketing, 25(7), 535–546. https://doi.org/10.1108/
tional Marketing, 23(3), 41–65. https://doi.org/10.1509/jim.14.0071 08858621011077763
Nutt, P.C. (2008) Investigating the success of decision making processes. Souchon, A.L., Hughes, P., Farrell, A.M., Nemkova, E. & Oliveira, J.
Journal of Management Studies, 45, 425–455. https://doi.org/10. S. (2016) Spontaneity and international marketing performance.
1111/j.1467-6486.2007.00756.x International Marketing Review, 33(5), 671–690. https://doi.org/10.
O’Brien, R.M. (2007) A caution regarding rules of thumb for variance 1108/IMR-06-2014-0199
inflation factors. Quality & Quantity, 41(5), 673–690. https://doi. Sousa, C.M. & Coelho, F. (2011) From personal values to creativity:
org/10.1007/s11135-006-9018-6 evidence from frontline service employees. European Journal of
O’Reilly, C.A., III, Chatman, J. & Caldwell, D.F. (1991) People and Marketing, 45(7/8), 1029–1050. https://doi.org/10.1108/
organizational culture: A profile comparison approach to assessing 03090561111137598
person-organization fit. Academy of Management Journal, 34, Sparks, J.R. (1994) Machiavellianism and personal success in market-
487–516. https://doi.org/10.5465/256404 ing: The moderating role of latitude for improvisation. Journal of
O’Toole, J., Gong, Y., Baker, T., Eesley, D.T. & Miner, A.S. (2020) the Academy of Marketing Science, 22, 393–400. https://doi.org/10.
Startup responses to unexpected events: The impact of the relative 1177/0092070394224008
presence of improvisation. Organization Studies, 42(11), 1741– Stajkovic, A.D. & Luthans, F. (1998) Self-efficacy and work-related
1765. https://doi.org/10.1177/0170840620937859 performance: A meta-analysis. Psychological Bulletin, 124, 240–
Panagopoulos, N.G., Lee, N., Pullins, E.B., Avlonitis, G.J., 261. https://doi.org/10.1037/0033-2909.124.2.240
Brassier, P., Guenzi, P. & Oksanen-Ylikoski, E. (2011) Inter- Strutton, D., Pentina, I. & Pullins, E.B. (2009) Necessity is the mother
nationalizing sales research: Current status, opportunities, and of invention: Why sales account manager creativity is more impor-
challenges. Journal of Personal Selling and Sales Management, tant now than ever and what we can do to encourage it. Journal of
31(3), 219–242. https://doi.org/10.2753/PSS0885-3134310302 Selling and Major Account Management, 9, 20–35.
Parhizgar, M.M., Elhami, S. & Gheysari, K.H. (2017) The effect of Theodosiou, M. & Katsikea, E. (2007) How management control and
sales account manager improvisation on sales performance in job-related characteristics influence the performance of export
3 hypermarkets. Journal of Administrative Management, Education sales managers. Journal of Business Research, 60, 1261–1271.
and Training, 13, 302–311. https://doi.org/10.1016/j.jbusres.2007.05.003
Pina e Cunha, M., Clegg, S., Rego, A. & Neves, P. (2014) Organiza- Tom, G. & Lucey, S. (1997) A field study investigating the
tional improvisation: From the constraint of strict tempo to the effect of waiting time on customer satisfaction. The Journal
power of the avant-garde. Creativity and Innovation Management, of Psychology, 131, 655–660. https://doi.org/10.1080/
23(4), 359–373. https://doi.org/10.1111/caim.12076 00223989709603847
Pina e Cunha, M., Vieira da Cunha, J. & Kamoche, K. (1999) Organi- Vera, D. & Crossan, M. (2005) Improvisation and innovative perfor-
zational improvisation: What, when, how and why. International mance in teams. Organization Science, 16(3), 203–224. https://doi.
