Toward a design theory of strategic enterprise management business intelligence (SEMBI) capability maturity model

Abstract

Purpose

The purpose of this study is to demonstrate that Strategic Enterprise Management (SEM) and Business Intelligence (BI) have the potential to integrate management decisions vertically through an organization’s hierarchy. This study also aims to present a design theory framework and build a model dimension using eight principles serving as mid-range theories.

Design/methodology/approach

This study uses a design science perspective to posit how organizations can successfully implement SEMBI (a union of SEM and BI). This study then completes the design theory by building the method dimension using two principles. Finally, the study presents testable hypotheses for the theory and an evaluation using stakeholder attitudes and judgments as proxies for objective measures.

Findings

In the search for a prescription for SEMBI success, this study finds that the notion of the Capability Maturity Model (CMM) is a good artifact with which to organize the principles the authors are seeking. CMM has since been adapted to suit different contexts by incorporating relevant principles from those domains. Hereafter, this study refers to SEMBI–CMM as the adapted solution for SEMBI's success.

Originality/value

This study coins and uses the term SEMBI to represent the union of SEM and BI. This term retains its distinct identities and principles and forms a holistic and integrated view of SEM and BI implementation strategies. In an effort to advance this line of research, this study employs a design science perspective to address the question of how an organization can successfully implement SEMBI.

Citation

Luo, X.(R). and Chang, F.-K. (2023), "Toward a design theory of strategic enterprise management business intelligence (SEMBI) capability maturity model", Journal of Electronic Business & Digital Economics, Vol. 2 No. 2, pp. 159-190. https://doi.org/10.1108/JEBDE-11-2022-0041

1. Introduction

Evidence-based management (EBM) or evidence-based decision-making is a recent paradigmatic movement emphasizing two types of decision-making evidence: “Big E Evidence” and “Little e evidence” (Rousseau, 2006). Big E Evidence stems from validated theories that decision-making can be done deductively, whereas Little e evidence holds that decision-making can be done inductively. The extant studies indicate that Strategic Enterprise Management (SEM), as a Big E Evidence-based DSS, and Business Intelligence (BI), as a Little e evidence-based DSS, are becoming the dominant DSS applications in industry and have piqued increasing interest among information systems researchers (Phillips-Wren, Daly, & Burstein, 2021; Ain, Vaia, DeLone, & Waheed, 2019).

Compared to its enterprise management application offspring (i.e., Enterprise Resource Planning [ERP]), SEM emphasizes strategy-focused characteristics (Brignall & Ballantine, 2004). While ERP systems can provide integrated solutions for planning, executing and controlling business processes horizontally across the value chain, SEM will extend these principles vertically to support strategic management processes, such as strategic planning, risk management, performance monitoring and value communication (Norton, 1999).

In essence, there is a trend to converge SEM and BI because of their inextricably complementary nature. Real-time or operational BI emphasizes the integration of technologies with business processes and with the management thereof: “Business value of BI is in its use within management processes that impact operational processes (which in turn, drive revenue or reduce costs), as well as its use within those operational processes themselves” (Williams & Williams, 2003). Thus, we see some overlap with SEM, as BI begins to emphasize integrating tactical and operational levels of management with the business processes themselves (Ain et al., 2019). Since SEM is a strategic and systematic management approach to plan, monitor, control and manage the implementation of business strategy, its principles may integrate well with BI, which is more focused on tactical and operational levels in the firm. For example, SEM can not only provide a strategic process to leverage BI processes (Frolick & Ariyachandra, 2006), SEM can also provide strategic metrics with which to leverage the BI output (Golfarelli, Rizzi, & Cella, 2004). The alignment of the SEM and BI may take one of two forms: (1) SEM-driven top-down or (2) BI-enabled bottom-up.

In this study, we coin and use the term SEMBI to represent the union of SEM and BI. This term retains its distinct identities and principles and forms a holistic and integrated view of SEM and BI implementation strategies. In an effort to advance this line of research, we employ a design science perspective to address the question of how an organization can successfully implement SEMBI. Hevner, March, Park, and Ram (2004), Venable, Pries-Heje, and Baskerville (2016) and Baskerville, Baiyere, Gregor, Hevner, and Rossi (2018) identify seven guidelines for design science, the sixth of which claims that design science research includes the search for an effective design artifact. In our search for a prescription for SEMBI success, we find the notion of the Capability Maturity Model (CMM) is a good artifact with which to organize the principles we are seeking. CMM was originally developed to assess the maturity of software development processes (Paulk, Curtis, Chrissis, & Weber, 1993); it has since been adapted to suit different contexts by incorporating relevant principles (Lacerda & von Wangenheim, 2018) from those domains. Hereafter, we refer to SEMBI-CMM as our adapted solution for SEMBI's success.

In the rest of this paper, we review the literature and outline our research method (Section 2); we then construct the design theory for SEMBI-CMM (Section 3). Finally, we test and evaluate the theory and present the theoretical contributions and practical implications of the current work (Sections 4–6).

2. Literature review and research method

2.1 SEMBI as an information systems artifacts

According to March and Smith's classification of IT research outputs (March & Smith, 1995), we discovered that SEM is an IT artifact, and its associated research outputs can be categorized into constructs, models, methods and instantiations. The findings of our analysis are presented in Table 1. When it comes to constructs, scholars create a conceptual framework that outlines the issues in the field and specifies their solutions (Teece, 2019). Regarding models, researchers devise a set of propositions or statements that articulate the relationships between structures (Brignall & Ballantine, 2004; Sen, Bingol, & Vayvay, 2017; Regent, Glinkina, Ganina, Markova, & Kozhina, 2019; Khudyakova et al., 2020; Melnyk & Zlotnik, 2020). For methods, scholars put forward a series of steps to accomplish a task (ZAITSEVA, 2014; Faizova, Ivanova, & Pozhuieva, 2018; Mathrani, 2021). In the case of instantiations, IT studies instantiate specific information systems and tools to address various aspects of designing information systems (Sestino, 2016; Akhmetshin et al., 2018). Moreover, with the advancement of big data analysis technology, the concept of BI has been incorporated into the SEM artifact (Sestino, 2016; Teece, 2019; Mathrani, 2021). As a result, there is some overlap between SEM and BI, as BI now emphasizes merging tactical and operational-level management with business processes themselves. Furthermore, as an IT artifact, the theoretical functions (Gregor, 2006) of related research can be categorized into analysis (which states what is) (Akhmetshin et al., 2018; Regent et al., 2019; Melnyk & Zlotnik, 2020), explanations (which outlines what is, how, why, when and where) (ZAITSEVA, 2014; Sen et al., 2017; Teece, 2019; Khudyakova et al., 2020), prediction (which anticipates what is and what will be) (ZAITSEVA, 2014; Sen et al., 2017; Teece, 2019; Khudyakova et al., 2020), explanation and prediction (which specifies what is, how, why, when, where and what will be) (Sestino, 2016; Faizova et al., 2018; Mathrani, 2021) and design and action (which illustrates how to do something). To further advance this line of research, we adopt a design science perspective to tackle the question of how organizations can effectively implement SEMBI.

The knowing-doing gap is a challenging problem in CMM design (Mettler & Rohner, 2009): CMM is sometimes criticized for lacking rigor because it is an ad hoc method needing theory and empirical support (Bach, 1994; Biberoglu & Haddad, 2002; Hansen, Rose, & Tjørnehøj, 2004). In addition, CMM suffers when it lacks organizational relevance and does not describe how to effectively perform requisite actions (Porte, 2018).

We believe that design science as a research paradigm can bridge this gap by simultaneously addressing both the rigor and relevance concerns of CMM design methods. We base this study on the design science research framework proposed by Hevner et al. (2004) who assert that two research activities; i.e., building and evaluating, interact with one another in the construction of an IT research artifact. In our case, the IT artifact is SEMBI–CMM.

2.2 SEMBI–CMM as an information systems design theory

March and Smith (1995) distinguish constructs, models, methods and instantiations as four basic IT artifact types. CMM is rooted in the Quality Management Maturity Grid comprising two dimensions: capability and capability improvement (Maier, Moultrie, & Clarkson, 2009). In this context, a capability is the application of competences to effect the desired end. Capability improvement describes the development of capabilities over time, progressing through different levels of repeatability from ad hoc toward some form of idealistic, perfectly repeatable state. Thus, we can see that the capability dimension tends to be a model IT artifact type because it uses constructs to represent capability problems and solutions. Likewise, the capability improvement dimension tends to be a method IT artifact type because it describes the process guiding capability improvement.

Mettler and Rohner (2009) assert that CMM fits between a model and method IT artifact. In fact, models and methods are more closely interrelated than other artifact types of information systems design science research (Winter, 2008). Models and methods are two sides of the same coin. In essence, models focus on the result and imply procedural aspects, whereas methods focus on procedures and imply results.

Jones and Gregor (2007) divide design artifacts into material or abstract artifacts. They argue that abstract artifacts including constructs, models and methods are theory or components of the theory. Kuechler and Vaishnavi (2008) present a similar view, “…our interpretation is that March and Smith’s use of the terms ‘model’ and ‘method’ – specified as desirable outputs of [design science research] spans the meaning of the term ‘prescriptive design theory’ .” We adopt this broad view of theory in our research to treat SEMBI–CMM as a theory including both a model and a method corresponding to the two dimensions of any CMM.

