The polyphony of deviance: the impact of deviant workplace behavior on digital transformation

Abstract

Purpose

The purpose of this study is to explore the impact of deviant workplace behavior on digital transformation in the public sector. This contributes to the current literature on public sector digital transformation as well as to that of deviant workplace behavior in public sector contexts.

Design/methodology/approach

The authors conduct a qualitative case study of a digital transformation initiative in a Swedish municipality.

Findings

The authors identify three types of institutional drift related to digital transformation, i.e. decelerating digital transformation, maintaining infrastructural stability and accelerating digital transformation. The authors categorize mediators for said drift and theorize on the role of deviant workplace behavior as a strategic driver for digital transformation in public sector organizations.

Research limitations/implications

With the study being a qualitative case study, it is limited in terms of generalizability and transferability. Through this study, the authors sensitize the notion of digital transformation and show how deviant behavior results in strategic polyphony. Future studies are informed through offering a new perspective to public sector digital transformation strategy.

Practical implications

Practice should view deviant workplace behavior as simultaneously constructive and destructive in lieu of planned digital transformation, as well as see its presence as a potential sign of subpar prerequisites for digital transformation in the public sector.

Social implications

Through this study, deviant workplace behavior is highlighted as a source of strategic polyphony and hence an important aspect of public sector digital transformation strategy.

Originality/value

Through being the first paper, to the best of the authors’ knowledge, to apply the theory of institutional drift to digital transformation settings as well as identifying the impact of deviant workplace behavior on digital transformation, the study offers novel insights.

Citation

Magnusson, J., Carlsson, F., Matteby, M., Ndanu Kisembo, P. and Brazauskaite, D. (2024), "The polyphony of deviance: the impact of deviant workplace behavior on digital transformation", Transforming Government: People, Process and Policy, Vol. ahead-of-print No. ahead-of-print. https://doi.org/10.1108/TG-09-2023-0144

Introduction

Since the dawn of the formal organization, organizational change has been cumbersome and riddled with caveats. Through studies of inertia and organizational resistance (Malhotra et al., 2021), we have learned that organizations are poised in opposition to change. Change is a non-natural state for the organization, and organizational actors will engage in direct conflict with change initiatives (Tangi et al., 2021; Val and Fuentes, 2003). Or to put it in the words of Newton’s third law of motion: for any action, there is an equal but opposite reaction.

Previous research has addressed these issues of active resistance through a plethora of constructs such as guerilla warfare, cynicism, inertia and rigidness (Sarkar and Osiyevskyy, 2018). Far from conclusive, the findings indicate that resistance impacts everything from the outcome of specific initiatives (Frick et al., 2021) to the institutional order (Voronov et al., 2021) and strategic direction of the organization (Baptista et al., 2021). Resistance is not solely acting against the change initiative resulting in standstills but prodding the organization into new directions. Resistance is hence a generative process (Dillard‐Wright, 2022).

Digital transformation is a specific type of organizational change instigated through the utilization of digital technologies (Hanelt et al., 2021). Previous research highlights that the transformation involves both incremental improvements in existing operations and radical changes in the underlying business model (Vial, 2019). With transformation being both incremental and radical, resistance is often abounded and will influence the direction of digital transformation (Chanias et al., 2019). There have been multiple studies of how resistance impacts digital transformation (Frick et al., 2021), yet so far there have been but few studies of how practice deviations on the micro level, over time, create drift in the strategic direction of an organization.

Voronov et al. (2021) proposed a theory for how practice deviations result in institutional drift, i.e. an incremental reshaping of the institutional order of an organization. Previous studies have mostly refrained from addressing the impact of digital transformation on the institutional order in the public sector, yet as argued by Orlikowski and Scott (2023) digital transformation invariably results in institutional displacement, i.e. significant changes in the institutional order. They describe this as “the undertow” of digital transformation. We posit that the theory of institutional drift warrants testing in relation to digital transformation in the public sector to increase our understanding of changes in institutional order. To this end, we equate practice deviations with the previously studied construct of deviant workplace behavior (Appelbaum et al., 2007), henceforth referred to as DWB. Our research is guided by the following research question:

RQ1.

What is the impact of deviant workplace behavior on digital transformation?

