Strategy, strategic management accounting and performance: a contingency perspective

The contemporary business environment, characterized by rapid technological advancements, intense competition, evolving stakeholder expectations, military and trade conflicts and sustainability pressures, is becoming increasingly dynamic and uncertain. The key mechanism for companies to cope with these challenges and uncertainties is to develop effective business strategies and make informed strategic decisions (Teece et al., 2020), a process that can be facilitated by properly designed and effective decision support systems (Luan et al., 2019) including management accounting (Bhimani and Langfield-Smith, 2007).

The focus of this study is strategic management accounting (SMA), an element of a decision support system focused on facilitating strategic decisions and securing competitive success (Cadez and Guilding, 2008). In the literature, SMA is typically conceptualized as a set of strategically oriented management accounting techniques that are used for planning, controlling and informing strategic decisions (Guilding et al., 2000; Hadid and Al-Sayed, 2021). This prevalent conceptualization takes the perspective of accountants and implicitly assumes that if SMA techniques are being used by accountants (information preparers) then the resulting outputs (information) are also used by the decision-makers (information users).

As a relative novelty to the theory of SMA, we challenge this implied assumption and advance a new conceptualization of SMA. Contrary to the prevalent approach, taking the perspective of accountants, we propose a construct taking the perspective of SMA information users, that is managers who make decisions. When developing strategies and making strategic decisions, managers can exhibit different levels of reliance on SMA-generated information, hence we label the proposed construct managerial reliance on SMA. This is clearly distinct from the established SMA usage construct. While SMA usage reflects the extent to which accountants deploy these techniques, managerial reliance on SMA refers to the degree to which decision-makers use the information generated by these techniques. For example, management accountants may exhibit high SMA usage (i.e. deploy a wide array of SMA techniques extensively), while managers may show low reliance on SMA-generated information (i.e. not use this information extensively to inform decisions), and vice versa (Cinquini and Tenucci, 2010).

Management accounting scholars have long argued that the quality of decision-making, facilitated by management accounting decision support systems, enhances organizational performance (Burkert et al., 2014; Milgrom and Roberts, 1995). Contingency theorists posit that organizational performance is a function of fit or alignment between organizational structures (including SMA as an element of a decision support system) and organizational contingencies (Cadez and Guilding, 2008; Gerdin and Greve, 2004; Otley, 2016). The most important contingency for designing a functional SMA system, as both the label “strategic” and empirical evidence suggest, is organizational strategy (Nixon and Burns, 2012). Prior evidence indicates that SMA aligned with strategic choices positively influences the relationship between strategy and performance (Cadez and Guilding, 2008; Oyewo, 2022).

Motivated by calls for more research on comprehensive modeling of the roles of strategy and SMA in facilitating organizational performance (Cadez and Guilding, 2008; Hadid and Al-Sayed, 2021), and drawing on contingency theory, we propose a conceptual model that includes six constructs: three strategic choices, two elements of a decision support system for strategic decisions, and organizational performance. The three strategic choices examined are highly relevant for companies operating in dynamic and competitive markets: strategy deliberation, market orientation and nonfinancial priorities. We theorize that these strategic choices do not influence performance directly, but rather indirectly through more informed decisions, facilitated by two elements of a decision support system: managerial reliance on SMA and accountants’ involvement in strategy processes. In line with this theorization, we ask and attempt to answer the following research question:

Building on the research question, we develop and empirically test two alternative mediation mechanisms. The first is a simple mediation, where managerial reliance on SMA serves as the sole mediator linking strategic choices to performance (S → M → P). The underlying theoretical logic is that strategic choices shape managerial reliance on SMA, which enhances the quality of managerial decision-making and, ultimately, organizational performance. We then extend this logic by introducing an additional antecedent mediator – accountants’ involvement in strategy – thereby specifying a serial mediation (S → M1 → M2 → P). In this specification, strategic choices are expected to influence the degree of accountants’ involvement in strategy, which, in turn, affects managerial reliance on SMA and, ultimately, performance. Consistent with the foundations of contingency theory, we formulate six mediation hypotheses.

The proposed conceptual model is tested on a sample of 138 firms from Czechia. There are two reasons for choosing this specific context. First, prior evidence on SMA focuses extensively on firms in developed Western countries, while non-Western contexts are underrepresented in research (Hadid and Al-Sayed, 2021). Second, countries that have recently undergone major social upheavals, such as a transition from socialist to market economy (Cadez, 2013), often demonstrate commendable and variable levels of business innovations, such as SMA (Cadez and Guilding, 2008), a property desirable for empirical investigation.

Empirical evaluation of the conceptual model provided support for four of the six posited hypotheses. As for the simple mediation hypotheses, strategy deliberation and nonfinancial priorities were found to be indirectly positively related to performance through managerial reliance on SMA, while no such effect was observed for market orientation. The same pattern was observed for the serial mediation hypotheses – strategy deliberation and nonfinancial priorities were indirectly related to performance through accountants’ involvement in strategy and managerial reliance on SMA, while market orientation was not. Answering our research question, these findings demonstrate that both elements of the decision support system are powerful mediators between two of the three strategic choices and performance.

Our study advances SMA literature and theory in two important ways. The first theoretical contribution is advancing a novel construct of SMA – managerial reliance on SMA. Unlike the conceptualization prevalent in prior literature, taking the perspective of accountants (information preparers), the newly proposed construct takes the perspective of managers (information users). The second contribution is advancement of a contingency model of SMA in response to calls for comprehensive modeling of strategy and SMA in facilitating performance (Hadid and Al-Sayed, 2021). We propose theoretically and provide empirical evidence that the nature of the relationships between strategic choices, SMA (as an element of the decision support system), and organizational performance is one of mediation: strategic choices influence decision support systems, which, in turn, influence performance.

The remainder of the paper is structured as follows. Section 2 reviews the SMA literature and defines the construct managerial reliance on SMA. Section 3 develops the conceptual model and hypotheses. Section 4 outlines the research method. Section 5 reports the empirical results. Section 6 discusses the findings and Section 7 concludes the paper.