Gregor (2006) distinguishes between theory for design and action and other types of theories for analysis, explanation, prediction, explanation and prediction. Theory for design and action is about the principles of form and function, methods and justificatory theoretical knowledge that are used in the development of IT artifacts (Iivari, 2015). Such theories give explicit prescriptions on how to design and develop an artifact, whether it is a technological product or a managerial intervention. Drawing on this interdisciplinary philosophical underpinning, CMM as a theory can be a type of design and action or a type of analysis (i.e., CMM gives an unprejudiced reproduction of some aspects of reality). CMM can be explanatory (i.e., CMM delivers a depiction of causal connections to better understand reality) or predictive (i.e., CMM recommends an efficient solution state of a future reality) (Mettler & Rohner, 2009). To search for a SEMBI solution, we position SEMBI–CMM as a theory for design and action known as information systems design theory (ISDT) (Walls, Widmeyer, & El Sawy, 1992; Gregor, 2002). However, this construction certainly does not prohibit SEMBI–CMM from absorbing other types of theories from the knowledge base.

As a common agreed-upon language that is recognizable and repeatable, ISDT can be used as a guiding framework to provide a way of systematically structuring the “how” design with a “why” foundation. It also helps generate insights that would remain hidden without the use of ISDT (Gregor, 2002). ISDT argues that design is both a product and a process, which means that a design theory concerns two aspects: the design product and the design process. The product and process perspectives of ISDT closely correspond to the model and method dimensions of CMM. This alignment provides a way to theorize the model dimension of CMM as an ISDT product and the method dimension as an ISDT process. However, since design leads to satisfactory solutions without explicitly specifying all possible solutions (Hevner et al., 2004), a designer’s orientation may lead to different design methods to fulfill the requirements.

In our search for a SEMBI–CMM solution, we find a variety of kernel theories from different sources. For the purpose of tightening the scope of kernel theories and integrating them consistently into our design, we introduce mid-range theories in the form of design principles to reflect our design orientations (Kuechler & Vaishnavi, 2008). In the ISDT framework, mid-range theories fall between kernel theories and design methods to facilitate the extrapolating process. Table 2 is the ISDT framework as it pertains to SEMBI–CMM. In an effort to further explain the employment of mid-range theories in the theoretical context, we detail this customized framework in subsequent sections.

3. Building SEMBI, an information systems design theory

Although SEMBI is becoming paramount in today’s business environment, the increasing problem is how to leverage this theory to successfully realize business value from its use (Ain et al., 2019). Therefore, our purpose is to formulate a design theory to help SEMBI stakeholders deploy the SEMBI approach successfully. These stakeholders may include organizations that are using or plan to use SEMBI, SEMBI consulting companies and vendors of SEMBI systems.

3.1 Meta-requirements for design product/model

SEMBI success is the meta-requirement in our design theory that describes the class of goals to which the design theory applies (Walls et al., 1992; Gregor, 2002). We decompose SEMBI success as a general meta-requirement to achieve deeper levels of detail and IS success researchers generally agree upon a three-step process model for so doing (Petter, DeLone, & McLean, 2012).

Some researchers identify the decomposition steps as the creation of a system, the use of the system, and the consequences of system use; they claim the model is only a process model (Seddon, Staples, Patnayakuni, & Bowtell, 1999). Others conceive of the three steps as production, use and net benefit; because of their conceptualization of the variables, they further claim that the model is also a variance model (DeLone & McLean, 2003). In particular, they argue that “simply saying that more use will yield more benefits, without considering the nature of this use, is clearly insufficient. Researchers must also consider the nature, extent, quality, and appropriateness of the system use” (DeLone & McLean, 2003; Petter et al., 2012).

Building upon this prior work, we decompose SEMBI success into the following three meta-requirements:

• MR1: SEMBI–CMM should support the SEMBI production

• MR2: SEMBI–CMM should support the SEMBI use

• MR3: SEMBI–CMM should support the SEMBI net benefit realization

3.2 Meta-design of SEMBI capability model

This section focuses on the meta-design of the model dimension of CMM. These comprise the mid-range theories governing the model design. As summarized in Table 3, we define and explore eight model design principles in this section; we then define and explore the method dimension in the next section.

1. Principle 1: Capability in SEMBI–CMM is constructed within resource-based theory (RBT) and the IS Capability lens.

As noted in Table 4, Resource is the lowest level construct and is defined as stocks of available factors owned or controlled by a firm (Amit & Schoemaker, 1993). RBT suggests that organizations should invest in the resources they believe will best help them succeed in achieving a sustainable competitive advantage (Barney, 1991, 1996). From a competitive perspective, however, not all resources are equally valuable, as it has been argued that the primary source of an organization’s competitive advantage will come from those resources that are simultaneously valuable, rare, not fully imitable and irreplaceable – the so-called VRIN conditions (Barney, 1991).

While resources are clearly a key element of RBT, there is a growing recognition that resources alone do not create value. Rather, value is created by an organization's ability to mobilize, integrate and utilize these resources by applying competences (Black & Boal, 1994; Bowman & Ambrosini, 2000). Thus, in the SEMBI–CMM, Competence plays an integrating role. On the one hand, Competence links to Resources “to deploy resources, usually in combination, using organizational processes” (Amit & Schoemaker, 1993), and on the other hand, Competence also links to Capability “to effect a desired end” (Amit & Schoemaker, 1993).

Capability is a higher level construct than Competence (Stalk, 1992), and it is used to achieve specific organizational goals (McGrath, MacMillan, & Venkataraman, 1995). Within this context, defining and creating the desired organizational Capability would be determined by the organization’s future goals. In turn, this specification establishes the need to improve or develop specific Competences. At any point in time, an organization’s current Capability, derived from its existing Competences, either enables or inhibits goal achievement, at least in the short-term (Peppard & Ward, 2004).

Just as these three constructs are adapted in IS context as IS capability, IS competence and IS resource (Bharadwaj, 2000; Peppard & Ward, 2004; Wade & Hulland, 2004; Aral & Weill, 2007; Prasad, Heales, & Green, 2009), we adapt them in SEMBI context as referring to SEMBI capability, SEMBI competence and SEMBI resource, and we use these constructs with their underpinning principles in SEMBI–CMM design.

1. Principle 2: SEMBI is a dynamic capability distilled from operational capabilities

Not all Capabilities are of the same impact on an organization. Santhanam and Hartono (2003) and Stoel and Muhanna (2009) suggest that capabilities should be distinguished into different orders. There is a broad consensus in the literature that “dynamic capability” is more potent than ordinary (or “operational”) capability (Teece, Pisano, & Shuen, 1997; Zollo & Winter, 2002; Winter, 2003). Teece et al. (1997) define a dynamic capability as “the firm's ability to integrate, build, and reconfigure internal and external competences to address rapidly changing environments”. Zollo and Winter (2002) note that dynamic and operational capabilities have different effects on the generation and appropriation of rents, depending on the pace of change in the environment.

Of course, effective operational capability is always a necessity, and superior operational capability is always a source of advantage. But in a context where technological, regulatory and competitive conditions are subject to rapid change, persistence in the same operational capabilities quickly becomes hazardous. Systematic change efforts are needed to anticipate and respond to environmental changes, and both superiority and viability will prove transient for an organization that has no dynamic capability.

Dynamic capability is, in essence, a repeatable process that governs the rate of change of operational capability (Winter, 2003). While operational capability enables the organization to function under given resource conditions, dynamic capability changes the resource base of the organization (Helfat et al., 2009). SEMBI is subject to a rapidly changing environment, and the adaptability to such change is fundamental for enterprises to leverage SEMBI for their competitive advantage. Therefore, in the SEMBI capability model, the dynamic capability is emphasized and purposefully distilled from the operational SEMBI capabilities.

1. Principle 3: SEMBI capability model is prescribed a priori

We define the SEMBI capability model as an a priori model comprising four categories of competences spanning operational and dynamic capabilities (see Figure 1). Further, unlike an a posteriori approach, which is generally considered to be theory-free and is adopted only in those cases where little theoretical basis exists (Venkatraman, 1989), we arrive at the SEMBI model based on design principles and their underlining kernel theories.

As shown in Figure 1, strategizing, deploying and complementing (governance and culture) competence categories are operational capabilities and the adapting competence category is a dynamic capability. Strategizing competences define the relationship between capability and competence and are purposefully designed “to effect a desired end” (Amit & Schoemaker, 1993). In contrast, deploying competences define the relationship between resource and competence “to deploy resources, usually in combination, using organizational processes” (Amit & Schoemaker, 1993).

While strategizing and deploying competence categories are fundamentally important to link resources to goals, recent IS capability research based on Complementary Theory notes the need for other competences to complement them (Prasad et al., 2009). Complementary Theory asserts that, to maximize organizational payoff, complementary factors must be changed in a coordinated fashion, in the right direction and in the right magnitude. When Prasad et al. (2009) apply this theoretical perspective to the IS environment, they find that complementary competences can not only contribute to other competences but also can directly contribute to the business value realization. In the SEMBI capability model, as depicted in Figure 1, we define two complementing competences: governance and culture.

While these three competence categories (strategizing, deploying and complementing) enable SEMBI to function under the current resource conditions, organizations also need competence to adapt to a changing environment (Teece et al., 1997; Helfat et al., 2009). The “Adapt to change” competence is specially designed for this purpose.