The question is answered through a qualitative case study of a large Swedish municipality engaged in a significant program of digital transformation. We contribute to research through answering the calls from Voronov et al. (2021) on empirical tests of the theory of institutional drift. Our study finds that DWB and institutional drift help explain and predict the emergent nature of digital transformation, supporting the notion introduced by Chanias et al. (2019) and offering potential new avenues for research. We further find that DWB trifurcates digital transformation in the case organization, introducing strategic polyphony into public sector digital transformation.

Previous research and theoretical framing

Digital transformation, here understood along the lines of Hanelt et al. (2021) as organizational change brought about through the utilization of digital technologies, has received ample scholarly and practical attention during the past couple of years. Through a multitude of studies, research has addressed issues ranging from its micro- to macrofoundations (Chanias et al., 2019; Wilson and Mergel, 2022). Core to previous findings lies the perception of digital transformation being a composite of parallel actions to exploit existing opportunities (e.g. decreased cost of continued operations and increased efficiency) and explore new opportunities (e.g. develop new value offerings, increase innovation). At the same time, digital transformation also challenges underlying institutional order (Hinings et al., 2018), resulting in what Orlikowski and Scott (2023) referred to as institutional displacement or the “undertow” of digital transformation.

Due to its oftentimes disruptive character, digital transformation is reportedly met with resistance from managers and coworkers (Syed et al., 2023). In a conceptual study by Voronov et al. (2021), the impact of resistance in the form of practice deviations is proposed to over time change the institutional order of the organization in question. Through coworkers not sanctioning the norms, the organization experiences what Voronov et al. refer to as “institutional drift.”

The notion of organizational actors not following rules, regulations and norms has been previously studied within a variety of perspectives. As noted by Ciborra et al. (2000), organizations are poised in the tension between control and drift, destined to move in directions that are sometimes intentional and other times unintentional. Baptista et al. (2021) studied the implementation of a digital check-in solution at a UK airport and found that deviations in terms of use lead to strategic drift for the organization as such, Rahrovani (2020) studied the strategic impacts of social media platforms and Nielsen et al. (2022) studied how translation of constructs over organizational sets creates conceptual drift. While there have been several contributions to the study of unintentional change in the previous literature on drift, most of the previous studies have so far not been applied in the study of public sector digital transformation.

Following Voronov et al. (2021), this study builds on a theoretical foundation of institutional work, calling for a more nuanced understanding of the relationship between agency and structure, as well as seeing institutional processes as emergent, nondeterministic and nonlinear. This strand of institutional theory has a long tradition yet has only more recently opened for an increased emphasis on the role of the individual in institutions and on the role of unintended consequences in the shaping of social reality.

As organizational actors engage in DWB, they invariably change the organization. Robinson and Bennett (1995) proposed a typology for DWB, differentiating between the organizational vs inter-personal level and minor vs serious consequences of the behavior. As seen, the DWB categorized by Robinson and Bennett as mainly negative with little or no potential positive consequences for the organization as such. Warren (2003) acknowledges that certain DWB may have a positive impact on the organization and develops an integrative typology with both constructive and destructive behavior. In terms of the constructive, this behavior may be in the form of refusal to comply with dysfunctional configurations and as proposed by Spreitzer and Soneshein (2004) behavior that is associated with honorable intent. Magnusson et al.’s (2020) study of how the coworkers at the Swedish Tax Authority engaged in unsanctioned digital innovation activities (i.e. shadow innovation) to save the long-term relevance of the organization, offers an example of said constructive DWB.

To increase our understanding of how DWB impacts digital transformation, we use the theory of institutional drift as proposed by Voronov et al. (2021). According to this theory, practice deviations (i.e. DWB) may be either unnoticed, ignored or deemed threatening by coworkers. If the practice deviation is unnoticed, it has no effect on the institutional order, but if it is either ignored or deemed threatening (i.e. in both cases acknowledged) it will have an effect and result in institutional drift or even institutional doubt. This theory will be used as the basis for answering the research question.

As argued by Kuipers et al. (2014) in their literature review of the management of change in public sector organizations, there is a dearth of detail and depth in empirical studies, which requires additional bridging of theories. Here, we argue that the theory of institutional drift offers a potential novel approach to studying change in public organizations, both in terms of the apparent changes in organizing and operations (Weerakkody et al., 2016) and in the institutional displacement inferred through digital transformation (Orlikowski and Scott, 2023). Given the increased prevalence of digital transformation in public sectors, recently accelerated through the COVID-19 pandemic (Agostino et al., 2021), we expect to see increased studies of change through the empirical phenomenon of digital transformation in the future.

