Conceptualizing Business Process Standardization: A Review and Synthesis

Conceptualization helps to describe a phenomenon, and is an essential pre-requisite to theorizing (Gerring 2011; Van de Ven 2007). Constructs designate conceptual abstractions of a phenomenon, whose meaning is specified through conceptualization (Schwab 1980). The success of research depends on clear understanding of constructs (Zhang et al. 2016). Well-defined constructs form the basic building blocks of a theory (Weber 2012; Markus and Saunders 2007). But, unclear constructs with ambiguous meaning will jeopardize construct validity (Schwab 1980), hinder quality research outcomes and complicate comparison and accumulation of knowledge in the area. The importance of conceptualization has been well argued (MacKenzie et al. 2011; Zhang et al. 2016).

Perspectives on conceptualization vary. Zhang et al. (2016) presents a concise summary of different views on how conceptualization is understood. To many researchers, conceptualization entails providing a definition of a construct that captures its dimensions thereby communicating what it is (MacKenzie et al. 2011). This definition of the construct is then used to guide operationalization and construct validation procedures by devising a set of items that fully capture the meaning of the construct. Confused and ambiguous conceptual definitions engender misunderstanding and inconsistent measures of a concept (Wacker 2004). Many researchers commence with a seemingly clear idea, but often after considerable effort has been expended (i.e., by collecting data) come to realize their original thinking was less clear than they had first believed, where after, consequences of changes to the research can be costly (DeVellis 2012). Undetected failure to adequately define a construct may result in invalid conclusions about relationships with other constructs due to deficient indicators (MacKenzie et al. 2011), hence producing unreliable theoretical conclusions. These dangers strongly convey the need for a clear conceptual definition from the outset, to enable accurate measurement and to make valuable cumulative contributions to an evolving knowledge base.

MacKenzie et al. (2011) suggest examining how the focal construct has been used in prior theoretical and empirical research and then to use that knowledge to inform conceptualization of the construct of interest. Burton-Jones and Straub (2006) suggest identifying the ‘structure’ of the concept as a first step towards conceptualization. Having identified the structure, they recommend identifying measures for each of the elements representing the construct.

A business process is commonly defined as “a set of logically related tasks performed to achieve a defined business outcome” (Davenport and Short 1990, p. 13). A business process comprises several events and activities. Events happen automatically and trigger the execution of a process. They do not have any duration. An activity is a single unit of work, also known as task (Dumas et al. 2018). A feature of business processes is that several activities are logically ordered and have a clearly defined input as well as output to achieve a business goal (Weske 2019). Business process activities can be performed manually by employees or with the help of information systems (Weske 2007). There are four major sub-dimensions of a business process: (i) activities—the main actions in a process, (ii) workflow—refers to the linking of activities, (iii) resources—agents that are used for adding value to activities, and (iv) entities—“objects processed by resources” (Muenstermann and Weitzel 2008, p. 3).

For business processes to be standardized, there must be a ‘standard’—an approved referent-process to adopt and benchmark against. According to ISO (1996, p. 1) “Standards are documents, established by consensus and approved by a recognized body, that provide, for common and repeated use, rules, guidelines or characteristics for activities or their results, aimed at the achievement of the optimum degree of order in a given context”. In a BPS context, the standard is usually an enhanced master process used as a point of reference, against which existing business processes (and their variants) are modelled and adapted (see Sect. 4.1 for further details).

A variant of a business process is an alternative version of the same process; it achieves the same outcome, but by following a different sequence of activities (Zellner et al. 2015b). A business process model is a graphical representation of a business process (Reijers et al. 2011), which enables visualizing the flow of activities and interdependencies occurring in the business process.

Standards, in the context of BPS should not be confused with abiding by governance standards set for process management initiatives; for example, using organization-wide modelling-notation standards (such as Business Process Model and Notation (BPMN)) when deriving process models, or using project management templates/standards to manage the BPS initiative. While modelling-notation standards and/or project management templates may be applied in BPS efforts as supporting mechanisms, BPS is quite a different notion. BPS is centred around how a business process abides by and aligns with an enhanced master process, irrespective of the use of other tools and guidelines (such as modelling, project management standards).