In Principle 4, we prescribe the form of each competence and then in Principles 5–8, we specify the design of each of the aforementioned four competence categories.

1. Principle 4: SEMBI competence is operationalized by a practice-oriented process

Although capability is constituted from competences, and competence is prescribed as a process, there are different types of processes. Any process is a combination of experience, context, interpretation and reflection and involves more human participation than information (Davenport, 2005). Therefore, understanding what kinds of knowledge people create, share and reuse in the process should be an important perspective in determining process type.

Marjanovic and Seethamraju (2008) contrast two types of processes based on the difference in their knowledge nature. While the procedure-oriented process is characterized by using explicit knowledge, the practice-oriented process is characterized by using tacit knowledge. Unlike explicit knowledge, tacit knowledge is often very difficult to communicate and is, therefore, also difficult to automate, organize or manage by technology. However, tacit knowledge can be externalized, to some degree, through problem-solving, reflection, knowledge-in-action and “working things out”.

Externalization of tacit knowledge in an organization results in the development of organizational practices. Practice is a concept designed to capture the essence of “what people actually do,” based on their knowledge, skills and experience and demonstrated by their actions (Schultze & Boland, 2000). Practice is very concerned with “objects” and “ends,” which makes it more specific and observable than competence (Carlile, 2002). Practice is not “a mechanical reaction to rules, norms or models, but a strategic, regulated improvisation responding to the situation” (Schultze & Boland, 2000). In reality, all processes combine, to some degree, both procedures and practices. Therefore, a practice-oriented process means its practice component is more prominent than its procedural component and vice versa (Marjanovic & Seethamraju, 2008).

Popovič, Coelho, and Jaklič (2009) characterize SEMBI-related processes as often non-routine and creative (unclear problem space with many decision options). They note that SEMBI process specifications cannot be predefined in detail, nor is their outcome certain and yet their success often brings innovations and improvements. These characteristics determine that their management should include flexibility (Meredith, 2008). Notably, SEMBI processes tend to work better when not directly controlled and supervised, as those involved need freedom and encouragement to explore and experiment. Moreover, excessive administration, surveillance or a lack of autonomy tends to restrict SEMBI processes. All these points indicate that SEMBI success requires a practice-oriented process environment, where people work with autonomy and creativity to perform practices that apply their knowledge.

With this principle in mind, we prescribe SEMBI competence as a practice-oriented process comprising of a set of practices and aimed at deploying combinations of firm-specific resources to accomplish a given task. In so doing, we define what practices should be done to represent a process structure; however, we avoid overprescribing it by not restricting how practices should be done. Our design not only gives space for people to exercise creativity in using their tacit knowledge, but we also apply the explicit knowledge in the form of process structure to guide people to focus on important issues and stay within the normative boundaries of the problem context.

This principle reflects the socially defined nature of practices as suggested by (Wenger, McDermott, & Snyder, 2002) “a set of socially defined ways of doing things in a specific domain: a set of common approaches and shared standards that create a basis for action, problem solving, performance and accountability.” This principle is a natural extension of the SEMBI capability model. While the model prescribes the SEMBI capability comprising a set of competences, this principle goes further to provide each competence with a best practice structure.

Having established the form of each competence in Principle 4, we now specify Principles 5–8, each of which defines one competence category (Strategizing, Deploying, Complementing and Adapting). We operationalize each competence with a practice-oriented process.

1. Principle 5: Strategizing competences are designed from business-IT alignment lens

Traditional IS strategy researchers claim business-IT alignment is a way of IS/IT strategizing and business value realization (Reich & Benbasat, 1996). While this view is still important today, IS capability researchers go further to consider how well the organization develops, nurtures and utilizes their competences to support both “aligning” and “aligned” business and IT (Peppard & Ward, 2004). While “aligning” emphasizes the competence requirement of agility and sustainability, “aligned” emphasizes the competence requirement of efficiency and effectiveness.

In adapting business-IT alignment view for strategizing (Peppard & Ward, 2004), distinguish two kinds of competences: IS strategy competences and IT strategy competences. IS strategy competences translate the business strategy into processes, information and systems investments and change plans that match the business priorities, and IT strategy competences translate business strategy into long-term information architectures, technology infrastructure and resourcing plans that enable the implementation of the strategy.

Adopting the new lens of business-IT alignment for Strategizing competences, we design SEMBI competences for strategizing and then operationalize them in a practice-oriented shown process in Table 5. As first illustrated in Figure 1, we identify two separate Strategizing competences corresponding to IS strategy and IT strategy. The competence “Define SEMBI contribution” is the IS strategy competence that focuses on new opportunities resulting from businesses leveraging SEMBI. Within this competence, we design four practices. The first two practices concern the fit of SEM and BI at the strategic level (Frolick & Ariyachandra, 2006) of the business; the second two practices focus on the integration of BI with the operational process, i.e., the realization of operational BI, at the operational level of the business (Williams & Williams, 2003; Panian, 2009).

In contrast, “Determine SEMBI system architecture” is an IT strategy competence that focuses on building the SEMBI architecture. This competence also comprises four practices derived from industry and academic knowledge. The practice “Determine ETL (extract, transform, and load) plan” concerns the extraction of data from legacy systems and external sources, the transformation and pre-processing necessary to produce useful integrated data and the loading of resulting data into data warehouse structures (March & Hevner, 2007). Srivastava and Chen (1999) note that ETL is time-consuming and expensive and that businesses can choose from different strategies and techniques to accomplish it. So determining a plan for this process should be a necessary practice for competence building.

Similarly, a data warehouse serves as a repository for data from ETL processes, and Darmont, Boussaid, Ralaivao, and Aouiche (2005) note that warehouse architectures are also numerous and vary widely. As a result, “Determine Data Warehouse architecture” is also a necessary practice. Data mining techniques are specifically designed to identify relationships and rules within the data warehouse (Jackson, 2002). These techniques, while quite powerful, may be too complex and sophisticated for the average information consumer. Thus, identifying the right data mining techniques for different types of problems and users is a needed practice (March & Hevner, 2007). Finally, “Determine SEMBI application” is the fourth practice for the “Determine SEMBI system architecture” competence. This practice concerns deciding which applications should be developed in-house and which are best outsourced from a vendor.

1. Principle 6: Deploying competences are designed to match strategizing competences

Competence is a firm’s capacity to deploy resources to effect the desired end (Amit & Schoemaker, 1993). While strategizing defines where the “desired end” is, deploying competences addresses the need to deploy sufficient resources where and when needed to achieve that end. This pairing of deploying with strategizing competences implies a matching relationship exists between these two competence categories.

Table 6 shows the two Deploying Competences, also introduced in Figure 1, that support the strategizing competences in SEMBI–CMM. “Manage deliver and support” is one of the deploying competences and is concerned with supplying SEMBI solutions and support; it comprises two practices. The practice “Manage service level agreements” focuses on identifying service requirements, agreeing on service levels and monitoring the achievement of those service levels; while the practice “Manage third-party services” focuses on establishing relationships and bilateral responsibilities with qualified third-party service providers and monitoring the service delivery to verify and ensure adherence to agreements.

These “best practices” are built upon similar concepts from IS capability researchers (Feeny & Willcocks, 1998). The SEMBI–CMM “Manage third-party services” practice, for example, corresponds to Feeny’s “vendor development,” while SEMBI–CMM “Manage service level agreements” combines the “contract facilitation” and “contract monitoring” constructs of Feeny & Willcocks (1998).

Also shown in Table 5, “Manage SEMBI human resources” is the second deploying competence and is concerned with SEMBI human resource management. From the competence point of view, these individuals possess SEMBI-related business and technical skills, knowledge and experience. They possess what Wixom, Watson, Reynolds, and Hoffer (2008) call “business-IT hybrid” skills. When technical personnel have more business acumen and business people have more technical skills than most companies, these individuals possess such hybrid skills. As such, they are key individuals for the effective use of SEMBI.

The “Identify and develop SEMBI human resources” practice concerns identifying existing and also developing new SEMBI human resources, and the “Exploit SEMBI human resource” practice concerns mobilizing and marshaling (Bassellier & Benbasat, 2004) these individuals. Note that, just as we do not particularly strategize financial investments in SEMBI, we also do not emphasize the deployment of financial resources. We treat them as embedded in strategizing and deploying processes.

1. Principle 7: Complementing competences are designed from institutional lens

Our Strategizing and Deploying competences are designed from RBT logic, which claims that economically rational identification and use of valuable, rare, difficult-to-replicate and non-substitutable resources can lead to enduring corporate variability and extraordinary profits (Barney, 1991). Despite these significant insights, others argue that this logic has not looked beyond the properties of resources and resource markets to explain enduring firm differentiation. In particular, RBT logic has not examined the institutional context within which resource selection decisions are embedded nor how this context might affect sustainable firm differences (Oliver, 1997).

Neglecting institutional context can lead to a mismatch in resource strategizing and deploying. Based on the economic rationale, the resource-based view assumes that economic motivations drive resource procurement decisions and that economic factors in firm competition and resource environments drive firm behavior and outcomes. In contrast, a normative rationale from an institutional perspective asserts that firms operate within a social framework of norms, values and taken-for-granted assumptions about what constitutes appropriate or acceptable economic behavior (Oliver, 1997).