Method

This study follows a qualitative, single case-based approach. The case was selected on two main criteria. First, we wanted to find an incumbent organization with a predigital heritage (Chanias et al., 2019). The rationale for this was that we wanted to study a case where the digital logic could be seen as clashing with something that existed previously, because this was believed to be a good condition for studying DWB in relation to digital transformation. Second, we wanted to find an organization with a substantial and explicit digital transformation program. The rationale for this was that the explicit program would increase the likelihood of the organizational actors being aware of the digital transformation happening (Wilson and Mergel, 2022).

Using these two criteria, we identified the case of Sundsvall municipality. Sundsvall is a large municipality in Sweden, with 6 000 employees serving some 100,000 citizens. In 2019, the organization initiated a substantial digital transformation program as a response to a forecasted financial crisis. As the forecast read, the municipality would not be able to continue functioning without significant changes to its operations. With decreasing tax revenues, increasing demands for public services and decreasing access to competence, the municipality identified digital transformation as of the highest priority to safeguard continued operations. Digital transformation is initiated by the politicians, and executed through a digital transformation department working in close collaboration with the service center for information technology (IT). The municipality sees digital transformation as a method for business development, and as such they have moved away from a supply-demand model where digital initiatives were internally procured from the supporting functions into a setup where the digital transformation experts support local business development following a DevOps setup (Gall and Pigni, 2021).

To sensitize the researchers to the case organization, we used secondary data from a larger, programmatic research initiative where the researchers had been engaged in clinical inquiry designed to support the digital transformation of the municipality. The data set comprised of 65 transcribed interviews conducted with stakeholders in the municipality. These stakeholders did to a large extent overlap the respondents of the primary data collection for this study, yet also included other individuals from, e.g. politics and other administrations as well as municipality-owned firms. This data set informed the primary data collection but was in no way part of the analysis of findings.

The data used for the analysis was collected through a series of 15 semistructured interviews conducted during the spring of 2021. The interviewees were selected following three criteria from the literature on DWB: hierarchal position (Moon, 2021), proximity with digital transformation (Christ-Brendemühl and Schaarschmidt, 2019) and tenure (Appelbaum et al., 2007). Out of the 15 respondents, three were upper-level managers, four mid-level managers, three team-leaders and five public servants. The interviews were conducted through Zoom, recorded and transcribed. The transcribed interviews were analyzed using NVivo 2020 in two phases. First, we coded individual accounts of DWB, inductively creating 12-s-order categories (Table 1). Second, we analyzed the mediating factors deductively from the previous literature to act as a basis for inductive categorization of ten categories (Table 2). This was followed by interpretation of the findings through the theory of institutional drift, where we first analyzed the patterns of DWB inductively to identify the different types of institutional drift. This was followed by a deductive approach where we used the elements of institutional drift (Figure 1) to trace both how the three identified types of institutional drift unfolded and to identify the mediating factors present in each type.

Results

The results are presented through first describing the patterns of DWB identified in the case study. This is followed by an exploration of the mediating factors associated with DWB. Finally, we present the three types of institutional drift identified in the case study to answer our research question. We have refrained from subcategorization of respondents based on hierarchical level (e.g. manager and public servant) on account of the analysis not identifying clear patterns in relation to differences between the levels. Hence, the analysis is conducted on the aggregate level of respondents, referring to all respondents as “coworkers.”

Deviant behavior in digital transformation

The most common DWB category found in the case is that of workarounds, where coworkers act in noncompliance with the existing governance to get things done. In all the observations, this category was related to wanting to either pursue something that was not possible, or to increase the pace and avoid what they regarded as inertia in the existing routines, i.e. the behavior was deemed constructive.

As for how workarounds were operationalized, the predominant method found in the data was that of using personal networks to bypass the formal governance. Instead of following the established routines, they would simply ask or bargain a favor from somebody sitting on the resources needed. Shadow IT was found to be the third most common category of DWB. Instead of following the governance in place and go through the formalized process for acquiring new IT, coworkers would simply handle the procurement through their own operating budgets (i.e. not the centralized IT budget), pay for the software out-of-pocket or simply use freeware solutions.