Current discourses about BPS often use terms which are conflated with the core notion of BPS. Though related, these concepts are different to BPS and it is important that these concepts are not confused with BPS (Goel and Bandara 2016). This understanding is essential for construct clarity, for the development of precise concepts (Zhang et al. 2016; Suddaby 2010). According to Weber (2012), clearly defined constructs can delineate the boundary conditions of what phenomenon are included and what are not included, when theorizing. The need for concept clarity in the BPS context is emphasized by Goel and Bandara (2016), who through a systematic literature review, list and analyse several concepts often conflated with BPS; such as ‘Process integration’, ‘Process Compliance’, ‘Process harmonization’ and ‘Unification of variants’. Additionally, process modularization and reference process models are also concepts that are conflated with BPS.

Process integration relates to the overall coordination of business processes and activities (Narayanan et al. 2011). It focusses on integration of systems and services as well as secure sharing of data across multiple areas of the business to derive well interconnected business processes. Process integration can serve as a useful precedent for standardizing a business process. Process compliance refers to business processes adherence to prescribed or agreed upon norms (Seyffarth et al. 2017; Papazoglou 2011). Process compliance is an outcome of BPS. Process harmonization refers to aligning variants or homogenizing variants with a standard process (Muenstermann and Weitzel 2008), which is a stage of BPS, as will be explained in Sect. 4.1. Process harmonization is also referred to as unification of variants by some authors (e.g., Muenstermann et al. 2010a; Davenport 2005).

The concept of Process modularization is another closely related concept to BPS discussed by several authors (e.g., von Stetten et al. 2008; Muenstermann and Weitzel 2008). Process modularization entails breaking a process into parts which can be recombined to create custom alternatives of a business process (Tuunanen and Cassab 2011; Böhmann et al. 2018), and is a stage of BPS, as will be discussed later.

Another closely related topic is the notion of reference process model, which by definition is a process model that is reusable, displays exemplary practices, and has universal applicability (Rehse et al. 2017; vom Brocke 2007; Dumas et al. 2018). Examples include Supply Chain Operations References (SCOR) (Bolstorff and Rosenbaum 2007) and Information Technology Infrastructure Library (ITIL) (AXELOS 2020). While most reference models, as per the examples of SCOR and ITIL, are derived and governed by external bodies (such as the Supply Chain Council for SCOR), reference models can also be derived inductively from within a company (Scholta et al. 2019) and hence need not be an external process. A company may during the BPS process, choose a reference model as their ‘master process’. Even if a reference process model is chosen as a master process, additional steps need to be conducted (as will be explained in Sect. 4.1) to standardize the process. For example, the adaption of the reference model may still vary from process-to-process or case-to-case; especially at the more detailed levels. This is why simply choosing a reference process model is not the same as business process standardization.

During our analysis of what BPS is and the process of conducting BPS (as explained later in Sect. 4 ‘Study Findings and Contribution’), we were able to confirm that certain of these concepts are antecedents, others are consequences, and yet others are a part of the process of BPS. These concepts are revisited in succeeding discussion as useful.

Research with a focus on conceptualizing BPS is scarce, there being calls to improve understanding of the BPS concept (e.g., Muenstermann 2015; Wurm and Mendling 2020). Authors tend to talk around the notion of BPS, some addressing factors that influence BPS (e.g., Romero et al. 2015a; Schäfermeyer et al. 2010), with most focusing on the impact of BPS (e.g., Schäfermeyer et al. 2012; Wüllenweber et al. 2008), but with little consensus on what BPS is.

Muenstermann and Weitzel (2008) make strong arguments for the need to develop a better conceptual understanding of the BPS construct. In their attempt to conceptualise BPS, they suggest four criteria/steps essential to standardize a business process: document process, modularize process, isolate specificities, and ensure process excellence (see Table 4, p. 9 of Muenstermann and Weitzel (2008)). Their final definition of a standardized process is; “to standardize a process means to homogenize it against a standard process” (Muenstermann and Weitzel 2008). However, they note that they could not locate a clear definition of BPS and indicate that their four criteria are based on their professional experience, with some evidence from literature. Their limited evidence-base raises doubts regarding the sufficiency of their definition. Muenstermann (2015) reports a further attempt to address the question “What is BPS?”, where 119 publications deemed related to BPS were analysed. He argues that the analysis of these definitions identified conceptual themes which substantiate their 2008 definition of BPS (Muenstermann and Weitzel 2008). Although comprehensive and rigorous, Muenstermann’s (2015) analysis has limitations. First, the papers included in the pool were not all about BPS. This is evident from the definitions extracted (see Table 4 of Muenstermann (2015)), which pertained to themes such as ‘IT standards’, ‘Work standards’, ‘Reference model standards’ and more (the precise scope of the review cannot be ascertained as the search strategy¹ is not provided). Such a mixing of concepts is fraught. A good construct definition is derived by collating the key conceptual themes specifically pertaining to the focused-construct (MacKenzie et al. 2011). Also, there is little detail in Muenstermann (2015) on the analysis process followed (i.e., how they arrived at their definition from the conceptual themes). These identified limitations are acknowledged by Muenstermann (2015) himself who calls for future research to arrive at a ‘concise and precise definition of BPS’ and to ‘develop adequate measurement scales’ to empirically investigate BPS.