Journal of Management Reviews, 1(3), 299–341. https://doi.org/10. org/10.1287/orsc.1050.0126
1111/1468-2370.00017 Wang, G. & Miao, C.F. (2015) Effects of sales force market orientation
Ping, R.A. (1995) A parsimonious estimating technique for on creativity, innovation implementation, and sales performance.
interaction and quadratic latent variables. Journal of Journal of Business Research, 68(11), 2374–2382. https://doi.org/
Marketing Research, 32(3), 336–347. https://doi.org/10.1177/ 10.1016/j.jbusres.2015.03.041
002224379503200308 Wang, G. & Netemeyer, R.G. (2002) The effects of job autonomy, cus-
Podsakoff, P.M., MacKenzie, S.B., Lee, J.Y. & Podsakoff, N.P. (2003) tomer demandingness, and trait competitiveness on sales account
Common method biases in behavioral research: A critical review manager learning, self-efficacy and performance. Journal of the
of the literature and recommended remedies. Journal of Applied Academy of Marketing Science, 30, 217–228. https://doi.org/10.
Psychology, 88(5), 879–903. https://doi.org/10.1037/0021-9010.88. 1177/00970302030003003
5.879 Wang, G. & Netemeyer, R.G. (2004) Sales account manager creative
Proctor, R.A. (1991) The importance of creativity in the management performance: conceptualization, measurement, and nomological
field. British Journal of Management, 2(4), 223–230. https://doi. validity. Journal of Business Research, 57, 805–812. https://doi.org/
org/10.1111/j.1467-8551.1991.tb00028.x 10.1016/S0148-2963(02)00483-6
Shalley, C.E., Zhou, J. & Oldham, G.R. (2004) The effects of personal Wiltbank, R., Dew, N., Read, S. & Sarasvathy, S.D. (2006) What to
and contextual characteristics on creativity: Where should we go do next? The case for non-predictive strategy. Strategic Man-
from here? Journal of Management, 30(6), 933–958. https://doi.org/ agement Journal, 27(10), 981–998. https://doi.org/10.1002/
10.1016/j.jm.2004.06.007 smj.555
 17404762, 2022, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/emre.12535 by University of Ghana - Accra, Wiley Online Library on [19/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
434 HULTMAN ET AL.
Wolfgramm, R., Flynn-Coleman, S. & Conroy, D. (2015) Dynamic
interactions of agency in leadership (Dial): An integrative frame- How to cite this article: Hultman, M., Boso, N.,
work for analysing agency in sustainability leadership. Journal of
Business Ethics, 126(4), 649–662. https://doi.org/10.1007/s10551- Yeboah-Banin, A.A., Hodgkinson, I., Souchon,
013-1977-7 A.L., Nemkova, E. et al. (2022) How agency and
World Bank. (2019) The Little Data Book on Africa 2019. Washington, self-efficacy moderate the effects of strategic
DC: World Bank. improvisational behaviors on sales performance:
Moorman, C. & Miner, A.S. (1998) The convergence of planning and Evidence from an emerging market. European
execution: Improvisation in new product development. Journal of
marketing, 62(3), 1-20. Management Review, 19(3), 417–435. https://doi.
Perry, L.T. (1991), “Strategic improvising; how to formulate and imple- org/10.1111/emre.12535
ment competitive strategies in concert”. Organizational Dynamics,
19, 51-64.