Not asking for approval was also found to be a category of DWB. Here, the coworkers worked under the impression that approval should really be asked for, but because this would slow down the pace of digital transformation, they simply disregarded it and circumvented formal decision gates. During and directly adjacent to meetings displayed, a separate category of DWB was categorized as destructive. Here, coworkers would either express negative comments about the competence of other coworkers or parts of the organization or simply engage in soldiering and being overly passive during the meetings.

Another destructive category of DWB was found in the continuation of the use of legacy systems over newly implemented solutions. Despite new systems and solutions having been implemented, some users would actively engage in counteraction, avoiding onboarding regardless of the organization having clearly communicated policies related to which solutions should be used for what. The municipality had a hard time decommissioning systems on account of factors such as their having data that needed to be saved for 10 years while the system was left active. We also found examples of extra-role or out-of-role behavior, where coworkers could be seen as misappropriating and spending resources on going the extra mile for both citizens and other coworkers that were not perceived as being part of their regular jobs.

Closely related to this category of DWB is the misaligned prioritization of tasks or projects. Here, the coworker would actively reprioritize, in direct conflict with the set priorities and with the expressed strategy of the organization, to attain objectives they perceive as taking precedence. In relation to potential breaches of laws related to staffing, we identified a separate category of DWB in the circumvention of the hiring process. With the current governmental regulation on recruitment in place, hiring is seen as a cumbersome process with significant elements of red tape where the rationale of said regulation is not always visible. To circumvent this, individuals would, e.g. engage in DWB by extending existing contracts rather than initiating a new hiring process.

Circumvention was also present in relation to the project management methodology used in the organization. Here, coworkers are obliged to halt projects until the formal requirements for passing a certain decision gate have been met. Instead of following these regulations, we saw examples of coworkers circumventing the process and flagging projects as “done” without having reached the formal requirements. As for the least prevalent categories of DWB identified, these were related to voicing concerns/explicitly questioning legitimacy and running shadow digital initiatives.

Mediating factors of deviant workplace behavior in digital transformation

In addition to identifying the types of DWB, we also studied the mediating factors involved for said behavior (Table 2).

As seen in Table 2, the most common category of mediator of DWB was Organizational. If the organizational structure is deemed counterproductive to the espoused direction that the coworker feels is legitimate, this will increase the likelihood of DWB. This may include both aspects of simply poor design (i.e. processes that are not purposive in any shape or form) or aspects of bureaucratization (i.e. regulatorily or in other ways normative driven design patterns such as red-tape). Other types of organizational mediators identified in the study include lack of resources and role ambiguity, both the direct consequence of the design and enactment of governance in the organization.

The second most common mediator category was Individual. Here, we see both tenure and knowledge and skills positively associated with DWB. In other words, the more competent the composition of coworkers are, the higher the likelihood of DWB. In addition to this, the personality traits and beliefs of the individual coworker is also found to be a mediating factor, with connotations of ideas about justice, responsibility and honor.

The third most common category of mediator was Technological. Current limitations in the installed base in the organization (technological limitations) are found to be a mediating factor. If the system does not support what is perceived as “best practice,” users will be prone to circumvent use through the introduction of hybrid routines. In addition to this, the fast-paced development and digital transformation (DT) process is also found to be a mediating factor for DWB, in the form of both destructive (e.g. resisting the fast-paced development) and constructive (e.g. circumventing legislation) behavior.

The least common category of mediator was Social. Factors include the organizational culture and climate, as well as the social pressure to conform to group. If the organizational culture is one heavily steeped in continuous improvements and minor (or no) changes over time, i.e. one of stability, this will increase the probability of DWB targeted at increasing the pace of change. If the social pressure to conform to group is high (i.e. low acceptance of behavior variations), then this will increase the likelihood of DWB.

Three types of institutional drift in digital transformation

Through our study and the identification of categories and mediators of DWB, we identify three types of institutional drift (Figure 2). The names of the categories of DWB have in some cases been shortened and thereby changed to afford illustration.