Romero et al. (2015a) also sought to synthesize existing research into a detailed conceptual model of BPS but focused on factors effecting Business Process Standardization and the effects of Business Process Standardization. While these antecedents and consequences of BPS are of interest, Romero et al. (2015a) give limited attention to carefully defining BPS. Their definition of BPS (as “the unification of business processes and the underlying actions within a company”, p 261) is arguably weak, with ‘unification’ not explained further. Commencing from an incomplete notion of what BPS is, limits the utility of their model.

A clear definition is an essential basis of any conceptualization effort (MacKenzie et al. 2011). Afflerbach et al. (2016) reviewed selected existing BPS definitions to understand how BPS has been conceptualized in the literature. The authors used this understanding to create a design objective for the economic decision model that they propose. While their work is of merit, the focus of that work is not on conceptualizing BPS. The authors themselves note a lack of consensus in BPS related literature, which further reinforces the motivation of this study.

Wurm et al. (2018) attempted to develop a measurement instrument of BPS. While the structured approach they followed has merit, we observe limitations. Any measurement instrument should start from a clear definition of the focal concept (MacKenzie et al. 2011), however, they give this light treatment. Also, their definition does not align with the measures of BPS presented later. Ultimately, the measures should map readily to the dimensions of the construct for which the measurement instrument is developed. The domains (or substrata) of measures in their measurement instrument (Wurm et al. 2018) do not capture the meaning of BPS, but align more to antecedents or consequences of BPS. For example, we consider culture and strategy (dimensions within Wurm et al.’s (2018) model) to be antecedents of BPS, rather than measures. Further, the authors use “process harmoni*” and “process Standardi*” as their search keywords. However, process harmonization can refer to several different activities not specifically associated with BPS² (Muenstermann and Weitzel 2008), which may conflate results.

Wurm and Mendling (2020) present a theoretical model for business process standardization, which aims at providing the contextual factors that influence BPS. Their overall objective was to identify and interrelate antecedents of successful BPS and connect them to different contextual elements. While the authors provide a good start to the theoretical model, their work again gives inadequate attention to conceptualizing the core construct, BPS. Fleig (2020) is another extensive and recent BPS study. Their focus is also on the antecedents of BPS. More specifically, they address how organizations need to consider contingency factors (such as strategy, business, and application systems) in a data-driven way when making decisions about how to standardise business processes. Once again, Fleig’s focus is not on the concept of BPS itself and includes a minimalistic conceptualization of BPS.

As evidenced above, prior attempts at conceptualising BPS are few and with limitations. This study addresses this lack by building on prior work with a rigorous study design which is detailed next.

This section is presented as a narrative review (Paré et al. 2015, p. 185), which “attempts to identify what has been written on a subject or topic …”. In line with typical narrative reviews (Paré et al. 2015), the aim here was not to seek cumulative knowledge from all published resources, but to develop a simple structured understanding of what has been stated about BPS, using the published literature as the evidence-base. This understanding of what is stated in the current literature is fundamental in any conceptualisation effort (MacKenzie et al. 2011); it was also deemed essential to understand and make sense of the findings of the subsequent stages.

This analysis pointed out that most of the literature inclined on explaining the conduct of BPS. Thus, this section provides a summary overview of how the conduct of ‘Business Process Standardization’ is described in current literature.

The analysis suggested BPS is a staged process, where various activities need to be accomplished for Business Process Standardization to be achieved. The literature analysis revealed seven stages, as detailed following. Knowledge gaps pertaining to each stage are briefly discussed in this section and are later suggested as important areas for future research. Note that earlier mentioned concepts that are related to BPS (introduced in Sect. 2.2) and play a part in the BPS process, are revisited here.