This resource-based view is particularly problematic for information system practitioners who often embed an economically rational understanding when they look for rational solutions to organizational and technical problems (Svejvig, 2009). The economic mindset can also lead to frustration among project personnel when they encounter problematic unexpected situations or discover ambiguous management decisions that deviate from the “the rational path.” Institutions are multifaceted, durable and resilient social structures, composed of symbolic elements, social activities and material resources, and institutional theory attempts to describe the deeper and more resilient aspects of how institutions are created, maintained, changed and dissolved (Scott, 2013). Institutional theory suggests that motivations for human behavior extend beyond economic optimization to social legitimacy and social obligation (Zukin & DiMaggio, 1990).

Therefore, in these cases, the institutional theory might provide complementary understanding and explanations to stimulate alternative solutions or "just" to reduce frustration among project personnel, which may be good for the organization itself (Svejvig, 2009). SEMBI is particularly relevant to the institution. As an “Evidence-” and “evidence-” based decision support system, SEMBI is characterized as “chaotic” and “subversive” (Meredith, 2008). We will address each of these descriptors, in turn.

3.2.1 SEMBI is chaotic

First, “chaotic” means its future properties cannot be foreseen. This unpredictability is not caused primarily by the SEMBI system itself but rather by how the system is aligned with the business.

Different from other kinds of enterprise systems, SEMBI is more flexible and requires more creativity to align with business systems (Meredith, 2008). Researchers also distinguish between mandatory and voluntary systems (Alavi & Joachimsthaler, 1992). While other kinds of traditional enterprise systems mandate that users’ operational processes align with rigid enterprise processes, SEMBI falls in the voluntary category of system.

In this case, the “Evidence” or “evidence” to use and where to use such E/evidence for meaningful decision support is largely determined by users surrounded and influenced by an institutional context. From an RBT perspective, the E/evidences and the SEMBI system as technical enabler are both resources, and their use means strategizing and deploying these resources in organizations. In practice, however, management decisions often deviate from the best available E/evidence, and the resulting “research-practice gap” is a prevalent phenomenon in organizations (Rousseau, 2006).

From an institutional perspective, the failure to acquire and optimally use E/evidences is not due to their lack of value or that SEMBI systems are not qualified to acquire or produce them but rather because the E/evidences or the systems are inconsistent with the organization's historical, cultural or political context. In other words, when the company believes that available resources are disgusting or politically inconsistent, the competition for valuable resources between the company will be more limited (Oliver, 1997).

3.2.2 SEMBI is subversive

SEMBI is “subversive” in that it is intentionally designed to transform organizational structures because SEMBI is strategic and can directly influence decisions and change organizational goals and direction. This influence contrasts with other kinds of systems that are intended to reinforce existing organizational structures, and while they might also be transformative, their transformation of organizational structures is often unanticipated and unplanned (Meredith, 2008). So, SEMBI can not only be influenced by the host institution but can also dramatically influence that host. In this situation, any institutional mismatch between the organization and the SEMBI will be a significant obstacle impeding SEMBI success.

Oliver (1997) complements RBT with institutional theory and claims that a firm's sustainable advantage depends on its ability to manage resources. He believes that resource capital and institutional capital are indispensable for sustainable competitive advantage, where resource capital refers to value-enhancing assets and competences of the firm, and institutional capital comprises the context surrounding resources and resource strategies that enhance or inhibit optimal use of resource capital. Based on this understanding, SEMBI capability model prescribes Complementing competences from an institutional theory lens to enhance the Strategizing and Deploying competence categories that are more directly related to the resources and resource strategies.

3.2.3 Institutional competences

The chaotic and subversive nature of SEMBI also implies that the institutional competences should be designed separately from those of other enterprise systems. Where the latter is intended to provide competences for control, predictability and conformance with corporate governance structures, the former can only be successful in an environment that encourages flexibility and experimentation (Meredith, 2008). This is not, however, an argument for a completely free rein. Rather, it is an argument for a balanced approach on one side to limit governance within well-defined boundaries and on the other side to provide it with an environment that cultivates creativity. We organize the SEMBI-related institutional competences into two dimensions: governance and culture. Governance provides the leadership and political structure that function as formal rules, and culture establishes a comparatively informal climate.

3.2.4 Governance competences

By adopting this balanced idea, SEMBI governance is lightly weighted as only one competence “Establish SEMBI management leadership” that consists of three practices as shown in Table 7. The “Build SEMBI competence center” practice establishes a transparent, flexible and responsive organization that we name SEMBI competence center. Members of SEMBI competence center are representatives of multiple units who are entrusted with roles and responsibilities for improving SEMBI capability. This concept is adapted from the business intelligence competence center concept (Unger, Kemper, & Russland, 2008) and enhanced by the IT steering committee concept from IS governance research (Karimi, Bhattacherjee, Gupta, & Somers, 2000).

The “Make SEMBI capability improvement plan” practice focuses on benchmarking and prioritizing the competencies for improvement. This practice connects the SEMBI capability model and the evolution method of SEMBI-CMM is covered later in Section Meta-design of SEMBI capability improvement method. The third practice “Communicate SEMBI goal and direction” is designed to establish a shared vision and build commitment among stakeholders. This practice is also used in successful IT governance (Bowen, Cheung, & Rohde, 2007).

3.2.5 Culture competences

In the culture dimension, EBM is emphasized (Rousseau, 2006). As a movement, EBM emphasizes using the best evidence in decision-making. Its purpose is to identify ways of closing the prevailing “research-practice gap” – the failure of organizations and managers to base practices on the best available evidence (Rousseau, 2006). While SEMBI is congruent with this movement by providing a technical enabler, EBM is essentially a culture of evidence: “Building a culture in which managers learn to learn from evidence is a critical aspect of effective evidence use” (Pfeffer & Sutton, 2006).

Thus, the task of designing competences for the culture dimension focuses on institutionalizing evidence use as endorsed by a culture environment. For this purpose, as shown in the first row of Table 8, we define three related evidence-based competences: “Define EBM process”, “Establish EBM ba” and “Cultivate EBM culture”.

Starting from the right side of Table 7 and working left, “Define EBM process” competence makes evidence-use effective; this design view is introduced by Zollo & Winter (2002) who use a systemically designed learning mechanism going beyond semi-automatic stimulus-response mechanism to actively influence the evolution of other competences. Their mechanism addresses the role of experience accumulation, knowledge articulation and knowledge codification processes in this evolution and is similar to Nonaka (1994) knowledge conversion theory (Schreyögg & Kliesch‐Eberl, 2007).

Nonaka (1994) knowledge conversion theory asserts that an organization creates knowledge through a spiral process in which explicit knowledge and tacit knowledge interact in four modes of knowledge conversion: Socialization, Externalization, Combination and Internalization (SICA). As described below, knowledge conversion theory aligns well with the two other competences in the culture dimension,

The practices within the “Define EBM process” competence are designed to correspondence to Nonaka’s SECI modes of knowledge conversion. “Define EBM Socialization mode” converts tacit evidence into new tacit evidence through social interactions among members. “Define EBM Externalization mode” codifies tacit evidence into explicit concepts. “Define EBM Combination mode” converts explicit evidence into more systematic sets of explicit evidence by combining key pieces, and finally “Define EBM Internalization mode” embodies explicit evidence into tacit evidence. By installing this learning mechanism, the users interact with one another and with the SEMBI systems to systemically create and use evidence, which in turn, contributes to the evolution of the other competences.

User commitment would enable the foregoing practices to become more deeply rooted in the institution’s culture. The “Establish EBM ba” and “Cultivate EBM culture” competencies are designed for this purpose. “Ba” is a Japanese word meaning “a shared context in which knowledge is shared, created, and utilized” (Nonaka, Toyama, & Konno, 2000). According to Nonaka et al. (2000), ba is the key to knowledge creation, generation and regeneration, as it provides the energy, quality and place to perform each of the SECI modes and to move along the knowledge spiral. There are four types of ba corresponding with each of the SECI modes: they are, originating ba, dialoguing ba, systemizing ba and exercising ba (Nonaka et al., 2000). We design our “Establish EBM ba” competence and its component practices accordingly.

The “Establish originating ba” offers a context for socialization where individuals share experiences, feelings, emotions and mental models; the “Establish dialoguing ba” offers a context for externalization where individuals' mental models and skills are shared, converted into common terms and articulated as concepts; the “Establish systemizing ba” offers a context for the combination of existing explicit knowledge using collective and virtual interactions; and the “Establish exercising ba” mainly offers a context for internalization using individual and virtual interactions.

Culture as an influence on the knowledge process and ba is broadly discussed in the literature (Nonaka et al., 2000; Gray & Densten, 2005). For our purpose to cultivate culture to foster the SECI mechanism through ba, we design a competence “Cultivate EBM culture” by adopting Gray and Densten (2005) cultural framework. As shown in Table 7, the “Define EBM process” results from the “Establish EBM ba” a formulation supported by Nonaka et al. (2000) work. Gray and Densten (2005) go further to foster these two competencies through a “Cultivate EBM culture” competence (Nonaka et al., 2000).

