Navigating role identity tensions — IT project managers’ identity work in agile information systems development

ABSTRACT

While the management of traditional, plan-based information systems development (ISD) projects lies with the information technology project manager (ITPM), this role is challenged in agile ISD team settings, characterised by self-managed and autonomous teams. Driven by the potential for individual, project, and organisational benefits through a deeper comprehension of these challenges, our study draws upon role identity theories to uncover identity tensions that this role creates and ways in which ITPMs engage in identity work to resolve these tensions. We conducted an in-depth, qualitative study involving interviews with ITPMs operating within agile ISD teams embedded in traditionally-managed organisations. A central contribution of our paper lies in the development of a theoretical model that unpacks how ITPMs navigate identity work tensions by engaging in specific identity work activities. By revealing the interplay between ITPMs’ role identity tensions, identity work activities, and contextual factors, our study moves beyond purely descriptive accounts and offers a nuanced understanding of the emotional and behavioural dimensions of ITPMs in their liminal position. Additionally, our work underscores the pivotal role of ITPMs as boundary spanners, bridging the gap between traditional and agile governance structures to drive digital transformation. For practitioners, we provide agile ISD team design recommendations.

KEYWORDS:

• Agile
• hybrid project setting
• IT project manager
• identity work
• identity tensions
• information systems development

1. Introduction

Ultimately, this is the key question: Does the project manager still exist in agile software development projects? And this is where you have to distinguish between theory and practice.

(ITPM 11)

To meet constantly changing requirements in volatile markets with high uncertainty, companies have increasingly adopted agile information systems development (ISD) (Rigby et al., 2018). In fact, a recent report finds that 86% of software development functions have already adopted agile principles and practices, while 94% of organisations claim to practise agile in some form within the company (Digital.ai, 2021). Agile ISD embraces a set of working practices rooted in the foundational principles articulated in the Agile Manifesto, such as responsiveness to change and the application of short iterations that focus on adding shippable increments to a software product (Fowler & Highsmith, 2001). Moreover, agile ISD teams are characterised by self-management and cross-functionality (Beck, 2000; Maruping et al., 2009). Agile ISD teams typically consist of 10 or fewer team members who are collectively responsible for adding value to the product in each iteration (Schwaber & Sutherland, 2020). Such agile ISD teams can operate with a high degree of autonomy because all product-related activities (e.g., stakeholder management, task prioritisation, making product decisions) can be performed within the cross-functional team. Consequently, teams are directly accountable for implementing, operating, and maintaining a feature or even an entire product (Schwaber & Sutherland, 2020). In contrast, traditional development methods like waterfall are facilitated by large functional teams with strict boundaries that work sequentially and in isolation to deliver a product or feature (Tripp et al., 2016).

Since agile ISD teams are self-managed and include a customer representative (Maruping & Matook, 2020), it is often proclaimed that the IT project manager (ITPM) role, typical of traditional development methods, is no longer necessary. This school of thought might originate in prominent agile literature such as the Spotify model (see Gerster et al., 2020) or the basic Scrum Guide (Schwaber & Sutherland, 2020), which do not presume any kind of manager within agile ISD teams. However, we have witnessed numerous descriptions of agile ISD team settings in practice that very much include an ITPM, thus suggesting the ongoing importance of the role (e.g., BCG, 2018). This is particularly true for organisations that undertake ISD activities using a hybrid model in which the surrounding organisation remains traditionally-managed and plan-based but deploys discrete ISD teams that operate on different projects using an agile approach (Cram & Marabelli, 2018; Port & Bui, 2009; Vinekar et al., 2006). Such hybrid configurations can introduce a diverse range of challenges, including the need to align and mediate between (either partly or fully) agile ISD teams and the surrounding, largely non-agile, organisation – a role that is often assumed by ITPMs (BCG, 2018). Within the greater context of digital transformations, these hybrid project settings serve a crucial purpose as they allow companies a gradual transition to agile working practices. Consequently, the lack of clarity on the ITPM role in hybrid settings leads to several concerns that motivate our research.

First, at the individual level, ITPMs within hybrid settings do not possess a well-defined role description, unlike other agile ISD team members such as the Scrum Master (SM) or Product Owner (PO) who have distinct responsibilities, tasks, and duties. The ITPM operates within an ambiguous and indeterminate space, occupying a liminal position that demands the fulfilment of divergent stakeholder expectations from both (i.e., traditional and agile) worlds. These dynamics inevitably give rise to role tensions among ITPMs, encompassing contradictory demands from top management and agile teams and the responsibility for projects without corresponding authority. In the absence of adequate guidance, ITPMs are left to grapple with these tensions independently, severely impeding their capacity to fulfil their designated roles. Consequently, this situation often precipitates work stress, diminished performance, and the potential departure of highly skilled, long-serving employees from organisations. Existing individual-level research in agile ISD settings has explored role challenges faced by developers (Benlian, 2022; Venkatesh et al., 2020; Windeler et al., 2017), client representatives (Maruping & Matook, 2020), and team leaders (Venkatesh et al., 2023). However, this line of research does not speak specifically to boundary-spanning roles that straddle diverse governance logics in hybrid ISD contexts.

Second, at the project level, past research has underscored the pivotal role of ITPMs in ensuring project success in traditional settings (Mata et al., 1995; Wiesche et al., 2018). However, the landscape shifts dramatically in hybrid project settings, marked by distinct team structures and unique challenges (Cram & Marabelli, 2018; Digital.ai, 2021; Gemino et al., 2021). Gaining deeper insights into the role of ITPMs in hybrid settings is crucial for IS research’s theoretical grasp of ISD project success in diverse contexts (Digital.ai, 2021; Lee & Xia, 2010). The growing acceptance of hybrid ISD initiatives, as opposed to rigid adherence to purely agile or traditional approaches (Gemino et al., 2021), accentuates the increasing significance of comprehending how ITPMs establish their role identity to contribute to project success in hybrid contexts. Previous research has primarily focused on structural and governance aspects (Haffke et al., 2017; Joehnk et al., 2019) or broader tensions and management practices in hybrid or bimodal IT ISD settings (Laux & Kranz, 2019; Toutaoui et al., 2022), yet specific insights into how ITPMs navigate their role identity within the liminal space bridging traditional and agile work structures remain limited.

Against this background, it is critical for research and practice to understand ITPMs and their role tensions in agile ISD team settings. Prominent theoretical underpinnings for ITPMs’ role are often based on command-and-control models, which seem unsuitable as theoretical guidance in hybrid contexts because they conflict with fundamental agile principles (Gregory & Keil, 2014; Napier et al., 2009). Therefore, we draw on work-related role identity theories, which provide a viable lens for our study. In the past, identity theories have explained numerous effects of role transitions, that is, changes in rights, duties, and obligations (Markovsky & Frederick, 2020). For instance, previous research has studied nurses becoming managers (Croft et al., 2015), individuals becoming priests (Kreiner et al., 2006), or artificial intelligence systems altering professionals’ roles (Strich et al., 2021). This strand of theories posits that role changes often engender role identity tensions that individuals try to resolve by engaging in role-based identity work—i.e., continuously forming or actively maintaining their identity or rejecting a new identity altogether (Caza et al., 2018). Accordingly, we seek to answer the following research questions:

RQ1:

What role identity tensions do ITPMs experience due to their shifting roles in agile ISD?

RQ2:

How do ITPMs resolve these tensions through identity work activities?

We adopted a qualitative research paradigm, grounded in the methodologies established by prior scholars (Deng et al., 2016; Dutton et al., 2010) to address our research questions. Our study involved in-depth interviews with 26 ITPMs operating within agile ISD team environments that were integrated into traditionally-managed organisations. This approach allowed us to delve into the intricacies of their work-related role identity tensions and the corresponding identity work activities, while also considering contextual factors influencing this dynamic.

Our study contributes to research on agile ISD, hybrid project settings, and digital transformation. Firstly, our research brings to light the work-related role identity tensions faced by ITPMs in the intricate landscape of hybrid project settings. This extends our comprehension of role challenges to encompass boundary-spanning roles, broadening our perspective beyond team-internal positions like developers, team leaders, or product owners (Venkatesh et al., 2018, 2020, 2023). Secondly, we deepen our understanding of how ITPMs meet these challenges in hybrid project contexts by revealing distinctive approach and avoidance patterns in their identity work activities, which complements research on managerial response patterns in hybrid, bimodal IT settings (Toutaoui et al., 2022). Notably, our primary contribution lies in the development of a theoretical model that synthesises the intricate relationship between how ITPMs navigate role identity tensions through identity work activities and the contextual factors shaping this dynamic. From a practical perspective, we provide helpful guidance for companies striving to overcome the obstacles they currently face in their agile adoption processes. We suggest proactively supporting ITPMs as they try to redefine their identities and cultivate a renewed sense of purpose within the evolving landscape of their hybrid work environment.