The term SMA was first introduced by Simmonds (1981), reflecting relative novelty of the discipline. The emergence of SMA during the 1980s occurred alongside a period of intense criticism directed at traditional management accounting practices (Johnson and Kaplan, 1987; Kaplan, 1984). These critiques proved influential, stimulating renewed innovation in management accounting. As a result, a range of new approaches and techniques evolved – spanning areas such as costing, decision support, planning and control, competitive analysis, and customer-focused accounting – which have since been collectively recognized under the broad concept of SMA (Cadez and Guilding, 2007, 2008; Guilding et al., 2000).

Until around the turn of the century, the pioneers of SMA primarily described and advocated individual management accounting techniques, even while using the label SMA in their writings (Bromwich, 1990; Brouthers and Roozen, 1999; Lord, 1996; Rickwood et al., 1990; Simmonds, 1981, 1982; Ward, 1992). These early writings achieved only limited consensus on which techniques constitute SMA. The turn of the century also marked a turning point in the conceptualization of SMA. In their seminal works, Guilding et al. (2000) and Cadez and Guilding (2008) proposed two criteria for the strategic orientation of management accounting techniques: environmental (outward-looking) and long-term (forward-looking) orientation. The outward and forward focus of SMA contrasts with the focus of traditional management accounting techniques on internal affairs and operational issues (Johnson and Kaplan, 1987), which have lost relevance in an increasingly competitive and complex world (Järvenpää, 2007; Rautiainen et al., 2024) preoccupied with the pursuit of competitive advantage (Bromwich, 1990). Cadez and Guilding (2008) provide a taxonomy of SMA dimensions and constituent techniques, widely adopted in subsequent empirical studies of SMA (Cescon et al., 2019; Hadid and Al-Sayed, 2021; Nik Abdullah et al., 2022; Pavlatos, 2015; Turner et al., 2017).

Contemporary work in SMA follows two paths (Carmona and Ezzamel, 2023). The more prominent approach is contingency-based, assuming that actors behave rationally when designing SMA systems to align with their organizations’ strategies and enhance performance. Much of this work is quantitative and survey-based (Baird et al., 2024; Cadez and Guilding, 2008, 2012; Cescon et al., 2019; Cinquini and Tenucci, 2010; Hadid and Al-Sayed, 2021; Namazi and Rezaei, 2024; Nazaruddin et al., 2026; Nuhu et al., 2025; Oyewo, 2022; Pavlatos, 2015; Rashid et al., 2023). Another vein of research is interpretive – this work is largely conceptual or case-based (Alsharari, 2023; Carlsson-Wall et al., 2015; Cuganesan et al., 2012; Dello Sbarba, 2024; Lapsley and Rekers, 2017; Ma and Tayles, 2009; Nixon and Burns, 2012; Roslender et al., 2023). The strength of this approach lies in its depth, although this depth comes at the expense of generalizability. Our study is ascribed to the contingency vein.

Contingency-oriented research on SMA can be grouped into four broad categories based on how contingency fit is conceptualized and operationalized (Cadez and Guilding, 2008; Chenhall, 2003). The first category comprises congruence-based studies, focusing on identifying contingency factors that explain the use of SMA, treating SMA as the dependent variable. As these models do not incorporate performance outcomes, they do not formally assess contingency fit. The second category consists of interaction-based studies, in which SMA is specified as a moderating variable in the relationship between contingencies and performance. In this approach, contingency fit is inferred from the presence of a statistically significant interaction effect (Burkert et al., 2014). The third category includes intervening or mediation-based studies where SMA is modeled as a mechanism through which contingency factors influence performance. In this case, contingency fit is reflected in a significant indirect (mediation) effect (Gerdin and Greve, 2008). Finally, systems or configurational studies adopt a holistic perspective by examining multiple contingencies simultaneously. While this approach captures complex patterns of alignment, it does not yield a measure of contingency fit. Table 1 provides an overview of contingency-based SMA studies across these four categories.

As noted earlier, the prevalent conceptualization of SMA in the literature is as the construct SMA usage which refers to the level of use or deployment of a range of SMA techniques by management accounting professionals (Cadez and Guilding, 2008, 2012; Guilding et al., 2000; Hadid and Al-Sayed, 2021; Nik Abdullah et al., 2022).

In this study, as a relative novelty to the theory of SMA, we propose a new SMA construct labeled managerial reliance on SMA. Unlike the traditional SMA usage construct, which adopts the perspective of accountants as information preparers, our proposed construct takes the perspective of SMA information users, that is managers who make decisions. Managerial reliance on SMA is clearly distinct from the SMA usage construct. High SMA usage indicates that management accountants deploy a broad range of SMA techniques extensively to garner strategically relevant information (Cadez and Guilding, 2008). In contrast, managerial reliance on SMA refers to the extent to which managers depend on information obtained through SMA for purposes of setting strategic objectives, monitoring strategic performance, taking actions, cross-level and cross-functional managerial dialogue, and other strategy-related decisions.

Though one can reasonably expect that correlation between SMA usage and managerial reliance on SMA ought to be high, the reality is often different. For example, it is possible that strategic management accountants exhibit high SMA usage (deploying a wide array of SMA techniques extensively) and yet managers can exhibit low reliance on SMA-generated information in their strategic decision-making (not using this information extensively to inform decisions). The opposite is also possible – managers may demonstrate high reliance on SMA information though SMA usage by accountants may be low. There is also empirical evidence supporting the view that strategic decisions and SMA are often decoupled (Cinquini and Tenucci, 2010). Figure 1 summarizes the difference between the traditional and newly proposed construct graphically.

Our conceptual model includes six constructs. At the heart of the model is the construct managerial reliance on SMA, described in the preceding paragraph. The most important contingency for the design of a functional SMA system, as both the label “strategic” and empirical evidence suggest, is organizational strategy (Cadez and Guilding, 2008; Cinquini and Tenucci, 2010; Nixon and Burns, 2012). Organizational strategy is a broad and complex concept, as strategic choices are many (Teece et al., 2020). In this study, we examine three strategic choices that are highly relevant for any company operating in dynamic and competitive markets. These are: strategy deliberation, market orientation, and nonfinancial priorities.