These relationships also hold between practices in “Cultivate EBM culture” competence and practices in both “Establish EBM ba” and “Define EBM process” competences. The four practices in “Cultivate EBM culture” competence are “Cultivate clan culture,” “Cultivate adhocracy culture,” “Cultivate market culture” and “Cultivate hierarchy culture,”

The “Cultivate clan culture” practice is designed for developing a culture of trust and belongingness that facilities evidence sharing and socialization. The “Cultivate adhocracy culture” practice is designed for developing a culture of flexibility, innovation and creativity that is congruent with the externalization mode. The “Cultivate market culture” practice is designed for developing a culture of rational goals that emphasize competitiveness, productivity, goal clarity, efficiency and accomplishment. This culture provides users with knowledge about how their efforts influence organizational outcomes and have a significant impact on organizational effectiveness. The “Cultivate hierarchy culture” practice is designed for developing a hierarchical culture that emphasizes information management, documentation, stability, standardization and control. These culture facilities the internalization mode.

It may seem that the four types of practices are contradictory, as the first two practices cultivate a culture of flexibility, while the latter two practices cultivate a culture of stability. As noted earlier, in SEMBI a culture of flexibility concerned with tacit knowledge use is emphasized; however, organizations are seldom characterized by a single cultural type (Gray & Densten, 2005). Leaders in organizations should be able to simultaneously master seemingly contradictory or paradoxical demands and balance them. Importantly, leaders should know where different types of culture are applicable and to what extent they apply.

Thus, the culture dimension is prescribed, as three competences corresponding to mechanism-context-culture institutions. Individually, these competences balance one another; together, they complement Strategizing and Deploying competences. Once institutionalized, a competence will tend to be enduring, socially accepted, resistant to change and not directly reliant on rewards or monitoring for its persistence (Oliver, 1992). An institutionalized competence also signals a cultural willingness among employees to commit resources (Oliver, 1997) in support thereof.

We now turn to the final principle defining the SEMBI product. We will then define the principles of SEMBI capability improvement processes next section and provide an initial evaluation of the entire SEMBI design.

1. Principle 8: Adapting competences are designed from a dynamic capability lens

A dynamic capability perspective is an extension of a capabilities approach and its appeal lies in its potential to offer a solution to the rigidities inherent in capabilities over time; thus, there is a tendency to make capabilities more dynamic (Barreto, 2010; Laaksonen & Peltoniemi, 2018). The guiding logic is the core of operating capabilities; additional dynamic functions are designed to overcome the inherent risks that become stiff and trapped (Schreyögg & Kliesch‐Eberl, 2007). While capabilities are means through which resources and competences are configured, dynamic capabilities can be thought of as means through which resources and competences are reconfigured and thus, over time, are central to competence building. Consistent with the capability approach, the dynamic capability approach argues that competitive advantage is not sustainable and needs to be updated. Therefore, dynamic capabilities cannot be the source of sustainable competitive advantage, but the source of renewed competitive advantage.

Although all dynamic capability approaches are directed toward effecting change, there are remarkably different theories of dynamic capabilities (Schreyögg & Kliesch‐Eberl, 2007). The work of Teece et al. (1997) is the most notable, and SEMBI adopts this approach for our kernel theory. Following Teece et al. (1997), a dynamic capability is conceived to be the mechanism for adapting, integrating and reconfiguring integrated clusters of resources and capabilities to match the requirements of a changing environment. The term “dynamic” refers to the capacity to renew competences. They conceptualize three dimensions: position, path and process. The four SEMBI practices in the “Adapt to change” competence is derived from these, as shown in Table 9.

“Determine the SEMBI capability position” practice corresponds to Teece’s position dimension and is used to determine both internal and external positions of SEMBI capability. The internal position relates to the specific set of SEMBI resources available in a firm. The external position refers to the specific market position/assets of the firm. This external position acknowledges that a firm’s current position limits, to a certain extent, the future decisions a firm can reach and realize.

The “Analyze SEMBI paths” practice corresponds to Teece’s paths dimension and is used to analyze the history of SEMBI of an organization. Dynamic capabilities point out that a company's current position is largely shaped by patterns that have evolved in the past. Furthermore, a company's future direction depends on its current path and shaping forces.

Both “Apply organization learning” and “Reconfigure the SEMBI resources” practices comprise the process dimension. The core idea of the dynamic capability process dimension is to extend the scope of process constructs by including the learning and reconfiguration of resource processes to ensure permanent adaptation and change in the organization. The practice of “Applied Organizational Learning” corresponds to the process of learning “learning,” which covers the process of incremental improvement (modifying current positions) and identifying new opportunities. The practice of “Reconfigure the SEMBI resources” corresponds to the reconfiguration of the resources process, changing a company's SEMBI resource structure through early-warning monitoring of environmental discontinuities and subsequent fundamental changes.

Finally, we summarize the SEMBI product design in Table 10.

3.3 Meta-design of SEMBI capability improvement method

SEMBI–CMM comprises two dimensions of concern. In the prior section, we focused on the SEMBI capability model; in this section, we address capability improvement and the method dimension.

CMM generally can be divided into two architectures: staged and continuous (Team, 2002), and both architectures define maturity as a rating level on a predefined path for improvement. The staged architecture rates the whole organization based on a proven grouping and ordering of competences and associated organizational relationships; the continuous architecture rates each of the competences in relation to strategic business objectives (Paulk, 1996; Team, 2002). Both architectures explicitly or implicitly share a rationale about maturity as a

… theory about how organizations or processes are ‘supposed’ to work and/or about how organizational change is envisaged. … It is important to be clear about this underlying rationale as it impacts on the interpretation of results and affects suggestions of how a chosen unit of analysis should change to improve its performance. (Maier et al., 2009).

Building on this prior work, SEMBI–CMM is designed as a staged model rated with six maturity levels, and each level comprises some of the SEMBI competences that we have prescribed in the SEMBI capability model shown above in Table 9. However, one might question the rationale behind staging competences into processes. In fact, SEMBI–CMM is a process theory that focuses on the sequence of activities to explain how and why particular outcomes evolve over time. While such a process theory is essential for CMM, no kernel theories exist to support it. Lacking kernel theories opens CMM to the criticism it is built upon ad hoc methods (Bach, 1994; Biberoglu & Haddad, 2002; Hansen et al., 2004). Our work is aimed at filling the gap in kernel theories for maturity models.

Prevailing wisdom holds that (1) maturity modeling is an evolutionary process and change cannot be achieved in one great leap (Kruger, 2005); (2) staging is a strategy to guide those few improvement activities that will be most helpful if implemented immediately because most organizations can only focus on a few process improvement activities at a time (Paulk, 1996); and (3) CMM embeds an organization change approach that is incremental and learning oriented (Maier et al., 2009).

Montealegre (2002) developed a process model of capability development that departs from most literature in RBT and shows how a firm develops, manages and deploys capabilities. This model illustrates that (1) capability development is a cumulative and expansive process where path dependency matters; (2) capability development can be strategically planned, one step at a time over time; and (3) capability development is not a black box, it is not random, and instead, is inherent in the firm's overall strategy formulation and implementation. The model further conceptualizes capability development into three stages: Establishing Direction, Focusing on Strategy Development and Institutionalizing the Strategy. Each stage calls upon the competences of strategizing, flexibility, integrating and engendering trust.

Helfat and Peteraf (2003) model of capability lifecycle differs from Montealegre (2002) and posits both a general pattern of organizational capability development and a set of possible paths. They argue that the framework is general enough to incorporate the emergence, development and advancement of virtually any type of capability into any type of organizational setting, from small start-ups to large, diverse companies. The life cycle consists of four stages: founding, development, maturity and finally transformation.

Aligned with the kernel theories advanced by Montealegre (2002) and Helfat and Peteraf (2003), SEMBI–CMM design treats the staged competences as a SEMBI capability development process, and we sequence the competences according to their nature and function. The SEMBI–CMM process offers finer-grained stages (Ad Hoc, Initial, Performed, Developed, Committed and Optimizing) that are similar to general CMM in form. This discussion leads to Principle 9.

1. Principle 9:Maturity is defined by a staged capability development process

Table 11 presents the stages in SEMBI–CMM and also depicts their mapping to process models from Montealegre (2002) and Helfat and Peteraf (2003). Note that the competences presented here were developed in Section B and summarized in Table 9. The SEMBI process defines the Ad Hoc Stage (Stage 0) to represent that no competences are formally or purposely developed. In Stage 1, three competences are developed. While the “Define SEMBI contribution” and “Determine SEMBI system architecture” competences help establish direction, the “Establish SEMBI management leadership” competence is the foundational competence in this Initial Stage.

In the Performed Stage “Manage SEMBI human resources” and “Manage deliver and support” competences are evident. These competences ensure that resources are deployed in line with the strategic direction of the organization.

The “Define evidence-based management (EBM) process” competence defines the Developed Stage, as EBM competence yields effective and efficient execution of strategy. The Committed Stage corresponds to the two culture competences “Establish EBM ba” and “Cultivate EBM culture” that institutionalize and operationalize the former competences. At the Optimizing Stage, the “Adapt to change” competence emerges. While we designed all prior stages to establish incremental capability within given resource constraints, the Optimizing Stage is designed to establish new capability by reconfiguring competences and resources in a dynamic environment.

SEMBI–CMM capability building and operationalization comprise stages designed to manage the conflicting requirements of exploitation and exploration (Schreyögg & Kliesch‐Eberl, 2007). This design requires that competences be developed in sequence; by extension, a stage qualifies as mature if and only if all the component competencies qualify.