In terms of the institutional drift of decelerating digital transformation (Figure 3), the primary rationale for engaging in this is associated with a deviant stance to the overarching idea of digital transformation for the business unit that is subjected to the initiative. The actors account to a feeling of loss of control, where agendas that they have not bought into and accepted are being pushed, as they see it, counter to the rationale of the organization and their own roles. We saw a prevalence of leaning on notions of red tape and bureaucracy as intimately necessary for the safeguarding of the viability of the organization. The very idea and norm of stability becomes a safeguarding of the existing institutional order. The organizational response to the DWBs identified are noticed but primarily ignored by the digital transformation team under the rationale of certain business units then not becoming prioritized. This is rational behavior given that the resources available to the digital transformation team are outweighed by both the objective and the existing demand from other business units, whereby it merely results in a reprioritization and acceptance of certain business units not engaging in digital transformation.

The institutional drift of decelerating digital transformation is mediated by all categories of mediating factors except Technology (Figure 2). In the Organizational category, both organizational structure and a lack of resources mediate the drift. In the Individual category, both personality traits and beliefs and knowledge and skills mediate the drift. In the Social category, organizational culture and climate, as well as social pressures to conform to group mediate the drift.

As for the institutional drift toward maintaining infrastructural stability (Figure 4), this is primarily seen within the IT department of the organization. Here, the actors engage in DWB to safeguard the structural integrity of the existing legacy environment. Instead of pushing for modernization and development activities with higher levels of risk than pure maintenance, coworkers reprioritize and focus their resources on making sure that the stability of the infrastructure is intact. Here, we see the respondents frequently returning to notions such as security and quality, as well as service and stability. As the organization needs to have a stable supply of secure, (high) quality services, the DWB is poised to make sure that this is never compromised. This DWB goes against the intended objective of digital transformation, i.e. in essence changing the logic of the municipality, and instead reverts the attention to quality management in the form of incremental and continuous improvements. DWB is largely noticed by the IT managers yet ignored on account of the stability of the existing infrastructure being the core objective of the IT department after development was re-sourced to the digital transformation department in 2021. Despite digital transformation being a prioritized initiative, it is not yet perceived by all to be part of the institutional ethos, whereby the DWB is not normalized but instead leads to institutional drift without institutional doubt.

The institutional drift of maintaining infrastructural stability is mediated by all categories of mediating factors except Social. In the Organizational and Technological categories, all factors are found to mediate the drift. In the Individual category, only tenure is found to mediate the drift.

In terms of the institutional drift of accelerating digital transformation (Figure 5), the primary rationale for the identified DWB is that of a pending doom of the existing order. Individuals subscribing to both the transformative power of digital and the existential threat to the municipality and the public sector (decreased supply, increased demand and fluctuating demands) latch on to the ethos that digital transformation is not only necessary but time critical. Here we see the respondents using terms such as courage, innovation, risk-taking and openness to justify their DWB. The behavior is by and large noticed and normalized, with the digital ethos to some extent threatening the traditional institutional ethos of the municipality, casting significant doubt as well as leading to changes in the institutional arrangements. This behavior was most prominent among the digital transformation team members as well as local champions for digital transformation in the business.

The institutional drift toward accelerating digital transformation is mediated by all categories of mediating factors except the Technological. In the Organizational category, organizational structure and a lack of resources are found to mediate the drift. In the Individual category, personality traits and beliefs, as well as tenure, are found to mediate the drift. In the Social category, only social pressures to conform to group is found to mediate the drift.

Discussion

DWB has been found to impact digital transformation in three directions in parallel. For instance, in terms of the identified institutional drift toward decelerating digital transformation, DWB in the form of shadow IT and projects, workarounds, sticking with legacy systems and soldiering during meetings all contribute to counteracting the agreed-upon direction and pace of digital transformation in the organization. With this conflicting with the espoused strategic direction of the organization, this aggregate DWB could be categorized as destructive (Warren, 2003). However, in line with Voronov et al. (2021), these practice deviations are not so much counteracting as changing the direction of the digital transformation per se.

Previous research on strategy has been criticized for overly emphasizing intentionality and instrumentality in the conception of strategy practices. Findings such as those of Chanias et al. (2019) and Magnusson et al. (2020) highlight the emergent nature of digital transformation strategies, but so far only a few studies have addressed the unintentional drift experienced through DWB. As found in our study, DWB results in three instances of institutional drift, whereby the strategic direction of digital transformation is trifurcated. In parallel with the original strategy, we find evidence of what we argue to be new strategic directions for the organization (accelerating, decelerating and maintaining).