Strategy deliberation is often advocated by normative strategic management scholars arguing that strategy should arise from a deliberate stream of decisions (Miles and Snow, 1978). In other words, organizations must first think about how to achieve their goals and then act according to the predetermined plan. This normative view was challenged by Henry Mintzberg (Mintzberg, 1987; Mintzberg et al., 1995; Mintzberg and Waters, 1985) who observed that, in reality, organizational strategies are often emergent rather than being predetermined. He sees strategy as a craft where strategic decisions are bound to be ambiguous and even messy as companies must continuously adapt to the unpredictable environment. In this view, thinking and acting take place simultaneously and constantly. In practice, the two extreme strategies (pure deliberation, pure emergence) are uncommon, most firms fall in between the two extremes (Mirabeau and Maguire, 2014).

Market orientation is often viewed as central to modern management and strategy (Andreou et al., 2020; Hult et al., 2005; Narver and Slater, 1990). This philosophy maintains that planning and coordination of all company activities focus on the primary goal of satisfying customer needs. This business culture emphasizes the effective and efficient creation of superior value for customers in a way that outperforms competitors (Morgan and Anokhin, 2020). Narver and Slater (1990) propose that this concept comprises three behavioral components: customer orientation, competitor orientation and interfunctional coordination.

Nonfinancial priorities refer to a managerial focus on nonfinancial issues – such as customer satisfaction, employee engagement, innovation, learning, sustainability, and others (Zhou et al., 2022) – rather than solely on financial matters like revenue, profit, and return on investment. There are four reasons why companies pursue these priorities:

• they are in the self-interest of the firm, including customer satisfaction, employee engagement, innovation, and quality (Tawse and Tabesh, 2023; Wanderley et al., 2021);

• companies respond to stakeholder pressures, such as reducing waste and pollution (Anzilago et al., 2024; Cadez et al., 2019; Monteiro et al., 2026);

• companies are genuinely socially responsible (Bouten et al., 2024; Galant and Cadez, 2017); or

• in a transitional context, these priorities persist as a legacy of a socialist past (Albu et al., 2020; Albu et al., 2023; Cadez and Guilding, 2012).

Nonfinancial priorities should not be seen as conflicting with financial priorities. Nonfinancial indicators are often leading indicators of long-term value creation and competitive advantage (Zarzycka and Krasodomska, 2021), two key elements of any business strategy (Carmona and Ezzamel, 2023), whereas financial indicators are typically lagging indicators of performance (Lopatta et al., 2024).

The fifth construct examined in our model is management accountants’ involvement in strategy. Although this construct addresses strategy, it does not refer to making strategic choices such as setting organizational priorities and actions, but rather to an element of a decision support system (Cadez and Guilding, 2008). Jarvenpaa (2007) championed the view that contemporary management accountants, unlike their traditional counterparts, are not only information collectors, analyzers, and preparers, but also business partners (Mucci et al., 2025) who are integral members of corporate decision-making teams (Aver and Cadez, 2009; Bajra and Cadez, 2018; Wolf et al., 2015), including in the digital age (Yigitbasioglu et al., 2022). This construct concerns the degree of involvement of management accountants in strategic decision-making processes and teams (Wooldridge and Floyd, 1990) and has been used in several prior SMA studies (Aver and Cadez, 2009; Cadez and Guilding, 2008, 2012).

The ultimate dependent construct examined in this paper is organizational performance.

Our model is based on the tenets of contingency theory, the dominant approach for investigating management accounting phenomena in general (Carmona and Ezzamel, 2023; Chenhall, 2003; Otley, 2016), and SMA phenomena in particular (Cadez and Guilding, 2008; Carmona and Ezzamel, 2023).

The fundamental tenet of contingency theory holds that company performance is a product of fit between the structure (including elements of a decision support system, such as SMA and accountants’ involvement in strategy) and organizational contingencies (such as strategic choices). It is a widely held convention in contingency theory that good fit enhances performance while poor fit diminishes performance (Cadez and Guilding, 2008; Gerdin and Greve, 2004). From a contingency perspective, this conditional logic requires that performance be explicitly incorporated into empirical models (Chenhall, 2003; Gerdin and Greve, 2008). When performance is omitted, such models do not test contingency relationships but, instead, reflect congruence-based analyses, as they focus on alignment between variables without assessing performance outcomes (Burkert et al., 2014; Gerdin and Greve, 2004).

Operationalizing contingency fit in empirical testing is varied and multifaceted (Aguinis et al., 2017; Boyd et al., 2012; Burkert et al., 2014; Gerdin and Greve, 2004). Two theoretical considerations are important when specifying a functional test:

1. is fit a continuum or a discrete state; and

2. is structural choice a mediator or a moderator (Burkert et al., 2014; Gerdin and Greve, 2008)? For the first consideration, we model relationships as a continuum, since all constructs in our model can change frequently and marginally (Cadez and Kalkhouran, 2026).

For the second, we model both elements of a decision support system as mediators, again on theoretical grounds. As elaborated later in the hypotheses development section, strategic choices in our conceptual model influence the two elements of a decision support system, which, in turn, influence performance, thus denoting the mediation effect (Aguinis et al., 2017; Chenhall, 2003). Following the Gerdin and Greve (2004) typology, a Cartesian-contingency-mediation form of fit is postulated. In such specifications, fit is depicted as a statistically significant indirect effect between contingency (in our conceptual model strategy choices) and performance (Cadez and Guilding, 2008; Gerdin and Greve, 2004).

The conceptual model is presented graphically in Figure 2. As evident, the model examines eight bivariate relationships between constructs, indicated by arrows. However, consistent with our research question and theoretical framework, our focus is not on testing these bivariate relationships, but on testing mediation effects. To assess the contingency-mediation form of fit, we propose six mediation hypotheses between the exogenous construct strategic choices and the ultimate dependent construct performance. Three hypotheses involve a shorter path with one mediating variable (managerial reliance on SMA) while the other three involve a serial path with two mediating variables (accountants’ involvement in strategy and managerial reliance on SMA).