In the SEMBI capability model, competence is prescribed as a practice-oriented process embedding a best practice pattern for problem-solving. However, although the pattern provides practitioners with an action goal as a guide, how the practitioners reach the goal is essentially determined by the practitioners’ continuous practices and through the tacit accumulation of experience and sporadic acts of creativity (Zollo & Winter, 2002; Marjanovic & Seethamraju, 2008).

1. Principle 10: Maturity is measured by practice-oriented processes

In SEMBI–CMM, when the practices of competence have reached a threshold level of reliability we deem that competence “qualified.” A qualified competence is repeatable and quite distinct from ad hoc problem-solving. Ad hoc problem-solving involves a high-paced, contingent, opportunistic and perhaps creative search for satisfactory alternative behaviors; as such, it might be characterized as non-repeatable “firefighting” (Winter, 2003). Similarly, a “taking a first cut” practice does not constitute a qualified competence, although it can trigger the building of a qualified competence (Helfat & Peteraf, 2003).

Certainly “qualified” is a concept of degree, which itself can be quantified to different levels. To avoid unnecessary complexity, we rate both competence and each component practice with two levels: qualified or unqualified. Therefore, if the number of qualified component practices reaches a threshold, the competence is qualified. Likewise, if all the competences of a stage are qualified, the organization achieves maturity at this stage. Thus, SEMBI-CMM evolution method is essentially a dual process method. While the staged process is external to competence and is used to define the maturity of the whole organization, the practice-oriented process is internal to competence and used to measure it.

Note that an organization acquiring a specific level of maturity does not imply that all the competences at this level have been maxed out. Organizations may differ in the efficiency or effectiveness of a specific type of competence. To say that a competence is qualified simply means that it has achieved some minimum level of functioning that allows the exercise to be performed repeatedly and reliably. Some versions of a competence are better than others (Helfat & Peteraf, 2003).

Moreover, in most cases, that a competence is qualified does not imply the competence cannot evolve again. Capability is essentially a historical concept that combines experience with current problem-solving practices to look to future directions (Schreyögg & Kliesch‐Eberl, 2007). Stressing the historical nature of a competence acknowledges that time is a basic dimension of a competence.

Competence evolution is somewhat of a chain reaction triggered by an initial event that establishes a competence trajectory. Competence evolution takes time, and the specific way in which time has been taken (i.e., the intensity, frequency and duration of social interactions) is relevant to the gestalt of a competence (Schreyögg & Kliesch‐Eberl, 2007). Note also that the competences in our design are not independent. The competences in Stages 3 and 4 are intentionally designed as “meta-rules” – routines for changing routines – to systematically and repeatedly produce improvement for prior competences established in Stages 1 and 2 (Zollo & Winter, 2002). So, after the competences in Stages 1 and 2 (Initial and Performed, respectively) reach the minimum level of functionality and enter Stages 3 and 4 (Developed and Committed, respectively), the competences in Stages 1 and 2 will continuously coevolve with competences in Stages 3 and 4. As noted earlier, the competence in Stage 5 (Optimizing) also has an impact on former competences by reconfiguring them.

The evolution method, although primarily a process theory, also embeds some variance relationships. The process and variance relationships function together to allow SEMBI capability to evolve both at an organization level and at each competence level. In Figure 2, the horizontal axis represents the staged process that directs SEMBI capability development and the rate of organizational maturing. The vertical axis represents the “quality” of a competence; i.e., a practice-oriented process or the overall skillfulness of the team in executing the practices of the competence.

Competences with quality above the threshold level are qualified; otherwise, they are unqualified. Along with the evolution of the staged process, each practice-oriented process (PoP) will coevolve not only due to the accumulation of practices but also due to the casual relationship embedded in the method. There are several practice-oriented processes in the method; and to illustrate possible interactions and co-evolution, we map two of them in Figure 2. The graphs show that PoP-A begins as an Ad Hoc practice with very low quality and gradually, steadily becomes more mature and of higher quality; plateauing well above the threshold for a qualified process. While PoP-B begins at the Initial Stage (more mature than PoP-A), the initial quality is also very low, remains lower than PoP-A and requires more time to qualify; PoP-B eventually achieves a higher level of quality on the maturity dimension than PoP-A.

To conclude the theory development portion of our work, Table 12 summarizes the mid-range and kernel theories comprising the SEMBI–CMM method design.

4. Testable hypotheses stemming from SEMBI-CMM meta-requirements

We use hypotheses to determine whether the meta-design satisfies the meta-requirements as per (Walls et al., 1992). For SEMBI–CMM, the meta-design is SEMBI capability model, and the meta-requirement is “SEMBI-CMM should support the SEMBI success.” We decompose this meta-requirement into three detailed meta-requirements comprising production, use and net benefit. Accordingly, the testable design product hypothesis states, “The development of the prescribed SEMBI capability has a positive effect upon SEMBI success.” This hypothesis can be deconstructed into three component hypotheses corresponding to the three meta-requirements. Table 13 shows these hypotheses and their relationship to the meta-requirements.

As shown in Table 12, design process hypotheses check whether the design method results in an artifact consistent with the meta-design (Walls et al., 1992). That is, SEMBI capabilities prescribed by the SEMBI capability model can be instantiated by using the prescribed evolution method. Therefore, the design process hypothesis proposes that “It is feasible to develop SEMBI capability based on the SEMBI evolution method.” Finally, the SEMBI–CMM model specifies that the design method is the SEMBI evolution method, and the meta-design is the SEMBI capability model.

5. SEMBI evaluation survey and results

Evaluation is essential to demonstrate the utility, quality and efficacy of the design output in design science research (Hevner et al., 2004; Baskerville et al., 2018). For a deep understanding of the utility, quality and efficacy of our research output, a longitudinal multiple-case study may be required. However, design science is also a spiral process in which building and evaluation activities interact cyclically (Baskerville et al., 2018; Venable et al., 2016). In that spirit and considering this early stage in our research, we conducted a pilot survey-based evaluation of two real-world environments based on the Content, Context and Process (CCP) framework (Stockdale & Standing, 2006). In the CCP framework, the Content construct specifies what is to be evaluated, the Context construct specifies who participates in the evaluation and what factors may influence the evaluation, and the Process construct specifies how the evaluation is to be implemented.

In our evaluation, the content construct is the hypotheses advanced by design theory. In contrast to the positivist view where the hypotheses focus on past actions, the two hypotheses in our design theory are based on the realist view, which focuses on future actions (Venable et al., 2016). The implication is that these hypotheses will be evaluated by their perceived utility to solve the problems. In the SEMBI–CMM framework, both the SEMBI capability model and the evolution method are prescriptions (or future actions) about a class of problems, and the two hypotheses concern whether these prescriptions can solve the problems. Our evaluation was conducted in two firms in China in 2011. We selected a convenience sample of 21 individuals from these firms and their respective consulting and implementation partners and ended up with 16 useable responses. Table 14 identifies our hypotheses (Content row), briefly describes the two firms (Context row) and identifies the 4-step process (Process row) we used.

The questionnaire is shown in Table 15, with the results provided in Table 16.

Table 16 converts raw results from Table 14 to percentages, with values above 50% shown in italic font. With this initial pilot survey, respondents generally indicated that all practices contribute to one or more SEMBI success stages. These results are preliminary per se, as we chose a small convenience sample and explicitly provided background SEMBI-CMM information before administering the survey.

6. Conclusion and contributions

EBM considers two types of evidence to support decision-making; i.e., “Big E Evidence” and “Little e evidence” (Rousseau, 2006). The extant studies indicate that SEM, as a Big E Evidence-based DSS, BI, as a Little e evidence-based DSS, are intertwined and complementary in nature.

In an effort to advance this line of research aligning strategic decision-making with operationalized business intelligence, we employ a design science perspective to address the question of how an organization can successfully implement SEMBI. We coined and use the term SEMBI to represent the union of SEM and BI in our work.

This piece focuses on three main topics related to SEMBI: building SEMBI, designing the SEMBI design product and the design process. First, the design model of SEMBI is discussed. The success of SEMBI is broken down into a general meta-requirement, which requires achieving deeper levels of detail and success in IS research. A three-step process model is generally agreed upon for achieving this success, which includes production, use and net benefit. Secondly, the capability model of SEMBI is examined. Eight model design principles are defined and explored, followed by an exploration of the method dimension. Finally, the capability improvement method of SEMBI is addressed. This includes a discussion of capability improvement and the method dimension.

To better understand the usefulness, quality and effectiveness of our research output, we conducted an initial evaluation of the SEMBI model with a convenience sample. However, to gain a deeper understanding, a longitudinal multiple-case study may be necessary. To determine if the meta-design meets the meta-requirements, hypotheses were formulated and tested. In addition, we conducted a pilot survey-based evaluation of two real-world environments using the CCP framework.

We summarize our research contributions to researchers and practitioners in Table 17, based on Hevner et al. (2004) seven guidelines for design science research. Design science is a spiral process where building and evaluation interact. Evaluation has meaning only to the extent that one can measure design output (Venable et al., 2016). Since acceptable objective measures are not yet available for our research at this stage, we use the stakeholders’ attitudes and subjective judgments as proxies (Baskerville et al., 2018). However, such measures are criticized by some researchers, especially when the measures do not tap into the cognitive components that reflect the direct experience and are held with confidence (Melone, 1990). Moreover, a more rigorously designed evaluation is warranted to adequately test stakeholder attitudes. Finally, a longitudinal and cross-section multi-case study needs to be introduced based on a real application of SEMBI-CMM. Our future work aims to attain a deeper understanding and perhaps provide a better design for SEMBI-CMM through an extensive spiral process of objective testing and re-building.