As discussed in Section 2.2, a master process can be referred to as a process reference model. Usually some process performance indicators (such as time, cost, and quality) are used to identify what is considered the best (or better) practice amongst several master process candidates (Muenstermann et al. 2010a). However, who decides on the derivation of the Master process and how this is done is rarely discussed in the current BPS literature.

An organization can also go through an internal Business Process Standardization effort which solely focuses on an internally derived master process; without attending to process improvements. Hence, we see Stage 5 as optional, in which case the master process (created in Stage 3 and enhanced in Stage 4) may (or may not) be subject to further improvements.

Zellner et al. (2015a) suggest a procedure to manage and unify existing variants with a standard process. They advocate documenting the processes to be unified in one table and documenting the variants along with the reasons for the variations in a separate table. Through interviews in an action research approach, they show how the tables help to identify, manage, and improve variants in a BPS context. However, their procedure derives from observation of a single process in a single organization. Overall, there is a dearth of literature in the BPS domain addressing how variants are best unified with a standard process.

While Fig. 2 depicts a sequential BPS process as evidenced in literature; we also argue that these stages can take place in a non-linear and iterative way. For instance, an organization may document a process, then modularize it, which may reveal things missing, which are further documented; or the modularization can be done as the processes are documented. Should the recognized body not approve the resultant master process, additional iterations may be required until the master process is established as the standard. Nor must these stages always occur in the depicted order. For instance, specificities may be isolated before or after the identification of the master process. Also, we consider Stage 5 to be optional (which we acknowledge runs counter to the views of some authors).

Importantly, although these stages represent a synthesis of the previous literature, most of the stages lack detail, revealing gaps which demand research attention to progress BPS knowledge (see Sect. 5.1).

An important step in conceptualizing any concept is to arrive at a clear definition. According to Wacker (2004, p. 631) “a ‘good’ definition is defined as: a concise, clear verbal expression of a unique concept that can be used for strict empirical testing”. Further, “definitions must be carefully designed to clearly represent the abstract concept” (Wacker 2004, p. 630). These conceptual definitions form the essential building blocks of a theory. MacKenzie et al. (2011) suggest identifying conceptual themes as a critical step to arriving at a conceptual definition. Conceptual themes are attributes or characteristics that are necessary and sufficient for something to be an exemplar of a construct. They advocate using past research to extract definitions and to uncover the characteristics associated with the construct of interest.

We later employ Wacker’s (2004) rules to assess various definitions of BPS. According to Wacker (2004), there are eight rules for a good formal conceptual definition: (i) it needs to be defined using terms that are clearly understood by the reader, (ii) it should be unique and try to exclude shared terms with other definitions, (iii) it should not contain unclear and ambiguous terms, (iv) it should be short (few terms), (v) it should be as similar as possible across studies, (vi) new definitions should not expand a concept and make it broader, (vii) it should not have any hypotheses (including outcomes), and (viii) content validity should be done (only) after there is a formal definition; measures should not be a part of the definition, as such suggests premature operationalization (before adequately defining the construct).

Following the approach suggested by MacKenzie et al. (2011), we discerned 22 explicit or implicit definitions from past literature on Business Process Standardization.

Each of the 22 distinct BPS definitions extracted was reviewed in depth (see Appendix B, Fig. B.1 for the summary results); firstly, by evaluating them against Wacker’s (2004) rules (mentioned above), and subsequently by extracting conceptual themes that characterize BPS. Thirteen (13) out of twenty-two (22) definitions, included a hypothesis (such as ‘process standardization results in efficient cost, time and quality’) thus infringing Wacker’s rule (vii). Several definitions were found to include measures, suggesting premature operationalization (before adequately defining the construct). Several use vague terms that require elaboration, without which the definition is unclear. No single definition has gained widespread acceptance (i.e., is cited by a majority of papers). These results further reaffirm the need for a satisfactory definition of BPS.

Our analysis confirmed that very little attention has to date been given to deriving a clear conceptual definition of BPS; and that there is no widely accepted formal definition of BPS. In the event of empirical measurement, some papers proceed without any formal definition of BPS (e.g. Wüllenweber et al. 2008; Chtioui 2009) raising serious concerns with the integrity of related results.