Firms differ in whether they follow a deliberate and predetermined strategy formulation or a more emergent strategy formulation where patterns develop in the absence of intentions or in spite of them (Mintzberg, 1987; Mintzberg et al., 1995; Mintzberg and Waters, 1985). In practice, most firms fall between the two extremes (Gomez-Conde et al., 2023; Mirabeau and Maguire, 2014). Since deliberate strategies invoke more active management of strategy, this suggests a greater call for strategically-oriented information, such as that provided by SMA, concerned with broad-scope information relevant for long-term business strategies (Cadez and Guilding, 2008; Hadid and Al-Sayed, 2021). The empirical evidence is also supportive of a positive bivariate relationship between strategy deliberation and SMA (Cadez and Guilding, 2008, 2012), hence we posit that strategy deliberation is positively related with managerial reliance on SMA.

Moving beyond the expected bivariate relationship between strategy and SMA, the fit between strategy deliberation and managerial reliance on SMA is expected to enhance decision-making quality and, in turn, deliver performance benefits. With greater managerial reliance on SMA-generated information, firms pursuing a deliberate strategy build the capacity to anticipate the unexpected (Huikku et al., 2018) and to adapt and react quickly when unanticipated market events occur (Teece et al., 2020), thereby securing competitive advantage and superior performance even in the face of inevitable future market upheavals (Luan et al., 2019; Teece et al., 2020). In more analytical terms, we posit that the impact of strategy deliberation on performance is not direct but operates indirectly through the mediating construct managerial reliance on SMA as an enabler for exploiting a deliberate strategy effectively. Stated alternatively, we expect that managerial reliance on SMA positively mediates the relationship between strategy deliberation and performance:

Firms also differ in their market orientation, that is, the extent to which company activities focus on satisfying customer needs (Hult et al., 2005; Narver and Slater, 1990). The key facets of market orientation – customer orientation, competitor orientation and inter-functional coordination – aiming to satisfy customers (Andreou et al., 2020) align closely with the key facets of SMA, namely, customer and competitor accounting (Cadez and Guilding, 2008; Guilding et al., 2000). This alignment suggests a positive relationship between market orientation and the use of SMA-generated information by managers focused on customer satisfaction. The empirical evidence also supports a positive bivariate relationship between market orientation and SMA facets (Cadez and Guilding, 2012; Guilding and McManus, 2002). Therefore, in line with the arguments presented, we posit that market orientation is positively related to managerial reliance on SMA.

Moving beyond the expected bivariate relationship, the fit between market orientation and managerial reliance on SMA is also expected to enhance decision quality and, consequently, performance. Market-oriented firms aim to satisfy their customers in ways that outperform competitors (Hult et al., 2005; Slater and Narver, 2000), therefore customer and competitor related information is highly valuable (Andreou et al., 2020; Slater and Narver, 2000). As customer and competitor accounting are the two cornerstones of SMA (Cadez and Guilding, 2012), SMA is well positioned to inform market-oriented strategies by providing valuable inputs to inform customer- and competitor-related decisions, thereby enhancing performance. In line with this discussion, we posit that managerial reliance on SMA positively mediates the relationship between market orientation and performance:

Similar to the two preceding strategic choices, firms also differ with respect to their strategic priorities (Chenhall and Langfield-Smith, 1998). We are particularly interested in their emphasis on nonfinancial priorities. These, as the name suggests, relate to nonfinancial matters, and can be in the self-interest of the firm, such as customer satisfaction (Tawse and Tabesh, 2023), responsive to stakeholder pressures, such as reducing waste and pollution (Cadez et al., 2019; Monteiro et al., 2026), or even altruistic, such as corporate social responsibility (Galant and Cadez, 2017). Given that information regarding all these issues (i.e. customers, competitors, quality, innovation, efficiency, sustainability) is the building block of SMA (Cadez and Guilding, 2008, 2012), a positive relationship is expected between nonfinancial priorities and managerial reliance on SMA.

Consistent with the logic of the preceding hypotheses, the fit between nonfinancial priorities and managerial reliance on SMA is also expected to yield performance benefits. Firms that pursue nonfinancial priorities (e.g. customer satisfaction, employee engagement, innovation, quality, efficiency, sustainability) seek information related to these areas to improve decision quality, and SMA is incredibly well positioned to provide such information as these aspects are core elements of SMA. Notably, an emphasis on nonfinancial priorities does not imply a trade-off between nonfinancial and financial performance. On the contrary, since most nonfinancial priorities are leading indicators of future financial performance (e.g. satisfied customers, motivated employees, excellent innovation, etc.) (Tawse and Tabesh, 2023), a good fit between nonfinancial priorities and managerial reliance on SMA can enhance overall performance, not just performance in nonfinancial areas. In line with this rationale, we posit the following hypothesis:

Relative to the simple mediation paths, serial mediation paths include an additional antecedent mediating variable accountants’ involvement in strategy. Unlike the simple mediation hypotheses, we posit that strategic choices in our conceptual model first influence accountants’ involvement in strategy and that their involvement, in turn, influences managerial reliance on SMA, which, ultimately, influences performance.

As noted earlier, strategy deliberation entails more active and predetermined management of strategy. In increasingly competitive and uncertain markets, interdepartmental teams can improve the speed and quality of an organization’s reaction to environmental developments, thus improving decision quality and performance (Baines and Langfield-Smith, 2003; Rowe et al., 2008; Scott and Tiessen, 1999). Heterogeneous senior management teams, including management accountants, are better equipped to recognize strategic opportunities, as greater breadth of functional perspectives enhances more informed strategy identification, as well as more effective strategy implementation (Bhimani and Langfield-Smith, 2007; Grabner et al., 2022; Naranjo-Gil and Hartmann, 2007). Accounting information such as financial analytics, cost evaluations, scenario-based projections, and the like, is vital for strategy development and implementation, hence in line with this argument, we expect that strategy deliberation is positively related to accountants’ involvement in strategy, an expectation supported by empirical evidence by Cadez and Guilding (2008).