2. Theoretical Background

2.1. It Project Managers in Agile ISD Team Settings

Twenty-two years ago, the Agile Manifesto introduced twelve principles that revolutionised the software development process (Fowler & Highsmith, 2001). These principles emphasise short development cycles, shared decision-making, welcoming requirement changes, and continuous collaboration among team members. As one of the various methods to implement these principles, Scrum emerged as the most popular, focusing on project management and creative teamwork (Digital.ai, 2021; Rigby et al., 2016; Schwaber & Beedle, 2002). At the core of the agile ISD process lies a self-managed team accountable for determining their workload and effective collaboration (Highsmith, 2004). The empowerment inherent in the self-management of the ISD team is in stark contrast to traditional ISD project settings in which teams are commonly managed by multiple layers of authority and in a strictly hierarchical command-and-control style (Hoda et al., 2012). Compared to previous and more traditional approaches, these agile principles signified a radical paradigm shift that led to significant changes in both team roles and responsibilities.

As core members of the agile ISD team, developers are collectively responsible for managing processes, making decisions and delegating tasks among the team members (Maruping et al., 2009; McAvoy & Butler, 2009). In addition to developers, the customer representative is part of the team and responsible for ensuring close collaboration between the agile ISD team and the customer (Maruping & Matook, 2020). In Scrum, this role is typically assumed by the PO, who takes care of the product vision in consultation with the customer and oversees the product backlog that includes the customer requirements (Schwaber & Beedle, 2002). Moreover, a facilitator (i.e., the SM in Scrum), who knows common agile ISD practices well and helps the team to implement them by removing impediments, is part of the agile ISD team (Shastri et al., 2017). The duties of these defined roles in agile ISD teams overlap with those of ITPMs in conventional ISD teams. Traditionally, ITPMs assume a role of top-down leadership that includes centralised decision-making, the power to assign tasks to specific team members directly, and generally, a large amount of control and authority (Benington, 1987; Royce, 1987). They manage personnel, costs, schedules, risks and stakeholders (Nelson, 2007) and act in the centre of the team as a “conductor” in their project “orchestra” who is accountable for the success and failure of the project. Hence, among the roles most profoundly affected by the transition from conventional to agile ISD practices stands the ITPM.

In general, agile literature has long concentrated on firm-, project-, and team-level consequences of agile ISD method application, such as agile method adaptation strategies (Cao et al., 2009), challenges to knowledge sharing in agile ISD teams (Ghobadi & Mathiassen, 2016), or emergent agile team capabilities (Vidgen & Wang, 2009). More recently, researchers have also started to examine individual-level consequences for agile ISD team members with a strong focus on developers (Benlian, 2022; Mueller & Benlian, 2022; Tripp et al., 2016; Venkatesh et al., 2020), team-leaders (Venkatesh et al., 2023), and customer representatives (Maruping & Matook, 2020). However, research on other key stakeholders involved in agile ISD, such as the ITPM, remains scarce.

The limited body of research addressing agile ITPMs portrays the role as intricate, provisional and – most importantly – marked by pronounced contradictions. Most of this research confirms the continued presence and necessity of ITPMs in agile ISD team settings, despite the lack of a formal role description in the agile literature (Drury-Grogan, 2014; Gandomani et al., 2020; Shastri et al., 2021). The role that these individuals assume in practice appears to be structurally different from the central, authoritarian role of ITPMs in traditional ISD teams (Gandomani et al., 2020). Some previous work reports that ITPMs’ duties partly overlap with existing agile roles like the SM (e.g., Crowder & Friess, 2015; Hoda & Murugesan, 2016) or the PO (e.g., Miller, 2019), which leads to tangible conflicts between these team members and adverse project outcomes (Maruping & Matook, 2020; McAvoy & Butler, 2009; Siddique & Hussein, 2016). Other publications describe an entirely new role (Shastri et al., 2021) that, for instance, coaches the traditionally-managed top management in agile principles and attempts to establish an agile mindset to push the adoption through the entire organisation (e.g., Shastri et al., 2017; Tjørnehøj, 2019).

Most previous research points to the dilemma that ITPMs in agile ISD team settings are forced to juggle the expectations and demands from various stakeholders throughout the overall company. On one side, the ISD team’s members exert pressure on the ITPM to work exclusively agile, while external stakeholders like the organisation’s top management exert pressure by holding ITPMs responsible for project success, timely delivery, and budget compliance. Figure 1 illustrates the liminal role of ITPMs in agile ISD team settings as described in the extant literature.

Figure 1. Descriptive model of ITPMs’ liminal position in hybrid ISD team settings.

As depicted, ITPMs find themselves as a critical intermediary, couched between agile ISD teams and other organisational stakeholders, such as top management, other business units, and functional departments like marketing. At the fringe of agile ISD projects, this position comes with a wide new range of duties and obligations towards their respective counterparts. The exact role they assume depends on several factors, including whether the agile ISD team covers all other roles that are expected by agile methods (e.g., SM and PO) or the proportion of agile compared to plan-based practices applied in projects. Consequently, the role of ITPMs in agile ISD team settings is characterised by “in-betweenness” and contradiction, and thus bears the potential for tension. They are caught between the distinct governance structures of the self-managed agile ISD team and the traditionally-managed surrounding organisation. The study of individual-level consequences that ITPMs face in light of such contradictory role requirements is worthwhile since ITPMs’ effectiveness and efficiency appear to be key determinants of successfully navigating agile ISD projects in traditional organisations. According to prior research, ITPMs continue to play a vital part in agile ISD project success (Crowder & Friess, 2015; Tjørnehøj, 2019), yet companies risk losing these valuable employees if their role tensions are neglected or even ignored.

2.2. Identity Theories and Role-based Identity Work

The concept of identity has been a subject of extensive interdisciplinary inquiry over several decades. In this study, we ground our research within the framework of established identity theory literature (Burke & Stets, 2009; Stets & Burke, 2014; Stryker, 1980). Notably, our focus diverges from the collective level, which encompasses group behaviour and intergroup dynamics as elucidated in social identity theory (Tajfel & Turner, 1978; Vignoles et al., 2011). Instead, we specifically draw on structural symbolic interactionist theories as they provide specific theoretical guidance on the individual level (Stets & Serpe, 2016; Stryker & Burke, 2000). These theories encompass identity theory (Stryker, 1980), identity control theory (Burke & Stets, 2009), and role identity theory (McCall & Simmons, 1978), collectively referred to as identity theories. This theoretical approach aligns with the trajectory of previous IS literature, which has increasingly explored individual-level identity tensions in recent years (e.g., Carter & Grover, 2015; Carter et al., 2020; Whitley et al., 2014).

According to identity theories, individuals maintain several identities at all times – for instance, “in terms of a group or classification (such as being an American or female), in terms of a role (a stockbroker or a truck driver), or in terms of personal attributes (being friendly or honest)” (Stets & Burke, 2005, p. 45). While role identities refer to both personal role identities (e.g., being a mother) and professional role identities (e.g., being a team leader), we focus on professional role identities in the domain of an individual’s occupation (Higgins, 1987). One of these is the role identity someone maintains due to their position in the company (Caza et al., 2018). When individuals participate in work-related activities or memberships within organisations, they construct and internalise a set of self-meanings (Dutton et al., 2010). Accordingly, the position one takes on in relation to others in the context of work is defined by the duties, rights and obligations associated with this position (Caza et al., 2018; Markovsky & Frederick, 2020; Sluss & Ashforth, 2008). In doing so, individuals create meaning and legitimacy in a particular work role (Caza et al., 2018). Prior literature has discussed, for instance, how managers position themselves compared to the definition of their role (e.g., Cuganesan, 2017; Epitropaki et al., 2017) or how entrepreneurs shape and express their role (e.g., Lewis, 2013).