Table 1 presents a synthesis of the conceptual themes extracted from Fig. B.1 (derived from the bolded text from Column 1 of Fig. B.1). Column 1 of Table 1 lists the initial (Level 1) themes extracted from Fig. B.1. Column 2 presents supporting evidence, listing the related sources supporting the theme(s). The ‘Level 1’ themes were subjected to axial coding (following the guidelines of Wolfswinkel et al. (2013)), which resulted in six ‘Level 2’ themes (Column 3). In Column 4, the Level 2 themes are mapped against the stages of BPS (presented in Sect. 4.1, see Fig. 1) to show how current definitions use stages of BPS to define BPS. The six Level 2 themes were then further grouped yielding two main phases of BPS: (A) Derivation of an enhanced and approved master process (as the standard), and (B) Unification of process variants with the accepted master process (Level 3—Column 5).

The derived Level 2 themes closely align with the stages of Business Process Standardization (see Table 1—Column 4 and Fig. 2), instantiating all but Stages 2 and 4—‘modularization of documented processes’ and ‘isolation of specificities’ respectively. Given that modularization of documented processes and isolation of specificities are identified as key stages of the BPS process in the literature (as explained in Sect. 4.1), we argue that they should be accounted for in a complete definition of BPS. However, this is not the case within current BPS literature.

Additionally, Table 1 includes a Level 2 conceptual theme, ‘best-known execution process’. As discussed in Sect. 4.1, once the organization decides on a master process, the master process may be subject to improvement by integrating best practices. However, this is context dependent as an organization may only be interested in standardizing its process without intending to assess and integrate best practices as part of the BPS effort. According to MacKenzie et al. (2011), a definition should only include dimensions (conceptual themes) that are necessary at all times for the concept to be complete. While it is important to acknowledge conceptual themes that may be relevant only to some contexts of a phenomenon (not universally relevant), its best to not include these in the definition. However, acknowledging these contextually unique themes enables researchers to integrate them, as and when the definition is to be adapted in future study contexts.

Our conceptualization of BPS has two main phases namely (A) derivation of an enhanced and approved master process, and (B) unification of variants with the accepted master process; with the former excluding the stage ‘possible improvements’ to the master process (optimization), which is optional. Therefore, the first phase comprises 5 stages. Table 2 provides the definitions of the phases and stages. This conceptual overview of BPS has been derived by integrating two sources of evidence, namely: (i) analysis of the way BPS is conducted as described in literature in Sect. 4.1, and (ii) analysis of definitions, presented above (and in Appendix B).

The resultant BPS conceptualization aligns somewhat with Muenstermann’s definitions across his three papers (Muenstermann 2015; Muenstermann et al. 2010b; Muenstermann and Weitzel 2008). Muenstermann and Weitzel (2008) define process standardization as: “To standardize a process means to homogenize it against a standard process”. In the other two papers, standardization is defined as “the unification of variants of a given business process by aligning the variants against an archetype process” (Muenstermann et al. 2010b). We agree with the definition provided by Muenstermann and Weitzel (2008), but not the definitions in the other two papers, mainly due to ambiguous terminology used, as described below.

BPS is the activity of unifying process variants against an enhanced and approved master process (standard process), not an archetype process (we equate archetype process with ‘master process’ in this paper). Recall (see Sect. 4.1) that once a master process is identified it is subject to isolation of specificities, optional enhancements, and then has to be approved as a standard by a recognized body (or accepted as a standard through consensus, where the group consenting is the recognized body), without which it is not considered a ‘standard process’. This is why an archetype process is not automatically an enhanced and approved master process. Earlier on (see Sect. 4.1, Stage 3), we pointed out that the terms ‘archetype process’, ‘prototype process’ and ‘master process’ have been inconsistently used in the literature to explain much the same thing. In example, we observe in the two definitions by Muenstermann (2015); Muenstermann et al. (2010b) use of the term ‘archetype process’ inaccurately.

We therefore adapt and modify the definition proposed by Muenstermann and Weitzel (2008) to derive a definition of BPS on the basis of findings from literature. Unlike prior definitions, our definition captures the conceptual themes of BPS, is in accordance with Wacker’s (2004) rules of a good formal conceptual definition and has a stronger evidence base. We define BPS as follows:

Business Process Standardization (BPS) entails derivation of an enhanced and approved master process as a standard and unifying other process variants with that master process.