Beyond the bivariate relationship, the involvement of management accountants in strategic decision-making teams is expected to further influence managerial reliance on SMA-generated information. This operates through two mechanisms. First, greater involvement in strategy will inculcate accountants with a more profound appreciation of the nature of the information needs posed by strategic management which, in turn, is likely to result in accountants generating SMA intelligence better tailored to the needs of strategic managers (Abernethy and Bouwens, 2005; Cadez and Guilding, 2008). Second, the increased involvement of accountants in strategy brings accountants and managers into closer locational and temporal proximity facilitating the timely conveyance and use of SMA information. Together, these mechanisms increase managers’ reliance on SMA-generated intelligence.

Moving beyond simple mediation, we further hypothesize that the conditional association – strategy deliberation, accountants’ involvement in strategy and managerial reliance on SMA – facilitates more effective managerial decisions and, in turn, enhances organizational performance (Chattopadhyay et al., 2001; Grabner et al., 2022; Wooldridge and Floyd, 1990). By proactively analyzing business issues, being customer-oriented, and liaising across functional boundaries and levels of management, accountants provide valuable input as integral participants in key organizational decision-making processes (Cadez and Guilding, 2008; Scott and Tiessen, 1999), which not only leads to greater reliance on SMA (Grabner et al., 2022), but also results in more informed managerial decisions and enhanced organizational performance (Hoque, 2004; Hoque and James, 2000):

Given the close conceptual alignment between market orientation facets, namely, customer orientation, competitor orientation and interfunctional coordination (Andreou et al., 2020), and SMA dimensions, namely, customer and competitor accounting (Cadez and Guilding, 2008; Guilding et al., 2000), we posit a positive relationship between market orientation and accountants’ involvement in strategy. Management accountants clearly provide valuable inputs into market-oriented decision-making processes. Extending beyond this bivariate relationship and using the same rationale as in the preceding hypothesis, the involvement of management accountants in strategic decision-making teams is expected to influence managerial reliance on SMA-generated information which, in turn, is expected to positively influence performance:

Nonfinancial priorities, though somewhat counterintuitively, are also likely to influence accountants’ involvement in strategy. As noted earlier, companies pursuing nonfinancial priorities will likely promote interdepartmental teams to improve decision quality in relation to these priorities (Baines and Langfield-Smith, 2003; Rowe et al., 2008; Scott and Tiessen, 1999), as heterogeneous senior management teams are better suited to recognize strategic opportunities. These teams are likely to also include management accountants as they can provide invaluable financial information about customers, competitors, quality, innovation, efficiency, and sustainability, all of which are important elements of SMA (Cadez and Guilding, 2008, 2012; Guilding et al., 2000). Therefore, we expect a positive relationship between nonfinancial priorities and management accountants’ involvement in strategy. Moving beyond this bivariate relationship and using the same rationale as in the preceding hypotheses, the involvement of management accountants in strategic decision-making teams is expected to influence managerial reliance on SMA-generated information which, subsequently, is expected to positively influence performance:

The data collection method used was an online questionnaire survey. Questionnaires are vital tools for collecting primary data about internal behavioral phenomena (such as organizational practices and decision-making processes) which are not available in public databases or other archival sources (Oluwalana and Ibiwoye, 2020). In the field of management accounting research, surveys are the preferred method when large sample sizes are sought to ensure the generalizability of results (Speklé and Widener, 2020), which also applies to our study.

The questionnaire was pilot-tested with six specialists (three academics and three industry experts). Feedback from the pilot study prompted only minor wording refinements for clarity. In designing and administering the survey, we followed established guidelines for survey research (Fowler, 2013; Speklé and Widener, 2020). Because survey participation rates are typically low – particularly in Central and Eastern Europe – we implemented response-rate enhancement strategies (Baruch and Holtom, 2008; Dillman et al., 2014), including initial telephone invitations and follow-up e-mail reminders.

Our focus was on examining medium and large firms located in the Czech Republic. Large and medium-size firms are likely to have formalized strategic processes and management accounting functions. The target population was defined as firms with over 50 employees and turnover exceeding 256 million CZK. The Albertina CZ Gold Edition database was used to identify the population size. Altogether, approximately 3,900 firms in the database met these criteria. From this population, we randomly selected 1,000 firms, contacted them by telephone, and invited them to participate in an online survey in 2019. In the first wave, we received 90 responses. This was followed by a second wave, in which the firms were contacted by telephone again, an effort yielding additional 48 responses and bringing the total to 138 (response rate 13.8%). Relative to comparable published studies, our response rate is satisfactory. For example, a recent SMA literature review (Rashid et al., 2021) reported that responses in SMA surveys ranged from 26 to 134, with an average of around 101 responses. In a recent renowned study, Hadid and Al-Sayed (2021) report a sample size of 149.

The data were also screened for completeness. Because the online questionnaire required mandatory responses for all items, there were no missing data, and all 138 responses were usable. The details of the sampling process are outlined in Table 2.

All six constructs in the conceptual model are measured with multiple items using five-point Likert-type scales. This is the preferred method when measuring abstract concepts such as strategy and SMA as the measurement of such constructs inevitably carries a degree of measurement error (Hair et al., 2019). The measurement instruments are explained below.

Exogenous constructs (independent predictors)

Strategy deliberation (two indicators); Scale 1 – strongly disagree, 5 – strongly agree:

1. Detailed strategy formulation in our company takes place before the implementation of strategy.

2. Strategic decision-making in our company is carried out in accordance with a predetermined strategy.

Market orientation (four indicators); Scale 1 – strongly disagree, 5 – strongly agree:

1. Our business has a good understanding of the needs of our customers.

2. Employees ensure that customers are provided high-quality products and/or services.

3. Managers of our company are focused on satisfying customer needs and wants on well-defined markets.

4. Our company is strongly market oriented.

Nonfinancial priorities (six indicators); Scale 1 – very little, 5 – very high:

1. Customers (e.g. market share, satisfaction and customer retention);

2. Employees (e.g. employee satisfaction, employee turnover, employee skills and development);

3. Operational metrics (e.g. productivity, safety, optimization of time use);

4. Innovation (e.g. research and development, success of new products/services);

5. Quality (e.g. product/service quality, defects, awards received); and

6. Social responsibility (e.g. environmental compliance, impact on communities, public image).

Endogenous constructs (mediators)

Accountants’ involvement in strategy (five indicators); Scale 1 – not at all involved, 5 – fully involved:

1. identifying problems and designing strategic goals;

2. proposing options for strategic direction;

3. evaluation of strategic direction options;

4. finding information on variants; and

5. taking the necessary actions to implement strategic changes.