Role identities, as with any identity, continuously develop and are subject to an ongoing process that progressively resolves tensions when confronted with positive and negative options for development (Junglas et al., 2007; Phoenix & Rattansi, 2005). Hence, role identity tensions are within-person conflicts between a person’s existing role identity and potentially threatening new identities or identity facets, triggered by changes in central aspects of the work role. Triggers could take the form of a new role, a different role design and expectations, or conflicting roles (Caza et al., 2018). Individuals engage in role-based identity work activities to resolve these inner tensions and realign their self to the new meaning (Alvesson & Willmott, 2002).

Caza et al. (2018) identified four identity work modes: Identity work can occur cognitively (e.g., changing cognitions), discursively (e.g., engaging in discourse about new roles), behaviourally (e.g., behaving in line with new role expectations), and physically (e.g., changing physical representations). When enacting role-based identity work using one or more of these modes, individuals aim to form, repair, revise, maintain, strengthen, or reject their role-related self-meanings within their social contexts (Alvesson & Willmott, 2002; Caza et al., 2018). They form their role identity when they adapt their thoughts, actions, discourses, or symbols to construct a new subjective interpretation of who they are that is aligned with the changed role characteristics and expectations. When these changes are so pervasive that parts of the existing role identity are damaged (e.g., being a victim of bullying), individuals engage in remedial identity work to repair their interpretations (Lutgen-Sandvik, 2008). They may also revise their role identity in the sense that they reassess a previously accepted role identity and take corrective actions (Beyer & Hannah, 2002). Moreover, when individuals are confronted with role changes they cannot accept, they can engage in actively maintaining the existing identity, such as by interpreting and designing the new role characteristics or expectations in a way that they can hold on to their existing role-related self-meaning. Some individuals might even reinforce and strengthen their existing role identity to solve role-related tensions. Lastly, individuals may reject a new role identity when the new role characteristics are incompatible with the meanings they attach to themselves, when they feel unable to realign to the new meaning of the role and when they see no value or future in the existing role (Westwood & Johnston, 2012).

In the context of agile ISD, the characteristics of the ITPM role are changing (Gandomani et al., 2020; Shastri et al., 2017). ITPMs often find themselves couched between various stakeholders and are exposed to an array of diverging demands (Tjørnehøj, 2019), including the competing demands of the self-managed agile ISD teams and the overall organisation. In addition, their traditionally top-down approach to leading is no longer suitable due to the agile ISD team making its own project-related decisions (McAvoy & Butler, 2009). As their role changes from one of a central authoritarian role that is responsible for guiding the project team members to one tasked with serving as facilitators or intermediaries, ITPMs working in an agile ISD team setting are likely to construe themselves differently in their work domains than ITPMs in fully traditional ISD project settings (Stewart et al., 2017). With the inevitability of hybrid settings in many organisations (i.e., settings in which agile ISD teams are embedded in traditionally-managed, plan-based organisations), ITPMs have to deal with complexities, inconsistencies and tensions in the meaning they attach to themselves and their new role, causing role-based identity work to happen in order to solve conflicts and to align with new role expectations.

Research has demonstrated that work-related identities have the potential to affect important individual-level work outcomes, e.g., creativity or the capacity to deal with adversity and stress (Caza & Bagozzi, 2009; Cheng et al., 2008; Hobfoll, 1989). Hence, we believe that an individual’s work-related role identity and role-based identity work have the potential to provide a solid basis for understanding individual-level consequences (e.g., tensions) for ITPMs in agile ISD team settings as well as broader implications for the success of agile ISD projects and organisation-wide digital transformation processes. To study ITPMs’ role identities and their implications, we conducted a qualitative research study, which we present next.

3. Research Methodology

We draw on a qualitative research approach by using selected principles associated with grounded theory (Corbin & Strauss, 2008; Glaser & Strauss, 1967; Sarker et al., 2018; Strauss & Corbin, 1990). Broadly, grounded theory is suitable for research that does not begin with an a priori theory that is intended to be tested but one where the empirical data is used as a foundation from which to inductively elicit relevant concepts. This method is in keeping with the approach advocated by Strauss and Corbin (1990) and Myers (2009) that allows for an awareness of the relevant literature and theoretical foundations, but not to the extent that it constrains creativity and idea generation.

We adopted this approach for two primary reasons. First, while remaining systematic and adhering to existing theoretical foundations, our study is notably anchored in empirical phenomena. Our research aims to develop a profound understanding and generate rich descriptions of individual-level phenomena, making detailed statements and experiences of utmost importance. Therefore, employing a qualitative, inductive approach allowed us to construct and develop new concepts using identity theories as a theoretical lens within the novel context of agile ISD. Previous studies of a similar nature have demonstrated the appropriateness of selected grounded theory techniques (e.g., those focused on data analysis) for exploring detailed concepts that encompass practitioners’ personal positive and negative experiences (e.g., Cram et al., 2021; Gioia et al., 2012; Jones & Noble, 2007; Martin & Turner, 1986; Myers, 2009). Second, we employ a qualitative approach because we investigate role-based identity work which is based on subjective realities. This requires that we use real-life research settings that are as close to actual IS practitioners as possible – and qualitative research lends itself particularly well for these settings in comparison to alternative approaches (Trauth, 2013). Specifically in combination with identity theories, extant literature has already shown that a qualitative approach is effective in examining leaders’ in-depth role-based identity work (e.g., Stewart et al., 2017).

3.1. Data Collection

Our research questions are oriented around the role identity tensions experienced by ITPMs on agile ISD projects. As introduced above, ITPMs in agile ISD teams often act as intermediaries between the surrounding traditionally-managed organisation and the agile ISD team members. As a result, they can offer valuable insights on their own role identity tensions but also present a unique perspective into the consequences for project outcomes and – on a higher level – the successful agile adoption process. Consequently, in order to collect data from employees who experienced this specific scenario, we used purposeful sampling, following previous research (Palinkas et al., 2015; Wiesche et al., 2017), with the following sampling criteria: (1) the interviewee holds or held the position of an ITPM in an agile ISD team setting, (2) the interviewee has work experience as an ITPM in traditional IT project contexts, and (3) the interviewee works in a company that remains predominantly plan-based and traditionally-managed. Using a snowball approach, we posted calls on social network sites (e.g., LinkedIn), invited our personal network to forward our call, and asked companies in our research network to forward our call and selection criteria to their ITPMs.

All interviews took place via phone or face-to-face, lasted between 40 and 60 minutes, and were audio-recorded and transcribed. As we followed an iterative approach to data collection and analysis, these phases were closely intertwined. We increased the number of interviews until no new theoretical insights relevant for this study emerged and additional interviews were considered to have minimal benefit (Ramesh et al., 2012; Salo et al., 2020). Eventually, after conducting and coding 26 interviews, we reached the point of theoretical saturation and terminated data collection (Keutel et al., 2014; Stake, 2005).

The interviewees came from mid- to large-size companies in line with our intention to capture experiences from ITPMs in a predominantly traditionally managed organisational setting. Most of the interviewees worked for companies operating within the financial services sector. Among the participants, approximately one-third were responsible for managing a single agile project or team, while the remaining two-thirds oversaw multiple projects. The average age of our interviewees was 42 and around 85% were male. The majority of the interviewees began their careers as developers and had an average work experience of 19 years. On average, they had been employed at their current company for 10 years and had approximately 5 years of experience with agile practices. Additional information can be found in Table A.1, provided in the Appendix.

All interviews were semi-structured and followed an interview guideline (see A.2 in the Appendix) building on role identity theories. In line with our iterative approach to data collection and analysis, the interview guideline was adapted during our study according to emerging results of the analysis. The guideline covered the topics of separation of activities and responsibilities among the various stakeholders like the development team, the SM, or the PO (e.g., “Who is responsible for the project?”), relations between the stakeholders and the ITPM (e.g., “How would you describe your relation to the ISD team?”), and how ITPMs would describe their role (e.g., “How would you describe your role as manager of a self-managing team?”) (Myers & Newman, 2007). Although the concepts of role identity theory form the backbone of our interview guideline, we stayed open to field data and new emerging concepts by using open questions and asking our interview participants, for instance, if they could envision a successful agile ISD project without an ITPM.