APPENDIX: CONSTRUCT ITEMS AND MEASUREMENT STATISTICS
Constructs and their measures Loading Mean (SD) α CR AVE
(t value)
Creativity 5.168 (0.676) 0.808 0.748 0.517
I think out of the box 0.626 (7.653**) 5.392 (0.901)
I try new approaches to problems 0.738 (9.376**) 5.084 (0.841)
I aim at originality in generating solutions 0.729 (9.225**) 5.053 (0.863)
I am inventive in overcoming barriers 0.775 (9.970**) 5.152 (0.797)
Spontaneity 4.776 (0.964) 0.893 0.827 0.678
I respond in the moment 0.829 (11.482**) 4.980 (1.130)
I deal with it on the spot 0.858 (12.066**) 4.911 (1.112)
I act spontaneously 0.836 (11.616**) 4.744 (1.028)
I respond impulsively 0.769 (10.337**) 4.480 (1.167)
Action-orientation 5.213 (0.683) 0.713 0.764 0.503
I take action 0.688 (7.982**) 5.363 (1.058)
I become focused on dealing with the situation 0.695 (8.168**) 5.594 (1.015)
I don’t have any problem getting started on my response 0.742 (8.980**) 4.892 (0.782)
I become action oriented 0.711 (8.382**) 5.011 (0.843)
Adaptability 5.562 (0.718) 0.874 0.807 0.720
I am very flexible in the selling approach I use 0.860 (12.013**) 5.551 (0.808)
I can easily use a wide variety of selling approaches 0.940 (13.607**) 5.492 (0.826)
I am very sensitive to the needs of my customers 0.733 (9.724**) 5.653 (0.772)
Customer demandingness 6.155 (0.759) 0.798 0.724 0.573
The customers I serve demand very high standards of quality 0.759 (9.466**) 6.301 (0.931)
My customers require a perfect fit between their needs and our offerings 0.806 (10.151**) 5.994 (0.916)
My customers expect the highest levels of product and service quality 0.702 (8.642**) 6.171 (0.850)
Agency 5.543 (0.683) 0.889 0.825 0.673
I actively keep myself on track to complete my plans 0.737 (9.781**) 5.634 (0.770)
I am conscious of my actions because they define my personal identity 0.882 (12.566**) 5.510 (0.816)
When completing tasks I am conscious of what I can and cannot handle 0.827 (11.453**) 5.503 (0.805)
When completing tasks, I tend to evaluate the effectiveness of my choices 0.828 (11.484**) 5.533 (0.763)
Self-efficacy 5.678 (0.670) 0.882 0.824 0.667
I feel confident in my ability to perform my job well 0.687 (8.983**) 6.020 (0.731)
I feel very capable of dealing with the demands of the sales job 0.864 (12.243**) 5.562 (0.814)
I feel I have the capabilities to successfully perform this job 0.927 (12.583**) 5.584 (0.806)
I feel very capable of dealing effectively with job-related problems 0.769 (10.413**) 5.552 (0.768)
(Continues)
 17404762, 2022, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/emre.12535 by University of Ghana - Accra, Wiley Online Library on [19/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
HULTMAN ET AL. 435
Constructs and their measures Loading Mean (SD) α CR AVE
(t value)
Experience 4.540 (2.714) 0.875 0.810 0.731
How many years of experience do you have in a sales job? 0.854 (11.802**) 6.041 (3.596)
How many years of experience do you have in your current territory? 0.781 (10.512**) 3.442 (2.313)
How many years of experience do you have in the current industry? 0.925 (13.163**) 4.140 (3.057)
Compensation type - - - -
Please circle the point (x) on the scale that best reflects how you are paid … (0% 0.989 (15.484**) 3.412 (2.120)
to 100% salary vs. commission)
Firm size - - - -
Number of full-time employees (log) 0.902 (14.111**) 1.605 (0.186)
Performance 5.118 (0.805) 0.896 0.832 0.687
Increasing market share for my company 0.843 (11.736**) 4.140 (0.879)
Selling products with higher profit margins 0.793 (10.798**) 5.101 (0.914)
Selling to large volume customers in my territory 0.905 (12.991**) 5.172 (0.995)
Meeting the sales targets assigned to me 0.768 (10.336**) 5.072 (0.896)
Note: Fit indices: χ 2(507) = 570.298; NFI = 0.998; NNFI = 0.999; CFI = 0.999; SRMR = 0.061; RMSEA (90% confidence interval) = 0.024 (0.009–0.034).
 17404762, 2022, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/emre.12535 by University of Ghana - Accra, Wiley Online Library on [19/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License