In other words, DWB is not merely resistance (i.e. etymologically holding back), but directional in essence. These behaviors skew the overarching strategic direction of the organization. Because the behaviors happen in parallel, we can see drift as introducing polyphony into the strategy of the organization, where different parts of the organization will strive in different directions. Previous research has identified this idea about multiple strategies coexisting in a single organization, much like the literature on organizational identity highlights the concurrent existence of plural identities in ideographic organizations (Albert and Whetten, 1985) and organizational hybridization (Buffat, 2014).

As noted by Warren (2003), DWB comes in the form of both constructive and destructive behavior. Coworkers engage in DWB when they experience doubt regarding the current direction of the strategy. They choose to take responsibility rather than merely follow, often at considerable risk of personal expense. In other words, the DWB becomes a warning signal of doubt and subsequent polyphony. In this case, the respondents were clear in acknowledging that their behavior was noncompliant and even counterproductive, but that they saw their decision to act as virtuous and even natural.

Here, we see the identification of the mediating factors as a specific contribution to our study. As found, the existing organizational design (e.g. governance and settings for management) triggers constructive DWB. In other words, the existence of constructive DWB is not a bug but rather a feature of the organizational design. As such, the existence of DWB that conflicts with the aspired strategy becomes more of a sign of subpar organizational design than a question of individual behavioral variation. This supports previous findings from the governance and control literature on the inability for change and responsiveness in governance (Cram et al., 2016; Visser, 2023).

In relation to the existing plethora of research on digital government, we make two main contributions. First, we contribute by adding to the diversity of approaches for studying the emergent process of digital transformation in government by leaning on the theoretical underpinning of institutional work. Previous contributions in digital government inspired by this tradition (Weerakkody et al., 2016) have started what we believe to be a fruitful avenue of continued research. Through the direct study of individuals and their role in digital transformation (Wilson and Mergel, 2022) in general and in institutional displacement in particular (Orlikowski and Scott, 2023) in public sector organizations, we believe that future findings will offer valuable insight, particularly in settings signified by high levels of institutionalization and bureaucratization (Sordi et al., 2021).

Second, we contribute by introducing the perspective of DWB to understand the emergence of digital transformation strategies in public organizations. This perspective has so far been underresearched in the context of digital government, and DWB has primarily been studied on the citizen side (Tang et al., 2019) and from the perspective of DWB as destructive (Piazza et al., 2022). Our findings illustrate the innate complexity of DWB as a driver of drift, i.e. a generative force in the evolution of government per se. We believe that this avenue of research should be followed more intently to further increase our understanding of the public servant as a factual co-creator of strategy.

Third, with this study being the first (in our understanding) to apply and test the theory of institutional drift (Voronov et al., 2021), we believe that this offers a contribution to research. How and why the institutional arrangements of public sector organizations change over time is an important area of inquiry, and one that so far has seen only limited research attention in the form of longitudinal studies related to digital transformation (Scupola and Mergel, 2022). If digital transformation changes the institutional order of the public sector, answering the question of how this change occurs becomes an important issue (Weerakkody et al., 2016). Here, we believe future research should consider using the theory of institutional drift to further our insight into the more processual perspectives on the impact of digital transformation over time.

In addition to the contributions to research, this study offers two main contributions to practice. First, the identification of DWB as not merely counteracting intent but also accelerating its execution should be seen as relevant for practice. As noted by Sordi et al. (2021), overcompliance constitutes a pathology in public sector organizations, whereby the DWB can be seen as purposeful reactions from coworkers. Through DWB, the organization’s operations are upheld despite the fact that governance and control is designed to counteract the enactment of the intended strategy. This positive perspective on DWB should be further fostered in organizations while simultaneously balancing the tendencies for less-than-purposive institutional drift. Second, the presence of DWBs as identified in our study may be interpreted as an indication of subpar prerequisites for digital transformation in organizations. Through mapping DWB, managers will be able to design better prerequisites for the efficient enactment of strategic choices. We recommend that managers in public sector organizations search for instances of deviance as one source of input for redesigning their organizations.