3.2. Data Analysis

Our analysis of the interview transcripts included several iterative rounds of open coding (Corbin & Strauss, 2008). In general, we sought to identify patterns of experiences that were expressed across multiple ITPMs (i.e., cross-case analysis). In doing so, we remained open to individual-level role identity tensions, as well as broader project or organisation experiences. Throughout the entire analysis, we used role identity theories as a lens for our interpretation and to identify relevant text passages (Sarker et al., 2018). In the first phase, we used the techniques of Hatch and Ehrlich (1993) and Andriopoulos and Lewis (2009) to identify transcript passages in which interviewees directly referred to identity tensions or contradictions in their work-related roles. For instance, one ITPM hinted at conflicting work settings and role expectations by stating that despite employing agile methods, companies continue to have strict hierarchies: “I do believe that agile methods have changed management behaviour because an agile team is managed differently than a waterfall team. But the hierarchy we have is still the same” (ITPM 15). Another ITPM mentioned a new role design saying that “people don’t report to you anymore. You work with the team, and the team doesn’t work for you” (ITPM 14). This ITPM concluded that the shift in collaboration patterns is challenging for them: “Communicating on an equal footing, but still having a functional responsibility is a fine line, a huge challenge”. We summarised these passages in which the ITPMs motivate potential identity tensions triggered by new role expectations, designs, and settings and gave them short, descriptive labels – our first-order codes. This was done in isolation first – each author did their own separate coding – before we compared, discussed, refined and lastly consolidated our findings into a range of first-order codes that were coherent across several interview transcripts. We quickly agreed that many themes were shared across interviewees – despite them working in entirely different companies or even industries. Using the previous examples, the conflicting settings in the quote of ITPM 15 were the “Changed management behaviour according to agile principles” on one side, and the “Company hierarchy remains the same” on the other. ITPM 14 acknowledged that there is “No explicit or implicit hierarchical order between ITPM and team” anymore. However, because the “Responsibility remains with the ITPM”, they are in a perceived superior position impeding collaboration on eye-level and leaving the ITPM with the challenge to figure out their new role. According to the Gioia methodology (Gioia, 2020), we subsequently grouped similar first-order codes and arrived at 10 second-order codes. We then aggregated these second-order codes into dimensions according to the tension-generating patterns the interviewees experienced regarding their role identity. Figure 2 shows our coding data structure concerning role identity tensions.

Figure 2. Coding data structure regarding role identity tensions.

In a second phase, we closely followed the identity work framework of (Caza et al., 2018) and coded text passages that indicated how the ITPMs cope with the role identity tensions they experience. ITPM 4, for example, explained that building trust is vital to be able to let go of control and influence over the team, saying, “I have less influence on my team’s decisions on how to carry something out, and I had to overcome that at first, but I can only do that if I trust my team, and vice versa, my team trusts me. And when I look back on it that way, it’s good that we do it this way” (ITPM 4). We concisely labelled these explanations resulting in our first-order codes. Subsequently, we grouped similar codes and mapped them to the identity work modes presented in the theoretical background (e.g., cognitive, behavioural) – our second-order codes. Finally, we mapped these modes to the likewise aforementioned identity work activity goals (e.g., maintaining, forming). Figure 3 summarises the coding data structure regarding role-related identity work of ITPMs.

Figure 3. Coding data structure concerning role identity work activities and goals.

Following the coding of identity tensions and identity work activities, we conducted an analysis of our coded textual passages to identify potential patterns in how identity tensions are addressed through identity work activities. By systematically examining each role identity tension and investigating the ITPMs experiencing these tensions, we sought to determine the specific identity work activities employed in response. Similarly, we scrutinised each identity work activity and assessed the ITPMs who performed these activities to determine whether they encountered specific role identity tensions. For example, the ITPMs whose statements suggest tensions related to teamwork indicate that they respond by forming a new identity, either cognitively or behaviourally.

In a third phase, we conducted an additional assessment of our interview data, focusing on potential contextual factors that may influence the identified role identity tensions or work activities.¹ We contend that the organisational and individual environments in which ITPMs operate represent boundary conditions vital for understanding the origins of these tensions and their corresponding responses. These factors are instrumental in refining our interpretation of the findings and substantiating our claims. We identified transcript passages indicating organisational or individual-level context factors. For instance, ITPM 10 states a preference for granting autonomy to the team instead of imposing strict deadlines: “I’m not the type of person who dictates tight deadlines. So, if the team works collaboratively and achieves the overarching goal, I’m completely okay with that”. Similarly, ITPM 18 expresses a preference for an empowering environment, stating, “I’m also the type of person who chooses my own tasks. In this respect, agile suits me.” Consistent with the approach taken in the initial and second phases of our analysis, we labelled these statements as first-order codes and grouped them into dimensions, such as leadership style. Throughout this follow-up analysis, special attention was given to determining whether these factors had an influence on the previously identified identity tensions and work activities. This involved examining whether ITPMs expressing a specific contextual factor encountered particular identity tensions or employed specific identity work activities in response to tensions. Only those factors with a clearly discernible impact were included in our analysis.

4. Results

Our first research question sought to identify the within-person role identity tensions that ITPMs experience due to their ambiguous role in agile ISD projects. Figure 4 synthesises the findings from our analysis concerning the role identity tensions. It reveals five tensions (labelled A to E) that result from contradictory demands from the agile ISD team on one side and from the traditionally-managed organisation on the other. Pulled between the divergent governance structures of these two extremes, ITPMs are constantly reshaping their role boundaries (dashed line) in an attempt to construct fitting self-meanings for their work role. The following sections describe these tensions and delve into the subsequent identity work activities.

Figure 4. Role identity tensions of ITPMs in hybrid ISD settings.

4.1. Role Identity Tensions for ITPMs

4.1.1. Tension A — Responsibilities towards Top Management

The first tension we observed emerging from our data is about the ITPMs’ delivery responsibility towards their top management, despite having a highly self-managed ISD team. In traditional ISD project settings, ITPMs were in a central position as a kind of “conductor” in their project “orchestra” that incorporated both responsibility (i.e., for timely delivery of a product, staying within the allocated budget, level of quality, etc.) and the means to steer the team members such that the ITPMs could influence the project’s development accordingly. Our interviewees reported that they continue to carry this responsibility for the whole endeavour, although they are working in agile, self-managed ISD team settings:

There is always only one culprit, and that is the project manager. That is already clear, and that remains. No matter whether agile or classic or mixed, hybrid.

(ITPM 24)

But on the whole, you’re still responsible for the budget, for the timeline, for stakeholder management.

(ITPM 19)

Because teams are self-managed, ITPMs are no longer in a position to set schedules and deadlines or assign specific tasks to developers. One ITPM connected this high degree of the ISD team’s self-management to issues they experience in relation to the organisation’s top management.

Personally, I would like to see the project manager move more in the direction of a grey eminence in the background. That the team is able to organise itself on certain topics, to demand requirements, and so on. And that the project manager only sets the guidelines and intervenes in crisis situations. [But], when things don’t go so well, I think this gets very difficult within the team. This may sound stupid, but [our top management] is usually looking for a culprit. (ITPM 25)

This sentiment was shared among several interviewees. They reported that the organisation’s top management has not internalised that, according to fundamental agile principles, the responsibility for project outcomes (i.e., timely delivery, budget compliance, etc.) needs to be shared among all members of the agile ISD team. By still holding the ITPM accountable, they create contradicting expectations that result in tensions for the ITPM’s role identity. One ITPM expresses their respective negative feelings quite clearly, saying:

Suddenly being the one who does the work but no longer leads. No longer being able to influence things as one sees fit. But possibly having your head chopped off for it. For many, that feels like a demotion.

(ITPM 23)

4.1.2. Tension B — Top Management Communication

A second core tension mentioned by multiple interviewees deals with communication and reporting to the top management. ITPMs are used to representing the projects for which they are responsible to top management. They hold the relevant project information concerning status and performance to inform the top management about the project’s progress and potential risks. Our interviewees stated that in agile ISD settings, top management still requires this constant information flow from ITPMs, a role expectation that conflicts with agile structures.

I believe that in large corporations, things like status reporting should be covered by the project manager. They should take on these governance topics, but below this, [companies can] build an agile organisation. (ITPM 6)

The theory sounds nice, but only if I’m working on a greenfield project. Which means I could work exclusively in an agile way. That’s not the case in most projects or companies. We have classic structures, from budget to reporting. And there will always be a project manager.