Figures

Figure 1

SEMBI capability model

Figure 2

SEMBI capability development

Table 1

Summary of the relevant studies of SEMBI

Authors	Research purpose	BI included	Research outputs	Theoretical purpose
Brignall and Ballantine (2004)	To study the interrelationships among SEM systems, PMM and organizational change programs within context, content and process model		Models	Prediction
ZAITSEVA (2014)	A justified procedure for choosing flexible strategic business management methods is based on using their capacities for self-organization		Methods	Explanations
Sestino (2016)	By adopting a process-based approach, businesses can improve their competitiveness	V	Instantiations	Explanation and prediction
Sen et al. (2017)	Exploring the integration of a balanced scorecard into innovative strategic management, providing a framework of interrelated issues based on the existing literature and application areas		Models	Explanations
Faizova et al. (2018)	To outline the prospects and directions for improving the methodology of strategic enterprise management under the conditions of its evolution into balanced enterprise management		Methods	Explanation and prediction
Akhmetshin et al. (2018)	Development of a guideline for analysis of the economic activity of an enterprise to control and ensure the interaction of tasks and functions of management in the current and strategic aspects in the conditions of innovative development		Instantiations	Analysis
Teece (2019)	The application of capability theory allows intellectual blinders to be removed and an understanding of differential firm-level resource allocation and performance to emerge	V	Constructs	Explanations
Regent et al. (2019)	Development of approaches to improving a tourism enterprise's strategy management in the international market		Models	Analysis
Melnyk and Zlotnik (2020)	To systematize and compare the main approaches to strategic management		Models	Analysis
Khudyakova et al. (2020)	To improve the model and mechanism of strategic management of the investment policy of the enterprise in the context of the environmental and economic status of the business		Models	Explanations
Mathrani (2021)	To investigate the BI practices critical in creating meta-knowledge successfully for strategy-focused analytical decision-making	V	Methods	Explanation and prediction
Current study	To present a design theory framework and build a model dimension using eight principles serving as mid-range theories	V	Models and Methods	Design and Action

Source(s): Table by the authors

Table 2

ISDT framework for SEMBI–CMM

Design product/model dimension of SEMBI–CMM	Design process/method dimension of SEMBI–CMM
A. Meta-requirements	A. Meta-requirements
B. Meta-design	B. Design method
C. Mid-range theories	C. Mid-range theories
D. Kernel theories	D. Kernel theories
E. Testable design product hypotheses	E. Testable design process hypotheses

Source(s): Gregor, 2002; Walls et al., 1992; Stockdale & Standing, 2006

Table by the authors

Table 3

Design principles for SEMBI

Dimension	Principle#	Principle
Covered in Section: Meta-design of SEMBI capability model
Model	Principle 1	Capability in SEMBI–CMM is constructed within resource-based theory (RBT) and the IS Capability lens
Model	Principle 2	SEMBI is a dynamic capability distilled from operational capabilities
Model	Principle 3	SEMBI capability model is prescribed a priori
Model	Principle 4	SEMBI competence is operationalized by a practice-oriented process
Model	Principle 5	Strategizing competences are designed from a business-IT alignment lens
Model	Principle 6	Deploying competences are designed to match strategizing competences
Model	Principle 7	Complementing competences are designed from an institutional lens
Model	Principle 8	Adapting competences are designed from a dynamic capability lens
Covered in Section: Meta-design of SEMBI capability improvement method
Method	Principle 9	Maturity is defined by a staged capability development process
Method	Principle 10	Maturity is measured with practice-oriented processes

Source(s): Table by the authors

Table 4

SEMBI capability constructs

Level	Construct	Definition
Enterprise	Capability	The strategic application of competences (Kangas, 1999; Peppard & Ward, 2004)
Organizing	Competence	A firm’s capacity to deploy resources, usually in combination, using organizational processes, to effect a desired end (Amit & Schoemaker, 1993)
Resource	Resource	Stocks of available factors that owned or controlled by a firm (Amit & Schoemaker, 1993)

Source(s): Table by the authors

Table 5

Strategizing competences designed from business-IT alignment lens

Competence	Practice
Define SEMBI contribution (i.e., the IS strategy)	Identify and innovate SEM process
	Align SEM process with BI
	Identify and innovate operational BI process
	Align operational process with BI
Determine SEMBI system architecture (i.e., the IT strategy)	Determine ETL plan
	Determine Data Warehouse architecture
	Determine Data Mining technique
	Determine SEMBI application

Source(s): Table by the authors

Table 6

Deploying competences matched with strategizing competences

Competence	Practice
Manage deliver and support	Manage service-level agreements
	Manage third-party services
Manage SEMBI human resources	Identify and develop SEMBI human resources
	Exploit SEMBI human resource

Source(s): Table by the authors

Table 7

Governance competences

Competence	Practices
Establish SEMBI management leadership	Build SEMBI competence center
	Make SEMBI capability improvement plan
	Communicate SEMBI goal and direction

Source(s): Table by the authors

Table 8

Culture competences

Competences	Cultivate EBM culture	→	Establish EBM ba	→	Define EBM process
Practices	Cultivate clan culture	→	Establish originating ba	→	Define EBM socialization mode
	Cultivate adhocracy culture	→	Establish dialoging ba	→	Define EBM externalization mode
	Cultivate market culture	→	Establish systemizing ba	→	Define EBM combination mode
	Cultivate hierarchy culture	→	Establish exercising ba	→	Define EBM internalization mode

Source(s): Table by the authors

Table 9

Adapting competence designed from a dynamic capability lens

Competence	Practice
Adapt to change	Determine the SEMBI capability position
	Analyze SEMBI paths
	Apply organization learning
	Reconfigure the SEMBI resources

Source(s): Table by the authors

Table 10

Meta-design, mid-range theories and kernel theories for SEMBI product

	Categories	Competences	Practice
Meta-design	Strategizing	Define SEMBI contribution	Identify and innovate SEM process
			Align SEM process with BI
			Identify and innovate operational BI process
			Align operational process with BI
		Determine SEMBI system architecture	Determine ETL plan
			Determine Data Warehouse architecture
			Determine Data Mining technique
			Determine SEMBI application
	Deploying	Manage SEMBI human resources	Identify and develop SEMBI human resources
			Exploit SEMBI human resource
		Manage deliver and support	Manage service-level agreements
			Manage third-party services
	Complementing	Establish SEMBI management leadership	Build SEMBI competence center
			Make SEMBI capability improvement plan
			Communicate SEMBI goal and direction
		Define EBM process	Define EBM socialization mode
			Define EBM externalization mode
			Define EBM combination mode
			Define EBM internalization mode
		Establish EBM ba	Establish originating ba
			Establish dialoging ba
			Establish systemizing ba
			Establish exercising ba
		Cultivate EBM culture	Cultivate clan culture
			Cultivate adhocracy culture
			Cultivate market culture
			Cultivate hierarchy culture
	Adapting	Adapt to change	Determine the SEMBI capability position
			Analyses SEMBI paths
			Apply organization learning
			Reconfigure the SEMBI resources
Mid-range theories	Principle 1: Capability in SEMBI–CMM is constructed within RBT and the IS Capability lens Principle 2: SEMBI is a dynamic capability distilled from operational capabilities Principle 3: SEMBI capability model is prescribed as a priori Principle 4: SEMBI competence is operationalized by a practice-oriented process Principle 5: Strategizing competences are designed from a business-IT alignment lens Principle 6: Deploying competences are designed to match strategizing competences Principle 7: Complementing competences are designed from an institutional lens Principle 8: Adapting competences are designed from a dynamic capability lens
Kernel theories	CMM (Bach, 1994; Paulk, 1996; Biberoglu & Haddad, 2002; Team, 2002; Hansen et al., 2004; Maier et al., 2009) RBT and IS capability (Barney, 1991, 1996; Stalk, 1992; Amit & Schoemaker, 1993; Black & Boal, 1994; McGrath et al., 1995; Feeny & Willcocks, 1998; Kangas, 1999; Bharadwaj, 2000; Bowman & Ambrosini, 2000; Bassellier & Benbasat, 2004; Peppard & Ward, 2004; Wade & Hulland, 2004; Aral & Weill, 2007; Prasad et al., 2009) Dynamic capability (Teece et al., 1997; Zollo & Winter, 2002; Winter, 2003; Helfat et al., 2009; Schreyögg & Kliesch‐Eberl, 2007; Stoel & Muhanna, 2009; Barreto, 2010; Laaksonen & Peltoniemi, 2018) Business-IT alignment (Reich & Benbasat, 1996) SEMBI research (Srivastava & Chen, 1999; Williams & Williams, 2003; Darmont et al., 2005; Frolick & Ariyachandra, 2006; March & Hevner, 2007; Meredith, 2008; Wixom et al., 2008; Panian, 2009; Popovič et al., 2009) Knowledge management (Nonaka, 1994; Nonaka et al., 2000; Schultze & Boland, 2000; Carlile, 2002; Wenger et al., 2002; Janz & Prasarnphanich, 2003; Davenport, 2005; Gray & Densten, 2005; Marjanovic & Seethamraju, 2008) Evidence based management (Pfeffer & Sutton, 2006; Rousseau, 2006) Institutional theories (Oliver, 1992, 1997; Scott, 2013; Svejvig, 2009)