In accordance with Burton-Jones and Straub’s (2006) two-step method for conceptualizing constructs, once the focal construct (the BPS concept) ‘structure’ has been identified (i.e., see Table 2 and Sect. 4.2.1 above), the next step is to generate a set of items (henceforth herein referred to as measures) that represent and further refine the conceptual meaning of the construct (MacKenzie et al. 2011; Burton-Jones and Straub 2006). The goal of this section was to further progress the BPS conceptualization effort, by analysing how measures used in operationalizing BPS in prior research best represent the now defined conceptual domain of the construct. Ideally, a construct’s measures should fully encapsulate the essential aspects of the construct’s domain (and exclude concepts outside the domain); to foster research that accurately focuses on the construct of interest (MacKenzie et al. 2011; Burton-Jones and Straub 2006). Failure to identify proper measures may jeopardize progressive knowledge in the area of interest, as otherwise the concept(s) may not be correctly measured, resulting in misguided research outcomes.

In this phase of the study, we collated and analysed the existing (otherwise scattered) BPS measures from the literature with a multi-phased approach. Our analysis was conducted on the premise that ultimately, measures should fully capture the essence of the construct (MacKenzie et al. 2011), and thus could be used to both confirm or modify our understanding, and to inform the potential operationalization of BPS. The detailed evidence of this analysis is presented in Appendix C. Summary results are presented in Table 3 below.

All statements in the literature that directly or implicitly mentioned quantitative or qualitative BPS measures were captured as in-vivo codes under the ‘measures’ main node (see Column 1 of Fig. C.1, Appendix C) in the first round of coding. In subsequent rounds of coding, these were grouped into similar themes and hierarchically arranged when relevant (see Column 2 of Fig. C.1) forming measurement categories for BPS (see Columns 3 & 4 of Fig. C.1 and Columns 2 & 3 of Table 3). These measurement categories were then mapped to the BPS stages and core phases (as derived in Sect. 4.2). Table 3 provides a high-level overview of: suggested measurement categories (with descriptions—Columns 2 & 3) mapped against the different BPS stages (Column 1) as synthesized from the literature.

The analysis of the 60 papers within the paper pool, revealed that 26 out of 41 empirical papers reported measures (quantitative and/or qualitative in nature) used to operationalize BPS (the 15 other empirical papers did not measure BPS⁷). Our review evidenced much variation in measures across the studies. While measures can legitimately vary due to contextual differences in which BPS is studied, the diversity observed appears ad-hoc; similar to what Burton-Jones and Straub (2006) describe as ‘un-disciplined diversity’—which makes the research results incomparable and difficult to (re)use, thus impeding cumulative knowledge creation. This further confirms the value from having a clearly defined starting set of measures to operationalize BPS.

The analysis of measures revealed several interesting observations. First, while the initial literature analysis (see Sect. 4.2) revealed two main BPS phases, with the first phase having five stages (see Fig. 2), none of the studies that ‘measured’ BPS (25 papers) had measures that mapped against all the stages. This suggests that prior authors may not have considered all stages of BPS and/or their measures were overarching in nature (spanning multiple phases). Construct operationalization guidelines (such as Burton-Jones and Straub 2006; MacKenzie et al. 2011) describe the importance of being clear on what aspects are being operationalized and at what level of granularity. This has not been addressed by the BPS operationalization efforts reported to date.

While no single study’s BPS measures mapped against all stages, all stages were instantiated by the analysis of the pool of papers. This reinforces the pertinence of the seven stages (inclusive of one optional stage) of which BPS is comprised. The analysis also revealed measures that aligned with ‘possible improvements to master process’. Recall, ‘possible improvements to master process’ is an optional activity in the conduct of BPS, and hence was not considered to be a core dimension of BPS. While we consider the incorporation of ‘possible improvements to master process’ as optional in BPS, we acknowledge that its relevance will depend on the unique context in which BPS occurs. Measures presented in this paper for ‘possible improvements to master process’ (see incorporation of best-known execution options/possible improvements to master process under ‘Other Measures’) may be adapted by researchers or practitioners who wish to assess BPS for an organization which chooses to integrate best practices as part of the BPS effort.

Several measures were overarching in nature (see second row under ‘Other Measures’ in Table 3). For example, ‘highly standardized processes’, aims to gauge whether the process is perceived to be standardized or not, without addressing the separate stages. Such overarching measures provide only a broad sense of the construct being measured (Gable et al. 2008). And though they can be valuable in certain construct validation tests (e.g., as criterion measures used in combination with dimension-specific measures), when used alone Burton-Jones and Straub (2006) refer to these as ‘lean’ measures; which are considered convenient to use, but have been critiqued for being inexact and not capturing rich data related to the attributes being measured.

Several of the ‘other measures’ were not directly about BPS. We positioned these measures as ‘implied’ measures; measures which can be used as a proxy to understand BPS.