(ITPM 11)

ITPMs in agile settings are no longer part of the agile ISD team. They move to the fringe of the project, acting as “boundary spanners” to the rest of the organisation while having little access to the status and content of team tasks. Consequently, the obligation to report collides with their new position and the missing immediate access to project information. To communicate effectively, ITPMs need to incorporate an agile mindset towards their self-managed agile ISD teams and a traditional mindset towards top management that expects detailed project reporting concerning time, quality and budget. This tension was summarised by one interviewee as follows:

I think the fact that the organisation is not really set up in an agile way puts the project manager in a bit of an in-between position. You soften the impact in the direction of the team because you don’t want to let this [traditional] organisational structure affect them so that the team can still work in an agile mindset. Externally, however, you still have to communicate official timelines, budgets, and so on. So, it’s a balancing act between not allowing anything to get through internally that would somehow make you lose your way, but then translating it upwards into organisational communication.

(ITPM 14)

4.1.3. Tension C — Collaboration with Agile ISD Team

The third tension we identified originates in a perceived loss of hierarchy and authority between the ITPM and the development team. In traditional project settings, development teams are evidently hierarchically structured in that ITPMs have a perceived superior position. This is reflected by the fact that they can make extensive decisions and are entrusted to delegate specific duties to specific members of the team. In contrast, there are no hierarchies in agile. Instead, agile project settings aim to eliminate such rigid hierarchical structures and processes in favour of shallow or even non-existent hierarchies and more frequent stakeholder interactions. This shift leads to pronounced role identity tensions for some of our interviewees because they notice that their previous collaboration approach is not in line with agile ISD methods. One interviewee explicitly notes that despite agile settings, they still feel like the ITPM ranks hierarchically higher:

In practice, the project manager ranks higher than the team. That is a fact for now.

(ITPM 11)

The tensions caused by differing collaboration patterns intensify given that hierarchies still seem to exist even though agile principles have been introduced:

If everything is completely agilised [sic], then you can do away with team leaders and department heads. […] That doesn’t exist in most companies. They are still strongly hierarchically organised.

(ITPM 23)

This still exists, this old way of thinking in hierarchies. (ITPM 17)

This discrepancy in how hierarchies are perceived throughout the organisation presented a particularly challenging situation for ITPMs. They are forced to seek new ways of collaboration and communication with the ISD team members as their equal peers without losing the authoritarian and high rank within the overall organisation. One ITPM explicitly recognised the challenge of carrying functional responsibility within their hierarchical position while being on a par with the agile ISD team members and treating them as equal peers:

People don’t report to you anymore. You work with the team, and the team doesn’t work for you. […] Communicating on an equal footing, but still having a functional responsibility is a fine line, a huge challenge.

(ITPM 14)

4.1.4. Tension D — Control of Agile ISD Team

We identified a fourth role identity tension that is related to the distribution of tasks and decision-making in the agile ISD team. In their traditional role, ITPMs have full decision control, a great say in who does what in the ISD project team, and tight tracking mechanisms implemented. In an agile ISD setting, project teams take their own decisions and assignments and do not want the ITPM’s interference. The ITPMs are expected to let go of their top-down, command-and-control way of leading and come to terms with the new bottom-up approach, which defines their duties and responsibilities differently. This situation is new to many ITPMs, and when teams act in a very self-confident manner, the ITPMs do not know how to react and feel uncomfortable, as one interviewee indicated:

I tried to manage it “the project manager way” and got a lot of resistance. […] Some were happy that someone was there again to tell them when they could go to the bathroom again, but the ones who accept this new responsibility put you up against the wall. […] It’s a lousy situation. You think you know better. You think you know how to solve it, and you think you know who can solve it. Then you go, “Listen, guys, the problem is this. Can you take care of it?” And they look at you and say, “Well, no. You don’t have anything to say here”. That’s a huge mess. At that moment, you become very aware that your duties have changed. And that your leadership responsibility is gone.

(ITPM 23)

In addition, in agile ISD settings, the number of projects one ITPM is responsible for is high. This different project organisation makes it impossible for ITPMs to distribute tasks and be part of decisions in all of their projects. As one ITPM explained:

I’ve had to scale that down because when you’re primarily involved in agile projects, it’s, of course, difficult to do justice to everyone when you have to be responsible for so many projects at the same time.

(ITPM 19)

Both the agile ISD team’s expectations and the overall agile ISD project organisation trigger tensions in the ITPMs’ self-perception. The traditional idea of project leadership, hence, the rights and duties attached to the ITPM role, is challenged, which forces the ITPMs to rethink and redefine their role identity.

4.1.5. Tension E — Distribution of Duties and Responsibilities

We finally observed a fifth and final identity tension among ITPMs in hybrid ISD team settings concerning the ambiguity of their role when compared to the established roles of SM and PO. Duties of SM and PO are described and well-known across organisations and include, for instance, the PO being the interface to the customer and the SM coordinating the project team’s activities. The duties of ITPMs in agile ISD team settings, however, are not explicitly formulated, leaving the ITPMs alone to figure out a new role design:

This makes it more difficult for a project manager in an agile project because there is no designated role.

(ITPM 24)

Based on their past, ITPMs have their own understanding and perception of their duties which includes, for example, risk management with the customer and reporting to top management:

You get into a conflict with the Product Owner in this situation. Because if you say you’re responsible for risk management, then you want to talk to the customers about risk management […] It’s one thing Scrum doesn’t say anything about. Does the product owner do business risk management: yes, or no? According to Scrum, yes, he is mainly responsible for the requirements and prioritisation together with the team. Is he the interface to the business for all topics, or is he only the interface to the business regarding the requirements? And doesn’t every requirement also involve risk, so of course, he has to do the risk management? That’s the conflict because this is not documented or defined anywhere. The Product Owner says, “No, that’s my job. Why did you speak with the customer yesterday?” Then I say, “That’s not your job. That’s my job. I’m responsible for the risks”.

(ITPM 23)

Consequently, overlaps between the duties of SM, PO, and ITPM roles might happen, which leads to ambiguity and tension. One interviewee described this tension as follows:

I believe that it will take quite some time until this topic of agile projects is really anchored in such a way that everyone knows exactly what they have to do or refrain from doing in their role at that moment.

(ITPM 22)

SM and PO can build on their established roles with clearly assigned duties and experience this tension as weak, while ITPMs face an important tension affecting their self-perception. ITPMs’ duties, an integral part of their role, are not explicitly formulated and might not correspond to what the ITPMs expect from their role. Hence, ITPMs might redefine their duties together with the agile ISD teams (e.g., with PO and SM), thus, redefining their identity.

4.2. ITPM’s Role-based Identity Work

In response to our second research question, our data revealed different role-based identity work activities exercised by ITPMs aiming to navigate the identity tensions encountered in agile ISD team settings. In particular, as can be seen in Figure 3, we discerned both approach and avoidance patterns in role identity activities, where participants engaged in cognitive and behavioural activities that resulted in either forming a new, actively maintaining an existing, or rejecting a change to their role identity. Some changed or adapted their mindsets and behaviours to form their own and others’ perceptions of the new ITPM role. Other interviewees stuck to existing processes to preserve and maintain their existing self-meanings in agile ISD contexts. One individual refused to adapt to the new situation and resigned from her job, not being able to construct a role identity in a way that rejection would not have been necessary. As a preliminary overview, Table 1 previews the observed patterns, which will be explored in greater detail below, highlighting the relationship between the previously described role identity tensions and the identity work activities that will be examined next.

Table 1. Patterns in the relationship between role identity tensions and work activities.

Pattern	Experienced Role identity tensions	Identity work activity goal	Used Identity work activities
1	Collaboration with agile ISD team (C) Control of agile ISD team (D)	Forming a new role identity (Cognitive)	New understanding of soft skills and own importance (Cognitive)
2	Communication with top management (B) Collaboration with agile ISD team (C)	Forming a new role identity (Cognitive and Behavioural)	New understanding of soft skills and own importance (Cognitive) Pulling back from the agile ISD team (Behavioural) Interacting with other Scrum roles to redefine ITPM (Behavioural)
3	Responsibilities towards top management (A) Communication with top management (B)	Actively maintaining existing role identity (Behavioural)	Continuing with traditional duties (Behavioural)
4	All tensions (A-E)	Rejecting changes to role identity (Behavioural)	Leaving after feeling demoted/downgraded (Behavioural)

4.2.1. Forming a New Role Identity

According to our findings, ITPMs form a new identity in three different ways (see also Figure 3). To approach the new role or role expectations, they may work on a new understanding of the relevance of certain soft skills and of their own importance, thus fostering a cognitive evolution of their identity. Moreover, they may also withdraw from the agile ISD team or intentionally engage in interactions with the roles explicitly delineated within the Scrum framework to redefine their own role, thereby behaviourally forming a new identity.