There are two main limitations to our study. First and foremost, our use of a single, nonlongitudinal case study to analyze the impact of DWB on digital transformation could be criticized for being too small a sample to adequately answer the research question in a manner that would afford generalizability. In line with Eisenhardt (1989), we argue that, albeit a small sample for statistical generalizability, our case offers a sound basis for theoretical generalizability. Second, we acknowledge the perils of public sector research transferability, as noted by Bannister (2007). DWB is, as argued in the previous literature, dependent on a range of factors, some of which may be linked to the institutional environment of the organization. We have made no attempts at controlling for the institutional environment in our study, whereby the question of the transferability of findings remains a shortcoming. We would rather expect to see that the manner in which DWB impacts digital transformation will differ between contexts, but argue that the theory of institutional drift would, albeit, be a valuable perspective to further understand the process of said impact.

Figures

Figure 1.

Voronov et al.’s (2021) process model of institutional drift

Figure 2.

Overview of three types of institutional drift

Figure 3.

Drift 1: decelerating digital transformation

Figure 4.

Drift 2: maintaining infrastructural stability

Figure 5.

Drift 3: accelerating digital transformation

Table 1.

Overview of category, frequency and empirical example

Category	Frequency	Empirical example
Using workarounds to bypass routines	19	“Sometimes you have to do the workaround because people are sitting, can’t do their work or something”
Using social networks to bypass routines	12	“And one thing that’s a problem, at least for us, is that it depends on who’s asking. We have a … not like hierarchy, it’s more like if you have a good relationship with them then you can get away with more and get stuff”
Installing or using unsanctioned IT	11	“That’s more like the shadow IT stuff because we don’t have the tools that are maybe good to work with and we go around saying stuff like ‘Miro’ that we use or I would use in workshops instead of Microsoft Teams Whiteboard that sucks in comparison. But as long as we keep it small and just for us then OK”
Not seeking approval	10	“I think I should ask more for approval, but I don’t do it so often. It’s not right, but it’s not completely wrong.”
Not contributing or disrupting meetings	10	“Yeah, I’ve been in meetings where they called the, you know a very non-IT person and said, ‘Oh well, it’s just to connect it to the Wi-Fi’ and then really … like one of the best IT persons we have that does not have good social skills call them an idiot because well, no you can’t do that”
Use of legacy systems	6	“Yeah, some of them that don’t use the program they like … now we are like 95% using Teams, but some still use Skype because they think Teams is not good enough or not safe enough for one example”
Bypassing the hiring process	5	“We have a couple of consultants in the project … to help us until December 2021. Let’s say … the project is not finished and if we should be very formal we actually need to make a new and maybe get two new people in. Uh, but sometimes we you know we let the consultants finish the jobs anyway”
Releasing unfinished projects or running shadow projects	5	“But sometimes IT takes decisions on their own … Uh, because maybe the line is not very clear …. what responsibilities are on the IT service center and what responsibilities are on my side if you understand. And so … people on IT service center sometimes do things … make priorities that are not in the way they are supposed to do.”
Extra-role or out-of-role behaviors	5	“Yeah, but sure some people can walk over to the person that's calling with the problem and help them at their office. …even though we shouldn't because we have other personnel for doing the on-site and problem solving
Prioritizing tasks and projects that conflict with DT goals	5	“And of course, we feel pressure that we have to, you know, try to help them and launch the projects. So sometimes we launch projects that are not maybe … we’re not finished with the preparations”
Running shadow projects that conflict with or are not prioritized by DT departments	2	“We should not do this right now and so of course, it’s kind of easy for the organization that has the capacity to start this anyway and not ask us so that they do it anyway, because we if we had the capacity to handle each case rapidly and with good quality and actually be a good support, I think we would have more projects in the daylight.”
Challenging the status quo	2	“I was told that I had to sign it like handwriting, and I said no, no why do we have this? It's unnecessary administration <… > some people say that if you are employed within the government or within as I am in the municipality, you should just follow decisions that is like somebody has said to you. I say the opposite”

Source: Authors’ own creation

Table 2.

Mediating factors of DWB

Mediating factor	Frequency
Organizational	229
Organizational structure	129
Role ambiguity	59
Lack of resources	41
Individual	150
Personality traits and beliefs	74
Knowledge and skills	48
Tenure	28
Technological	88
Technological limitations	56
Fast-paced development and DT process	32
Social	76
Organizational culture and climate	39
Social pressures to conform to group	37

Source: Authors’ own creation