Imbalance was noted in the range and detail of measures observed across stages. For example, very few measures were found in relation to the ‘unification of variants with the accepted master process’ phase. One could argue that what was found here was too broad (see Phase 2 measures in Table 3); hence inviting ambiguity. Other elements such as applied modelling language, definitions of process components, and interdependencies of process activities need to be considered before merging modelled process variants, none of which are represented in the current measures. Consideration of such aspects will enable more detailed measurement, resulting in improved validity of the phase being measured (Burton-Jones and Straub 2006).

Several items mapped to the BPS stages are inadequately granular (too high-level—observed via Appendix C). For example, items for—‘modularity of business processes’, would benefit from more granular measures, to better capture the meaning of that stage. Lack of sufficient detail is also evident for ‘agreement of enhanced master process’, as it has only two high-level measures (‘maintaining governing documentation’ and ‘well-regulated process’), which could be usefully further decomposed, perhaps dependent on context.

Overall, this analysis provides a solid base for future researchers as it consolidates all prior measures of BPS; an essential task when conceptualizing a construct (Burton-Jones and Straub 2006). This presents a solid foundation to derive a formative index of BPS for future research. The consolidated list of BPS measures we arrived at, and the analysis presented, is more complete (and transparent) than prior such efforts.

This study on business process standardization (BPS) conceptualization is a response to continuing calls for a clear(er) conceptualization of the construct. Strong conceptualization is an essential prerequisite to building theory (Weber 2012) and the progression of a field (Corley and Gioia 2011). Our work coincides with Gregor’s (2006) ‘theory of analyzing’ and provides an in-depth explanation of what BPS entails. The primary contributions of this work are the (i) newly formed BPS definition, (ii) empirically derived ‘structure’ of the BPS concept, and (iii) collation of BPS measures—which provides valuable theoretical foundations to progress ‘disciplined’ (as explained by Burton-Jones and Straub (2006)) research in BPS.

The study comprehensively reviews prior BPS conceptualizations and definitions, identifying limitations, thereby better attuning future researchers in the area to the relative strengths and weaknesses of these past studies and of their prescriptions. With the increased insight this study achieves, prior work can be better and more appropriately leveraged.

We inventoried past related work from which we inductively synthesized the salient stages. Then we deductively applied that stage-model as a lens on past definitions of BPS, evaluating those definitions’ foci and completeness, ultimately arriving at the ‘structure’ (Burton-Jones and Straub 2006) of BPS which informed our definition and guides any adaptations (i.e., for new definitions or BPS measurement) in future research. In addition to a more reliable definition of BPS, the work also recognises potential conflation (Sect. 2.3) with related but not fully analogous concepts and guides the reader on how to position these other notions within the context of BPS. Further, we substantiate our conceptualization by extracting and analyzing measures of BPS from literature that addresses BPS operationalization efforts, thus yielding a solid base for future BPS operationalization efforts. We recognize that the collated measures reflect extant work in the area, and that should one alternatively take a zero-based view (i.e., start from the derived dimensions, ignoring existing measures), the measures may be different. Nevertheless, this paper provides a solid base for future researchers as it consolidates all prior measures of BPS; an essential task when conceptualizing a construct (Burton-Jones and Straub 2006).

Methodologically, this study goes beyond prior attempts to conceptualise BPS. It applies a systematic literature review approach to accounts for and build on prior BPS studies and does so with increased transparency as regards the evidence-base search criteria. The study is highly specific as regards the process of analysis and synthesis of evidence, through to interpretation and outputs. Such detailed, and open description of the study approach increases the likelihood of its widespread adoption in future research and facilitates the study’s extension in further work, while concomitantly increasing confidence in the current study results.

The study results are also of value to practice. With growing intent to standardize processes, clarification of BPS is important for business process practice as well. The stages presented enable an improved understanding of the process of BPS, providing guidance on the conduct of BPS and enabling organizations to plan accordingly. The definition of BPS provides a good starting point for practitioners to understand the concept of BPS. Finally, the measures assist practitioners to comprehend how they can evaluate the extent of their BPS efforts. The findings of the paper can act as input in planning, implementation, and monitoring of BPS efforts and hence support the large-scale investments involved in business process standardization efforts.