In addition to these three pathways, our results suggest that by forming a new role identity, ITPMs especially navigate identity tensions related to top management communication as well as collaboration with and control of the agile ISD team. In particular, our data indicate that cognitive identity development more often tends to address tensions surrounding team collaboration and control, while tensions linked to top management and team communication are more often handled by a combination of cognitive and behavioural identity development.

New Understanding of Soft Skills and Own Importance. Several ITPMs enact cognitive identity work by engaging in a mindset change. When forming this new mindset, they attach the highest importance to empathy and integrity in order to build and foster trust, especially in the relationship between the ITPM and the agile ISD team. Thereby, the ITPMs enable meaningful collaboration. Two interviewees describe the altered relevance of soft skills in the following quotes:

With a[n] [agile] team that still has to find itself, I think empathy is particularly important. Bringing people along and letting them find each other. Ultimately, helping to generate this trust […] is very important. In addition, supporting the team so that they build trust among themselves and bring empathy into the team and help them in this way. […] Integrity in any case [is important]. […] I think the most important thing is that you stand by your people and really back them up, even if something unpleasant happens.

(ITPM 13)

This concept of emotional intelligence, after all, means empathy. I think it is very important that you assess people in such a way that you give them the support that they need.

(ITPM 14)

Building a strong bond with the agile ISD team based on trust is additionally a cognitive strategy to overcome identity tensions triggered by the loss of control over the team’s decisions and the team members’ tasks. Trust creates the possibility for ITPMs to mentally let go of the will to directly control the team and accept a bottom-up leadership approach like this ITPM states:

There may be times when they say, “I’ve already got that on my radar. I’ll get it done”. Then that’s absolutely okay for me.

(ITPM 17)

This personal adjustment at times takes some effort but is worth it, as two of our interviewees report:

I have less influence on my team’s decisions on how to carry something out, and I had to overcome that at first, but I can only do that if I trust my team, and vice versa, my team trusts me. And when I look back on it that way, it’s good that we do it this way.

(ITPM 4)

The typical situation is: We got a few million and want a new feature. You need two persons from the team and thus throw the team together again to solve the problem. Versus you have a problem and give it to a well-functioning, running team to solve. That’s the difference for me. […] Again, that’s top-down versus bottom-up. Because top-down means you get people together versus bottom-up, where you say, “Dear team, you are good, you are finding the right way”. To do that, you need trust with your teams so that you can let go.

(ITPM 13)

Understanding and admitting to themselves that the agile ISD team does not necessarily need their input anymore to perform successfully is another form of ITPMs enacting cognitive identity work. If ITPMs accept that the less they get involved, the better, then they are not only able to reduce the team’s feelings of being micromanaged but also to lead many agile ISD projects to success in parallel. This mental adjustment is described by one ITPM as follows:

But at some point, I learned, or understood, to also say, the less this project actually needs me, the better it goes. […] That’s a bit of a thing where you have to be clear with yourself and admit to yourself that it’s great when it all works and the guys don’t actually need me. Then I’ve done everything right, then it works. But you have to be conscious of that. […] I’m happy when I say I have to intervene as little as possible in a controlling way.

(ITPM 18)

4.2.1.1. Pulling Back from the Agile ISD Team

As another identity work activity to form a new role identity, the interviewees emphasised behavioural actions by pulling back from the agile ISD teams – often hand in hand with a mindset change. To manage the expectations of top management concerning project reporting, some ITPMs decided to intervene less within the ISD project (e.g., by adjusting task prioritisation) and instead attempted to facilitate the process on a higher, organisational level (e.g., by increasing the project’s budget). One interviewee explains:

I don’t look at the details from the top at all, but I still define the key deliverables at a fairly high level. This actually has to do with the fact that I want to interfere as little as possible in the actual agile project. […] Sure, there are those [situations where I intervene]. That usually happens whenever I notice that one of the three criteria is starting to get out of hand, quality, time or resources. […] More on the organisational side, I then say, “What would help you now? Is it money that is missing? Is it resources that are missing?”

(ITPM 18)

Other ITPMs stepped back from being the single point of contact to top management. Instead, they bring the top management and the development team closer together, thereby designing the information flow as direct and transparent as possible, as one ITPM stated:

What I usually do is invite our board members. They come here and attend a Sprint Change Revision for 20 minutes. Anyone can talk to them because they’re just normal people. And they are actually always quite happy to see something tangible. […] So, what I try to do is to be as transparent as possible. That’s why I don’t buffer everything in the middle. When someone does something good, they can present it to the Board of Management—that’s not my job.

(ITPM 13)

ITPMs also decided to withdraw themselves from day-to-day operations within the agile ISD team and move closer towards the top management. By doing so, the ITPMs hold on to a certain level of hierarchy towards the agile ISD team and, at the same time, allow for the team’s self-management. Additionally, they can act as an agile coach to top management, as one ITPM highlighted:

If we now look at the mind-set, I see a very different important role. That is top management. A cultural change like this only works if management is fully behind it. […] I would say that the project manager takes on a superior role and is less directly involved in team activities.

(ITPM 11)

Actively forming new communication patterns with their agile ISD teams is another behavioural adaptation path that the interviewees took. Namely, they switch from a hierarchical way of communicating to communication on equal footing, as one ITPM explained:

I think the moment you give more responsibility to people in the team, you have to communicate with them differently. You no longer dictate things to the team; you communicate as equals.

(ITPM 14)

4.2.1.2. Interacting with Other Scrum Roles to Redefine ITPM

In addition to ITPMs pulling back from their agile ISD teams, we observed that ITPMs adapt their behaviour by deliberately interacting with the different roles that Scrum explicitly defines to redefine their own role. Some interviewees consciously involve themselves in the team’s day-to-day operations, for example, in terms of optionally attending team meetings. The ITPMs emphasised the importance of staying up-to-date to be able to pursue their other tasks, such as stakeholder management, reporting, and escalation management. One ITPM explained the following:

I try to get involved a bit so that I can at least have a say and understand why decisions are made the way they are. And that is appreciated. […] This stakeholder management that I mentioned at the beginning is essential, and from my point of view, you can only do stakeholder management if you know what you’re talking about.

(ITPM 19)

The ITPMs also discussed their role with the SM and the PO intensively. Initial interactions were used as experiments to refine the ITPMs’ role, including a split of responsibilities and duties between SM, PO, and ITPM and their perception of it. The following quotation underlines this point:

The Scrum Master simply has to signal to the project manager when nothing can be changed because the rollout of Sprint 5, for example, is imminent. […] It’s important to have someone who is in a position to give the project manager a leg up.

(ITPM 19)

4.2.2. Actively Maintaining Existing Role Identity: Continuing with Traditional Duties

Besides these different tactics to approach a new role identity, some ITPMs also reported that they very consciously avoided changing their behaviour and stuck to existing processes. Especially, ITPMs who experienced more identity tensions towards the top management, i.e., regarding responsibility towards and communication with the management, tended to engage in actively maintaining their existing identity. These ITPMs continued with traditional duties navigating their identity tensions by defying the agile ISD teams’ expectations of self-organisation. They legitimise the intervention by saying that Agile cannot control itself. Three ITPMs succinctly summarise this identity work:

I moderate that sprint planning as well. I don’t feel any different, then. Especially as I still have overall delivery responsibility for the entire project. And, of course, I still have to see whether the self-organising is going well. Is there any progress, and is something delivered or not?

(ITPM 21)

My management style has not really changed. […] Of course, I make sure that certain milestones are reached and break that down, keeping an eye on the overall project plan so that the target is not missed.

(ITPM 4)

For me, it’s like there really still has to be someone who keeps an eye on all the chaos. Individually, the teams are limited in their agile thinking.