Source(s): Table by the authors

Table 11

SEMBI–CMM staged process

Maturity level	0 Ad hoc	1 Initial	2 Performed	3 Developed	4 Committed	5 Optimizing
SEMBI Competences		Establish SEMBI management leadership	Manage SEMBI human resources	Define EBM process	Establish EBM ba	Adapt to change
		Define SEMBI contribution	Manage deliver and support		Cultivate EBM culture
		Determine SEMBI system architecture
Montealegre (2002)		Establishing Direction	Focusing on the Strategy Development		Institutionalizing the Strategy
Helfat and Peteraf (2003)		Founding	Development		Maturity	Transformation

Source(s): Table by the authors

Table 12

Mid-range and kernel theories for SEMBI design process

Mid-range theories	•Principle 9: Maturity is defined by a staged capability development process •Principle 10: Maturity is measured with practice-oriented processes
Kernel theories	CMM (Bach, 1994; Paulk, 1996; Biberoglu & Haddad, 2002; Team, 2002; Hansen et al., 2004; Maier et al., 2009) Process model of capability development (Montealegre, 2002; Helfat & Peteraf, 2003) RBT and IS capability (Zollo & Winter, 2002; Schreyögg & Kliesch‐Eberl, 2007) Practice-oriented process (Marjanovic & Seethamraju, 2008)

Source(s): Table by the authors

Table 13

Hypotheses resulting from SEMBI–CMM meta-requirements

Meta-requirements	MR: SEMBI–CMM should support the SEMBI success
	MR1: SEMBI–CMM should support SEMBI production	MR2: SEMBI–CMM should support SEMBI use	MR3: SEMBI–CMM should support SEMBI net-benefit realization
Design product	SEMBI capability model
Design product hypotheses	H1: The development of the prescribed SEMBI capability has a positive effect upon SEMBI success
	H1a: The development of the prescribed SEMBI capability has a positive effect upon SEMBI production	H1b: The development of the prescribed SEMBI capability has a positive effect upon SEMBI use	H1c: The development of the prescribed SEMBI capability has a positive effect upon SEMBI net benefit realization
Design process	SEMBI evolution method
Design process hypotheses	H2: It is feasible to develop SEMBI capability based on the SEMBI evolution method

Source(s): Table by the authors

Table 14

Evaluation framework

Content	H1: The development of the prescribed SEMBI capability has a positive effect to SEMBI success ■ H1a: The development of the prescribed SEMBI capability has a positive effect upon SEMBI production ■ H1b: The development of the prescribed SEMBI capability has a positive effect upon SEMBI use ■ H1c: The development of the prescribed SEMBI capability has a positive effect upon SEMBI net benefit realization H2: It is feasible to develop SEMBI capability based on the SEMBI evolution method
Context	Case 1: A famous retail company in Shenzhen, China The company is planning to use a balanced scorecard system A consulting company from Shanghai, China will help the company plan the implementation A “strategy management” department similar to a SEMBI competence center has been established An employee has been trained to know the basic balanced scorecard principles
	Case 2: The China branch of a famous logistics company The company has implemented a BI system A company in Shenzhen helped to implement the system An employee with one year experience is using the system
Process	Select stakeholders from Two consulting companies Implementation companies (clients of the consulting companies’ stakeholders)
	Educate the stakeholders on the SEMBI-CMM, so they understand the concepts in the survey
	Survey the stakeholders with a questionnaire
	Analyze the data

Source(s): Table by the authors

Table 15

Questionnaire and raw results

Questionnaire generated 16 useable responses of 21 requested
SEMBI capability model			SEMBI success stages (16 lots in total)
Categories	Competences	Practice	Production	Use	Net benefit
Part 1: If you think the practice prescribed in the SEMBI capability model has a positive effect upon the stage of SEMBI success, please mark a√in the correspondence blank
Strategizing	Define SEMBI contribution	Identify and innovate SEM process
		Align SEM process with BI
		Identify and innovate operational BI process
		Align operational process with BI
	Determine SEMBI system architecture	Determine ETL plan
		Determine Data Warehouse architecture
		Determine Data Mining technique
		Determine SEMBI application
Deploying	Manage SEMBI human resources	Identify and develop SEMBI human resources
		Exploit SEMBI human resource
	Manage deliver and support	Manage service-level agreements
		Manage third-party services
Complementing	Establish SEMBI management leadership	Build SEMBI competence center
		Make SEMBI capability improvement plan
		Communicate SEMBI goal and direction
	Define EBM process	Define EBM socialization mode
		Define EBM externalization mode
		Define EBM combination mode
		Define EBM internalization mode
	Establish EBM ba	Establish originating ba
		Establish dialoging ba
		Establish systemizing ba
		Establish exercising ba
	Cultivate EBM culture	Cultivate clan culture
		Cultivate adhocracy culture
		Cultivate market culture
		Cultivate hierarchy culture
Adapting	Adapt to change	Determine the SEMBI capability position
		Analyses SEMBI paths
		Take the organization learning
		Reconfigure the SEMBI resources
Part 2: If you think it is feasible to develop SEMBI capability based on the SEMBI evolution method, please mark a √ in the following blank

Source(s): Table by the authors

Table 16

Questionnaire results in percentage form

Useable questionnaire rate 0.76 (16/21)
SEMBI capability model			SEMBI success stages
Categories	Competences	Practice	Production	Use	Net benefit
Part 1: If you think the practice prescribed in the SEMBI capability model has a positive effect to the stage of SEMBI success, please mark a√in the correspondence blank
			Positive effect/Total
Strategizing	Define SEMBI contribution	Identify and innovate SEM process	0.63(=10/16)	0.31	0.56
		Align SEM process with BI	0.50	0.75	0.19
		Identify and innovate operational BI process	0.44	0.44	0.69
		Align operational process with BI	0.63	0.69	0.31
	Determine SEMBI system architecture	Determine ETL plan	0.94	0.13	0.13
		Determine Data Warehouse architecture	0.94	0.44	0.19
		Determine Data Mining technique	0.69	0.63	0.88
		Determine SEMBI application	0.44	0.69	0.63
Deploying	Manage SEMBI human resources	Identify and develop SEMBI human resources	0.94	0.94	0.19
		Exploit SEMBI human resource	0.69	0.94	0.69
	Manage deliver and support	Manage service-level agreements	0.69	0.56	0.63
		Manage third-party services	0.75	0.63	0.75
Complementing	Establish SEMBI management leadership	Build SEMBI competence center	0.94	0.75	0.06
		Make SEMBI capability improvement plan	0.19	0.88	0.69
		Communicate SEMBI goal and direction	0.88	0.38	0.38
	Define EBM process	Define EBM socialization mode	0.44	0.63	0.19
		Define EBM externalization mode	0.81	0.38	0.06
		Define EBM combination mode	0.75	0.38	0.13
		Define EBM internalization mode	0.44	0.63	0.13
	Establish EBM ba	Establish originating ba	0.25	0.69	0.38
		Establish dialoging ba	0.44	0.81	0.13
		Establish systemizing ba	0.31	0.69	0.25
		Establish exercising ba	0.44	0.81	0.25
	Cultivate EBM culture	Cultivate clan culture	0.31	0.75	0.75
		Cultivate adhocracy culture	0.38	0.75	0.69
		Cultivate market culture	0.31	0.69	0.75
		Cultivate hierarchy culture	0.38	0.69	0.69
Adapting	Adapt to change	Determine the SEMBI capability position	0.75	0.38	0.19
		Analyses SEMBI paths	0.75	0.38	0.19
		Take the organization learning	0.31	0.94	0.56
		Reconfigure the SEMBI resources	0.38	0.81	0.81
Part 2: If you think it is feasible to develop SEMBI capability based on the SEMBI evolution method, please mark a √ in the following blank

Note(s): Values exceeding 50% are displayed in italics

Source(s): Table by the authors

Table 17

SEMBI through the lens of Hevner et al. (2004) seven guidelines for design science research

Design as an artifact	SEMBI-CMM is designed as an abstract artifact with both a SEMBI capability model and a SEMBI capability evolution method
Problem Relevance	The research provides relevance to a real business problem of achieving SEMBI success
Design Evaluation	The utility of the prescribed design is evaluated based on a Context, Context, Process (CCP) framework
Research Contributions	The research provides a methodology for organizations to achieve SEMBI success and meanwhile provides new knowledge in the form of design theory for researchers
Research Rigor	The research is conducted based on an ISDT framework and is initially tested based on a hypothetic-deductive logic
Design as a Search Process	The research searches for a solution for SEMBI success from many sources of kernel theories and through a mid-range theory design. The build-evaluation interaction aids a cyclical search for better solutions
Communication of Research	The research provides rich information to management audiences on the importance of the SEMBI-CMM and meanwhile provides consulting and implementation practitioners a detailed model and method for their SEMBI practices. It also provides other researchers with an established theory that can be extended in the future

Source(s): Table by the authors