In this section, we acknowledge some limitations and identify opportunities for future research. As all research, this study also had inherent limitations triggered by the study approach. A literature-based synthesis and theorizing is influenced by the set scope and applied analysis approaches. While a rigorous method was designed and followed through, selection bias (e.g., of papers selected), coder bias (e.g., on what was coded and how things were interpreted) etc., are hard to completely avoid in a qualitative study as this.

One key limitation of the work is our inattention to neighbouring domains—e.g., business process harmonization. It is acknowledged that this study relied heavily on extant research in BPS. Yet, other disciplines like the service research community have interest in standardization and modularization (e.g., Voss and Hsuan 2009; Beverungen et al. 2018). Given overlap between these communities, with a process often considered a central element in service thinking, there is symbiotic value possible from extending the work herein to these areas of expertise, both drawing from and informing their thinking. As specific example, we found that several of the stages had coarse measures (as per the BPS literature reviewed), such as modularization and isolation of specificities. While looking at proxy domains was outside the scope of this study, we urge future researchers to refer to research related to proxy domains for analogous supporting measures of these stages.

While we believe the core intent of this work has been achieved; resulting in a more rigorous conceptualisation of BPS, we also acknowledge the diverse opportunities for further research. Though a comprehensive research agenda of such potential work is outside scope, we refer the reader to the various gaps and opportunities, mentioned throughout the paper, which we observed across the structured literature review activity. The gaps and opportunities were derived following Alvesson and Sandberg (2013, Ch. 3) and Müller-Bloch and Kranz (2015) techniques for gap/opportunity spotting.

Firstly, we call for future work that specifically focuses on the measurement of BPS. The study provides a customisable set of options that future research and practice can adapt and develop further. For example, BPS can be operationalised as a process (i.e., to measure how well a BPS effort is progressing) or as a formative index (i.e., to economically evaluate the state or degree of BPS, after a BPS effort) (Ortiz de Guinea and Webster 2014).

From a process perspective, the stages and measures of BPS presented in the paper help to understand what BPS entails and enable development of procedural guidelines for BPS. Given that the key stages defining a construct will depend on the purpose and context of each study (MacKenzie et al. 2011), we recognize and encourage the stages presented here be adapted to fit the specific purpose and context of future BPS research. Future empirical work to further test, validate, and re-specify the BPS stages presented in this paper is warranted to further establish their validity and applicability. This can be done by action design research case studies where the proposed stage-model is applied as a reference method to standardize a business process. It would be interesting to see if/how the applicability of stages (and their order) differs across different types of organizations and process standardisation purposes.

We further call for future research to further develop the measures and build the BPS conceptual model with more structured, relevant, and complete operationalizations. With this paper presenting the key dimensions of BPS, each dimension can be investigated in detail to extract specific and customisable measures and means to operationalise them (e.g., specific measurement items with different modes and forms of evidence). Furthermore, one can use the measures proposed in this paper to understand the extent of BPS at an organization, where new measures may emerge.

Secondly, we recognise that BPS occurs in diverse forms and that a clear understanding of these different types/forms of BPS can be helpful to better conceptualise, operationalise, and manage BPS. While some early related work exists (Goel et al. 2018, 2021), it is yet in its genesis and more extensive research here can usefully contribute to the field of BPS. For example, this will provide a deeper understanding on; how different forms of BPS are influenced by diverse contextual factors (across design, implementation, and maintenance decisions), and the impact of different types of standardizations on different organizational context.

Specific stages/activities within the BPS process needs expanded design knowledge. For example, while BPS literature mentions the significance of modularization in standardizing a business process, how it can be done remains unaddressed. Similarly, there is limited literature around the steps involved in the derivation of Master process. The BPS field will benefit from an understanding of how a process can be modularized and how a master process can be derived. Additionally, the paper we raise questions about the completeness of the phase; ‘unification of variants with the accepted master process’. We call for future research to understand the process of unification in more depth. Other work e.g., Romero et al. (2015b), has argued the need to understand in greater detail the process of unifying variants, but that work is yet in its genesis. Methods used herein for BPS more broadly, can be adapted to conceptualise process unification more specifically, and to derive measures of process unification. Furthermore, the significance of isolation of specificities is well understood, however, the kind of specificities that need to be taken into consideration remains inconspicuous. There is a need to understand the specificities that need to be considered for BPS, enabling smooth operationalization of BPS stages and also how best they can be managed. Moreover, there is little discussion about the governance structures regarding process standards in literature. The BPS field will benefit from an understanding of the key activities involved in the governance of process standards, the responsibilities, amongst others would benefit from research in sister domains.