(ITPM 1)

In the same way, neglecting the agile framework of self-managing teams, other ITPMs reserved the right to distribute tasks directly when they saw the need. Especially when important topics are not picked up by the teams themselves, but the ITPMs see the importance because they are acting outside of the agile ISD teams, an overruling of team-internal priorities is considered necessary. As the following quote of one interviewee illustrates, the ITPMs can combine their intervention with other project management tasks, such as employee development:

Sometimes other tasks come in that are now more important or subject blocks that really have to be done for some reason, then I also assign them. […] In the team, of course, it is perceived as a cut in the personal freedom of the development team. On the other hand, it can also be used to develop the employees. This is also a task of project management. Saying, “Hey, here. Please take a look at the task. I would like to do that now. Please familiarise yourself with it. Get in touch with new topics”.

(ITPM 25)

In addition, other interviewees retained traditional communication processes and continued to see it as their job to pass information to and from top management. They consider this the best solution for their organisational and personal situation. One interviewee stresses that especially the organisation in large corporations still needs traditional hierarchical communication with the management:

So, I’ve seen both, and I’m of the opinion that in large enterprises, you still have to have things like status reporting covered by the project manager to serve these governance topics, but underneath, a certain agile organisation is built.

(ITPM 6)

4.2.3. Rejecting Changes to Role Identity: Leaving after Feeling Demoted/Downgraded

Besides forming and maintaining as identity work activity goals, we identified a third goal by one interviewee, namely rejecting the new role and the changes that come with it. The ITPM experienced almost all tensions we identified above resulting in complete rejection of any further identity development. While identities evolve within certain boundaries, the transition into agile ISD team settings might overstretch these boundaries for some ITPMs. Managing a project without the clear possibility of steering and controlling the team felt more of a Project Management Office (PMO) role, which did not correspond to their expectations of an ITPM role and identity:

That felt like a very heavy demotion from a project manager who managed major projects in the millions of dollars to, it felt like a PMO

(ITPM 23)

By leaving the project(s) and even the organisation, one of our interviewees marked the rejection of the ITPM role in agile ISD team settings (without necessarily rejecting the concept of agile ISD):

Yes, I had this position [of an ITPM who is equal to the team] for a certain time. But that does not work with me.

(ITPM 23)

Rejection was a necessity because the ITPM was not willing or able to form a new identity or actively maintain it in the new setting. Leaving the organisation resolved the tensions for the ITPM in this instance. However, they are likely to reappear in other agile ISD team settings.

4.3. Supplementary Analysis on Contextual Factors

Through our interviews, we further probed into contextual factors, both organisational and individual, that significantly impacted ITPMs within their respective companies. While our study did not initially prioritise the examination of contextual elements, four factors surfaced organically during the analysis. The nature of these factors can variably shape the identity tensions encountered and the identity work activities undertaken by ITPMs.

First, on the organisational side, more than half of the companies operated within the tightly regulated financial services sector, mandating adherence to a multitude of regulatory standards and corresponding requirements. Interestingly, for certain ITPMs, the imperative to comply with regulatory requisites facilitated their transition to agile ISD practices. The advantages of embracing agile ISD practices, particularly in terms of project duration and cost management, were readily discernible to the ITPMs. Such projects fostered an environment conducive to the development of a new ITPM role identity, thereby aiding ITPMs in engaging in identity-forming activities. In particular, they found that the inherent flexibility of agile methods was advantageous when contending with continuously evolving and intricate regulatory demands. In contrast, the traditional waterfall methodology would have faced significant delays due to numerous change requests and budgetary adjustments.

The challenges in the project revolved around the fact that the regulator defined the “What” but not the “How”. This meant that it had to be continuously verified through inquiries what the regulator actually wanted. As a result, the process underwent a progressively detailed and changing set of requirements. In this context, agile methods proved advantageous compared to the traditional waterfall methodology because individual components could be easily modified. […] In the classic waterfall model, execution would have been significantly slowed down by numerous change requests and new budget applications.

(ITPM 12)

Second, on the individual side, some ITPMs mentioned having previously adopted a transformational leadership style, distinct from a directive one, before the introduction of agile practices and the transition to hybrid project settings. They emphasised trust, decision-making support, and team empowerment in their leadership approach.

I don’t do Control and Command. I neither have the inclination for it nor have I found it particularly helpful in my career here so far. When I’ve spoken positively about my bosses, it’s always been because they trust me and support me in making decisions. That’s something I’ve always appreciated and it’s what motivates me. Nowadays, they call it empowerment for the team. I believe that, regardless of what it’s called now, I’ve always operated with that leadership style.

(ITPM 16)

These ITPMs tended to reshape their identity by developing a new understanding of their role and collaboration model with the team. They had already experienced relinquishing control and embracing a self-managing team in a traditional project setting. Consequently, despite facing identity tensions with the introduction of agile practices, they possessed the leadership skills to effectively manage and resolve these challenges by forming a new role identity.

Third, while POs and SMs often benefitted from formal training and certifications, many interviewees noted a lack of introductory training for those taking on the role of ITPMs in hybrid project settings. In the absence of formal training, some ITPMs resorted to drawing on prior experiences in agile projects or informal coaching from peers to enhance their comprehension and proficiency in agile methodologies (“agile literacy”). However, a substantial majority find themselves thrust into their roles without adequate preparation or support.

The qualification for [being an ITPM in an agile setting] is kind of so-so. Those who are trained for their new roles are the POs and SMs. They got quite comprehensive training. The rest not. […] They were really thrown into the deep end.

(ITPM 17)

Regarding identity work activities, our interviews revealed that ITPMs who sought to maintain their role identity within the new setting typically received minimal training for this liminal role. They appeared to gravitate towards familiar, tried-and-true practices in an attempt to create a sense of security because of their lower degree of agile literacy. Consequently, agile literacy had a significant impact on shaping coping strategies adopted by ITPMs.

Lastly, only one interviewee explicitly stated that their role was clearly defined as an agile project manager by their organisation. More commonly, ITPMs described their new role as lacking a clear designation (role ambiguity) or emerging only through the shared responsibilities they held with SMs and POs.

I would say the role is less clearly defined. It feels like you’re responsible for a bit of everything because all the issues that couldn’t be resolved end up on my plate and need to be addressed so that people can continue their work. It’s certainly more ambiguous than it used to be.

(ITPM 15)

These ITPMs encountered a range of identity tensions, which predominantly stemmed from the intricate allocation of responsibilities and duties among themselves, SMs, and POs. These tensions often revolved around questions of authority, decision-making power, and the delineation of roles within the agile project teams, necessitating careful navigation and negotiation to maintain effective collaboration. However, they did not articulate a distinct approach to managing these identity-related challenges, except for one individual who outright rejected any alterations to their identity.

In summary, Figure 5 synthesises the insights we gained through our study in a new theoretical model. This model depicts how ITPMs engage in different role identity work activities to navigate role identity tensions with the goal to form, actively maintain or reject an identity. The work activities demonstrate how ITPMs employ approach or avoidance response strategies to contend with the opposing demands in the liminal zone between conventional and agile work structures. Moreover, contextual factors significantly influence this dynamic, impacting both the intensity of tensions experienced by ITPMs and the nature of identity work deployed in response.

Figure 5. A theoretical model on navigating role identity tensions with role identity work activities.

5. Discussion

In light of the increasing number of organisations employing a hybrid approach to agile, our primary goal with this work was to understand the liminal position and identity tensions ITPMs face in hybrid ISD team settings, as well as how they navigate these tensions. Using a qualitative, inductive study, we uncover five core identity tensions that ITPMs face and reveal distinct role-related identity work activities in which they engage to resolve these tensions. Our results show how ITPMs experience role identity tensions as a result of conflicting demands from their agile ISD team and the surrounding traditional organisation. Considering such contradictory demands, the ITPMs’ situation is clearly tense, and their relationship with the agile ISD team members and top management is strained. Caught in the middle of two opposing forces of the identified tensions, ITPMs engage in various identity work activities. While each ITPM comes up with a distinct mix of potential solutions to their respective tensions, our results point to common themes shared across participants. Specifically, as illustrated in Figure 5, all ITPMs’ identity work activities serve one of three goals: forming a new role identity that adheres to the contradictory demands, actively maintaining their pre-existing role identity while mostly ignoring their changed environment or rejecting the new role entirely. To achieve these goals, ITPMs employ both behavioural and cognitive approaches. Navigating this intricate terrain necessitates an acknowledgement of the impact wielded by individual and organisational context variables. Distinctions in project typology, the levels of ambiguity inherent in roles, the depth of one’s agility aptitude, and divergences in prior leadership styles have all emerged as crucial determinants shaping the identity tensions experienced by ITPMs and the manner in which they navigate and respond to these challenges.