Managerial reliance on SMA (ten indicators); Scale 1 – very low, 5 – very high:

1. identification of critical factors for achieving strategic objectives;

2. setting strategic objectives;

3. monitoring the achievement of strategic objectives;

4. providing information to address deviations from predefined strategic performance targets;

5. monitoring key areas of strategic performance;

6. use in top management meetings;

7. use in middle and lower management meetings;

8. use in ongoing discussions within all levels of management;

9. use in addressing issues related to strategic uncertainties; and

10. stimulating and facilitating dialogue and information sharing between different levels of management.

Endogenous construct (ultimate dependent variable)

Performance (3 indicators); Scale 1 – significantly below average, 5 – significantly above average:

1. operating cash flow;

2. net profit; and

3. sales.

The strategic choices (exogenous constructs in the model) were measured as follows. Strategy deliberation (two indicators) is measured using an instrument adapted from Cadez and Guilding (2008). It assesses the extent to which strategy formulation is predetermined and carried out in accordance with predetermination (Scale: 1 – strongly disagree, 5 – strongly agree). A market orientation (four indicators) instrument, also adapted from Cadez and Guilding (2008), evaluates the extent to which employees and management demonstrate commitment to customer satisfaction (Scale: 1 – strongly disagree, 5 – strongly agree). Nonfinancial priorities (six indicators) is an original instrument. It gauges the organizations’ emphasis on six areas: customers, employees, operational metrics, innovation, quality, and social responsibility (Scale: 1 – very little, 5 – very high).

Each element of the decision support system (endogenous mediating constructs in the model) was measured as follows. Accountants’ involvement in strategy (five indicators) is assessed using an instrument originally developed by Wooldridge and Floyd (1990) and later adapted to an accounting context by Cadez and Guilding (2008). It assesses the degree of involvement of management accountants in various stages of the strategic process, including designing strategic goals, proposing and evaluating strategic options, gathering information and implementing strategic changes (Scale: 1 – not at all involved, 5 – fully involved). Managerial reliance on SMA is an originally developed construct, so we also created a new instrument to measure it. This instrument assesses the extent to which management relies on SMA information in ten different strategic activities, such as setting strategic objectives, monitoring strategic objectives (Scale: 1 – very low, 5 – very high).

The ultimate dependent variable – performance – is measured with an instrument deployed by Cadez and Guilding (2008). Respondents were asked to indicate their relative performance compared to competitors along three indicators: operating cash flow, net profit and sales (Scale: 1 – significantly below average, 5 – significantly above average).

To evaluate the proposed conceptual model, partial least squares structural equation modeling (PLS-SEM) was used using SmartPLS 4 software. PLS-SEM is a composite-based form of SEM, recognized as a robust and well-established methodology within management and social science literature (Dijkstra and Henseler, 2015; Hair et al., 2022; Hair et al., 2018).

PLS-SEM is highly suitable for the task due to several factors:

• it can handle multiple relationships between observable indicators and unobservable latent constructs at the same time;

• it considers measurement error during the estimation process;

• it enables the testing of direct, indirect, and interaction effects among constructs within the same model; and

• it does not demand a large sample size (Hair et al., 2009; Hair et al., 2018; Henseler et al., 2014).

Following the recommended two-phase approach (Hair et al., 2022; Hair et al., 2019), the assessment began with the measurement (outer) model (reported in Section 5.1) followed by the evaluation of the structural (inner) model (reported in Section 5.2). Reflective measurement was used for all constructs because the indicators do not serve as defining characteristics of these constructs; rather, alterations in constructs lead to corresponding changes in indicators (Diamantopoulos and Siguaw, 2006). In this stage, two indicators of construct market orientation were excluded due to validity issues. This does not affect the model’s overall predictive ability, as reflective measurement does not depend on the number of items per construct (Diamantopoulos and Siguaw, 2006; Henseler et al., 2014).

As outlined in our hypotheses, we are primarily interested in mediation effects rather than direct effects between constructs. We proposed two types of mediation: simple mediation (H1a–H1c) and serial mediation (H2a–H2c). In path models, mediation is demonstrated when there is a statistically significant indirect effect between a predictor and a criterion variable through one (simple mediation) or multiple (serial mediation) mediating variables (Hair et al., 2022; Sarstedt et al., 2020). In all our hypotheses, we expect complementary mediation, meaning that all direct effects constituting an indirect effect are in the same direction (positive). Our analysis follows the most recent methodological recommendations, including testing the significance of indirect effects using bias-corrected bootstrap confidence intervals rather than p-values, as recommended by Hair et al. (2022) and Sarstedt et al. (2020).

As outlined by Hair et al. (2019), the evaluation of the reflective measurement model should encompass:

• scrutinizing the indicator loadings;

• evaluating the internal consistency reliability through composite reliability;

• examining the convergent validity of each construct via average variance extracted (AVE); and

• evaluating discriminant validity using the heterotrait-monotrait (HTMT) ratio of the correlations.

We now discuss these issues, in turn. Hair et al. (2019) suggested that indicator loadings exceeding 0.7 are preferable, as they indicate that the construct explains more than 50% of the indicator’s variance. Some quality studies in the SMA field applied less strict criteria, e.g. Hadid and Al-Sayed (2021) used an indicator loading threshold of 0.5. In this study, we applied a threshold value of 0.6. This threshold led to the exclusion of two indicators for market orientation. The outer loadings of the retained items are presented in Table 3.

Next, Table 4 presents internal consistency, reliability and convergent validity parameters. According to Hair et al. (2019), Cronbach’s alpha, though widely used, is a less accurate reliability measure because it involves unweighted items. In contrast, composite reliability considers the weights of items based on the individual loadings of construct indicators, resulting in higher reliability accuracy than Cronbach’s alpha. For both metrics, higher values indicate higher reliability, whereby values between 0.70 and 0.95 are typically considered acceptable. In our case, all values are within the suggested range. The third step concerns the evaluation of convergent validity, that is how well the construct accounts for the variance of its items. To assess a construct’s convergent validity, the AVE is used as a metric for all items associated with the construct. An AVE of 0.50 or higher is deemed acceptable, signifying that the construct accounts for at least 50% of the variance of its items. In our case, all AVE values are above the recommended threshold.

The fourth step involves evaluating discriminant validity, the degree to which a construct is empirically distinct from other constructs within the structural model. Henseler et al. (2015) advocate the HTMT ratio of correlations as a preferred method (Voorhees et al., 2016) whereby high HTMT values signal potential issues with discriminant validity. Henseler et al. (2015) recommended a threshold of 0.90 for models with conceptually similar constructs (values above the threshold indicate problems with discriminant validity) and a more conservative threshold of 0.85 for conceptually more distinct constructs. Table 5 suggests little concern for issues with discriminant validity.

After establishing that reliability and validity were satisfactory, we proceeded toward estimating the structural (inner) model. This model examines the proposed relationships (paths) between constructs. To determine the statistical significance of the path coefficients, the bootstrapping method with 10,000 iterations was deployed. This approach is favored over single model estimation because the final parameters are based on multiple model estimations across different samples rather than depending on the assumptions of a statistical parameter distribution (Hair et al., 2009; Hair et al., 2022; Henseler et al., 2014).

Prior to assessing the significance and relevance of the structural model relationships, we assessed the collinearity between constructs. The pairwise correlations between constructs are presented in Table 6. As evident, pairwise correlations are low to moderate, the highest correlation is observed between the constructs nonfinancial priorities and managerial reliance on SMA. As a formal test of potential multicollinearity, we also calculated VIF values for all constructs. Details are presented in Table 7.

As shown in Table 7, all VIF values are well below the conservative threshold of 3 (Hair et al., 2022), therefore we conclude that multicollinearity between the constructs is not a concern.

According to Hair et al. (2019), standard criteria for the assessment of a structural model include the extent of variance explained by the endogenous constructs (R²) and the statistical significance and relevance of the path coefficients. Table 8 presents the evaluation of the direct effects in the structural model.

As shown in Table 8, the model explains 16.8% of the variation for the construct accountants’ involvement in strategy, 56.9% of the variation for managerial reliance on SMA, and only 3.5% for performance. Although the explained variance for performance is low, this was expected, as organizational performance is influenced by many factors with SMA being just one of them. All direct effects are positive and, with one exception (path from market orientation to accountants’ involvement in strategy), statistically significant.

Since all our hypotheses concern the indirect (mediation) effects of SMA, Table 9 provides an evaluation of the indirect effects in the structural model. When assessing the statistical significance of indirect effects in PLS-SEM, Hair et al. (2022) argue that p-values are overly conservative parameters and propose that confidence intervals are more suitable for this purpose as they align better with the method’s nonparametric nature and provide richer information about parameter estimates. For this reason, we report both p-values and confidence intervals. The interpretation of confidence intervals is as follows: if the entire confidence interval for an estimated coefficient lies above or below value 0 (or in other words, the interval does not contain zero value), the effect is statistically significant.

As shown in Table 9, although all examined indirect effects are positive, they differ in statistical significance. According to the overly conservative p-value criterion, only one hypothesized indirect effect is at the threshold of significance (H1c, p = 0.051). In contrast, using the more appropriate confidence interval criterion, four of the six specific indirect effects examined are statistically significant; in all four cases, the entire confidence interval lies above zero. We therefore conclude that our model supports H1a, H2a, H1c and H2c. Both unsupported hypotheses involve market orientation as the exogenous construct.

Grounded in contingency theory, this study examines the mediating role of two decision support system elements in the relationship between organizational strategic choices and performance. The three strategic choices analyzed are strategy deliberation, market orientation, and nonfinancial priorities. We theorize that these choices do not influence performance directly, but rather indirectly through two elements of a decision support system: managerial reliance on SMA and accountants’ involvement in strategy. We proposed and empirically tested two mediation paths: a simple mediation path with managerial reliance on SMA as the sole mediator, and a serial mediation path with accountants’ involvement in strategy as the first mediator and managerial reliance on SMA as the second mediator. The proposed conceptual model, which includes six constructs and six testable hypotheses, was tested empirically by analyzing data from 138 Czech firms.

In line with the pair of hypotheses regarding the indirect relationship between strategy deliberation and performance, we find that both the short path with one mediator (managerial reliance on SMA) and the long path with two mediators (accountants’ involvement in strategy and managerial reliance on SMA) are statistically significant. This finding is consistent with the study by Cadez and Guilding (2008), who also found that SMA usage positively mediates the relationship between strategy deliberation and performance. This observation supports our theoretical argument that firms with a more deliberate strategy are likelier to deploy heterogeneous management teams, including management accountants; that greater involvement of accountants’ in strategy increases managerial reliance on SMA information; and, in turn, that greater reliance on SMA information increases decision-making quality and, ultimately, performance (Chattopadhyay et al., 2001; Grabner et al., 2022; Wooldridge and Floyd, 1990).

Regarding the second pair of hypotheses, which address the indirect relationship between market orientation and performance, neither mediation hypothesis was supported. This is somewhat surprising, as the key facets of market orientation and SMA – customer orientation, competitor orientation and inter-functional coordination – are conceptually very closely aligned (Andreou et al., 2020; Cadez and Guilding, 2008). In the simple mediation model, although both bivariate effects were significant (market orientation → managerial reliance on SMA, managerial reliance on SMA → performance), the indirect effect was not statistically significant. In the serial mediation model, the bivariate direct effect path coefficient for market orientation → accountants’ involvement in strategy was close to zero, indicating no effect. This means that market-oriented firms do not seek informational input from accountants which supports longstanding observations that customer accounting and marketing remain separate when making strategic decisions regarding customers (Guilding and McManus, 2002; Matsuoka, 2020; Roslender, 1995).

Finally, we provide support for the third pair of hypotheses that nonfinancial priorities are indirectly positively related to performance at a statistically significant level, both through a shorter path with one mediator and a serial path with two mediators. This is consistent with our theoretical argument that companies pursuing nonfinancial priorities are likely to promote interdepartmental teams that include strategic management accountants. They can provide valuable financial information on customers, competitors, quality, innovation and sustainability (Cadez and Guilding, 2008, 2017). This, in turn, will increase managerial reliance on SMA and, ultimately, performance.

Regarding our research question, these findings collectively demonstrate that both elements of the decision support system for strategic decisions, appraised in this study, are powerful and effective mechanisms for leveraging strategic choices strategy deliberation and nonfinancial priorities to enhance organizational performance. While this finding aligns with previous contingency-based research showing that firms with more deliberate strategies and a greater focus on nonfinancial priorities seek broader-scope information provided by SMA (Cadez and Guilding, 2008; Cescon et al., 2019; Chenhall and Langfield-Smith, 1998), it extends prior research by providing evidence that this fit also leads to positive performance outcomes. From a holistic perspective, this finding underscores the importance of aligning SMA practices with organizational strategies and strategic objectives for companies aiming to improve performance.

Overall, the study makes two important contributions to SMA literature and theory. The first is the introduction and development of the novel construct managerial reliance on SMA. While the prevailing conceptualization of SMA in the literature – SMA techniques usage – reflects the perspective of accountants (information preparers), our proposed construct adopts the perspective of managers (information users). Prior evidence indicates that these two perspectives are often decoupled in practice (Cinquini and Tenucci, 2010). Not all SMA information prepared by accountants is necessarily relied upon by managers in decision-making processes, and conversely, not all information sought by the managers can be prepared by accountants, as managers often rely on intuition and feelings in their decision-making (Hulpke and Fronmueller, 2021).

The second contribution is the advancement of a contingency model of SMA. Responding to calls for comprehensive modeling of alternative contingencies and the role of SMA in facilitating performance (Hadid and Al-Sayed, 2021), we propose a conceptual model that theorizes multiple indirect relationships between strategic choices, elements of a decision support system for strategic decisions and performance. We theoretically propose and empirically provide evidence that the relationships between these constructs are mediated: strategic choices influence accountants’ involvement in strategy and managerial reliance on SMA which, in turn, affect performance.

From a broader management accounting perspective, the logical sequence of the strategy-SMA relationship also raises questions about the theoretical suitability of moderation analysis when examining the relationships between strategy and management accounting. This is because a key assumption of moderation analysis is that the moderator is not caused by the predictor (Aguinis et al., 2017). Consistent with our findings and prior evidence, it appears unlikely that an effective SMA system can be designed independently of strategic choices (Cadez and Guilding, 2008). In other words, if an effective SMA system were independent of organizational strategy, the optimal SMA system would be universal for all companies.

This study aimed to answer the research question: Do elements of a decision support system mediate the relationship between organizational strategic choices and organizational performance? Our results demonstrate that managerial reliance on SMA and accountants’ involvement in strategy effectively mediate the relationship for two strategic choices – strategy deliberation and nonfinancial priorities – but not for the strategic choice of market orientation.

These findings offer two useful and practical implications. Firms pursuing deliberate strategies can benefit from greater involvement of accountants in strategy processes. This will, in turn, increase managerial reliance on SMA information, and, in effect, enhance decision quality and performance. The same implication applies to firms pursuing nonfinancial priorities. Although it may seem counterintuitive, these firms can benefit from greater accountants’ involvement in strategy processes: accountants’ financial input can strengthen managers’ reliance on SMA information and, in turn, improve performance. Surprisingly, but consistent with prior observations (Guilding and McManus, 2002; Matsuoka, 2020; Roslender, 1995), market-oriented firms do not instigate accountants’ involvement in strategy, possibly due to stereotypes attributed to accountants by marketers (e.g. marketers’ creativity and commercialism versus accountants’ dullness and bureaucracy) (Dimnik and Felton, 2006).

Like any study, this one has limitations. While the general drawbacks of survey research apply, some limitations are specific to this study. First, as with any conceptual model, the model is a simplified representation of reality and does not include all potentially relevant constructs. Second, the constructs examined are abstract and, therefore, subject to possible measurement error. Third, the study’s reliance on self-reported data introduces the potential for response bias, particularly regarding subjective measures of performance. To address these concerns theoretically, we followed the strategy-structure-paradigm, which is well established in the management accounting literature (Cadez and Guilding, 2008; Carmona and Ezzamel, 2023). To address them methodologically, we used structural equation modeling, the preferred method for testing complex mediation effects (Aguinis et al., 2017), and followed the most recent best practice recommendations for PLS-SEM analysis (Hair et al., 2022; Sarstedt et al., 2023), including an extensive assessment of measurement reliability and validity and estimation bootstrapping.

While the findings should be interpreted in light of these limitations, the study provides novel and valuable insights into the roles of management accountants and SMA in developing and implementing strategic choices. In addition, the study also offers useful pointers for future research. One interesting path forward is to use alternative methodologies to validate the proposed conceptual model, explore causal relationships and address endogeneity-related concerns addressed in Oyewo et al. (2025), while another promising path includes grassroots qualitative approaches for investigating the phenomena of interest in this study.

The authors thank Libuse Soljakova for her help in the development of the questionnaire and data collection.

This study involved human participants via a voluntary online questionnaire survey. Formal ethical approval was not required by the university, but the research complied with the Code of Ethics of the Prague University of Economics and Business (VŠE) [Link to Code of EthicsLink to the website of Code of Ethics].