More broadly, we find that in hybrid ISD team settings, ITPMs are no longer authoritarian “conductors” who orchestrate project team members single-handedly. Instead, they are positioned outside – yet adjacent – to agile ISD teams and act as “boundary spanners” between the agile and the traditional parts of the organisation. However, they continue to be the project leader and contact person in the eyes of the top management. Consequently, ITPMs experience tensions during their interactions with top management and agile ISD teams. ITPMs are forced to navigate the blurred lines between their once-established role and their new, emerging role. In attempting to resolve these tensions, they engage in role-based identity work and reflect on their new role’s duties and responsibilities – battling the non-existence of a formal role definition in agile and their own organisation.

5.1. Theoretical Contributions

By revealing role identity tensions that ITPMs experience and examining identity work activities that ITPMs use to cope with these tensions, we go beyond mere descriptions of ITPMs in agile and hybrid project settings (Gandomani et al., 2020; Shastri et al., 2021). Rather, we reveal the underlying complexity in the role of ITPMs, who seek to navigate the competing demands of agile teams and traditional organisations. Based on this, we believe that our results offer three important contributions to research on agile ISD, hybrid project settings and digital transformation.

First, we uncover and systematise tensions that produce important role identity conflicts for ITPMs working in hybrid project settings. We identify five notable work-related role identity tensions. Two concern the communication and responsibility towards the top management, and three are related to the collaboration with, control of, and duties towards the agile ISD team. ITPMs are couched between the agile and non-agile governance structures of these stakeholders and need to position themselves in the charged relationships with both the top management and the agile ISD team. Our results advance previous research by offering a more profound understanding of the enormous challenges encountered by ITPMs and the detrimental consequences arising from the integration of agile ISD teams within traditional organisational structures. Specifically, our study illuminates the intricate terrain of role-based identity tensions experienced by ITPMs operating within the dynamic landscape of hybrid project settings. In doing so, we extend our comprehension of role-related challenges to encompass the distinctive realm of boundary-spanning roles, thus broadening our perspective beyond team-internal roles, such as developers, team leaders, or product owners (Venkatesh et al., 2018, 2020, 2023).

Second, we deepen our understanding of how ITPMs meet these challenges within hybrid project contexts by conceptualising the varied approaches employed by ITPMs in addressing role identity tensions. Our study identifies cognitive and behavioural identity work activities undertaken by ITPMs as they strive to form a new role identity, actively maintain the existing role identity, or reject any changes to it. These findings illustrate diverse response patterns among ITPMs facing agile ISD projects within traditional organisations. Specifically, we discern two distinct response strategies employed by ITPMs to manage their role identity tensions: approach strategies, where ITPMs actively embrace a new role identity through cognitive or behavioural means, and avoidance strategies, where they preserve their existing role identity or reject alterations. The insights gleaned from these patterns significantly contribute to our understanding of how ITPMs effectively navigate the multifaceted challenges inherent to their hybrid project roles, shedding light on the mechanics of individual-level boundary spanning. This adds to prior research that has primarily centred on “bridging structures” and managerial response patterns within the broader organisational context, often exemplified in bimodal IT settings (Haffke et al., 2017; Toutaoui et al., 2022).

Finally, our proposed theoretical model contributes to the advancing discourse on the role of ITPMs in today’s organisations striving to adopt agile ISD by articulating the dynamic relationship between ITPM role identity tensions, the resulting identity work activities, and the associated contextual factors. While ITPMs continue to play a vital part in agile ISD project success (Crowder & Friess, 2015; Tjørnehøj, 2019), previous agile ISD research has frequently marginalised the role of ITPMs, inadvertently obscuring the intricate tapestry of tension that lies beneath the surface of this pivotal role. Our model offers a fresh perspective, illuminating a previously overlooked reality: ITPMs are immersed in profound identity tensions in hybrid ISD settings, catalysing a transformative journey to reshape their professional identities. This transformation is intimately entwined with the unique contextual landscapes they navigate.

5.2. Practical Implications

Prior research has concluded that companies face major challenges when employing a hybrid approach (Cram & Marabelli, 2018; Digital.ai, 2021; Gemino et al., 2021). A key risk lies with the ITPM’s personal role conflicts, which might endanger the success of individual ISD projects on one level and the agile adoption process of the entire organisation on another level. Our findings provide helpful guidance for practitioners to overcome these hindrances.

First, we identify how companies can proactively design their agile ISD team settings to accommodate ITPMs and their new demands more adequately. Our research investigates both the tensions that ITPMs experience and the identity work activities they engage in, aiming to resolve these tensions. As a result, we provide recommended actions that organisations should take if they seek to be successful in their ISD efforts. These actions can be structured in two distinct tiers. Tier 1 describes the bare minimum measures companies should take: organisations should actively support ITPMs in their identity work activities. By successfully and sustainably resolving their tensions, ITPMs can focus their energy on the actual work activities that are expected from them. A company’s failure to assist their employees in their identity work can significantly and detrimentally impact their employees’ work performance simply because they are occupied with navigating their individual-level challenges. This connection is relevant not only for ITPMs but also for team members that closely collaborate with them. The ITPMs’ distractions from their core work may spread to other employees if they are not contained. Targeted communication training, for example, may aid ITPMs before they resort to leaving the company. If companies want to avoid losing long-tenured, skilled employees, we urge them to take action. Given how significant the ITPMs’ role in ISD teams is, helping companies avoid losing them will ultimately ensure uninterrupted ISD operations. Otherwise, significant delays and unsuccessful projects are highly probable.

Beyond tier 1, we recommend that companies aim to prevent tensions from the outset. Such active prevention strategies, which can be viewed as tier 2, include but are not limited to proactively measuring employee satisfaction and stress levels, for example, via recurring surveys. When employees disagree with survey statements like “My role (my duties, obligations, and responsibilities) are clear” or “My position within the team is well-defined”, employers are alerted about a potential role identity conflict early. This early detection allows companies to take appropriate countermeasures before their employees resort to sub-optimal responses.

Second, we confirm the crucial importance of organisational culture for the overall agile transformation process. Echoing findings from previous literature (Cram & Newell, 2016; Iivari & Iivari, 2011), our results show that an agile mindset within a company’s management is imperative for a successful shift from traditional to either fully agile or hybrid ISD settings. In addition to previous findings, however, we want to contrast this by noting that ITPMs’ efforts to be effective in this new ISD setting are not thwarted by a fundamental disagreement with agile principles. Instead, their tensions arise on an individual level from uncertain and vague expectations towards their new work role in day-to-day operations. Consequently, while we found evidence that an agile mindset is a clear prerequisite for agile success, it is insufficient to instil an awareness of agile principles in ITPMs. Instead, organisations should focus on individual-level duties, tasks and responsibilities to build a collective mindset from the ground up. While we studied this for ITPMs specifically, we suspect this is also true for other management roles (Dong & Götz, 2021).

5.3. Limitations & Future Research

Despite the contributions mentioned above, we note several limitations that provide valuable opportunities for future research. First, our data set is based exclusively on the unique perspective of individuals working as ITPMs and their own perception of agile ISD team settings. ISD teams, however, consist of a wide and diverse range of distinct roles with potentially greatly varying perspectives. While we are primarily focused on the individual-level consequences of role identity tensions, we aim to make conclusions about the impact of individual-level role identity tensions on the ISD team and the broader organisation. Consequently, the findings we report might best be complemented by additional studies. For example, future studies could seek to further clarify the direct connections between the ITPM identity tensions and the associated solutions. By employing multi-level models and mixed methods, including more ISD team members’ roles, we could gain a more holistic view of inter-personal and inter-organisational role tensions. These – in addition to the individual-level tensions studied in the present paper – play an important role in shaping project outcomes. To make sure that adding more ISD team roles to the research scope does not introduce any undesired and avoidable side-effects and biases, we should exercise additional caution during data collection. Ideally, we should verify that all interviewees not only occupy the specific professional position that we want to study but also personally identify with their work role. This could, for instance, be accomplished by conducting an adapted version of the 20 statement test (Kuhn & McPartland, 1954) with all interviewees.

A second limitation lies in our cross-sectional data collection. Interviewing ITPMs and performing a cross-case analysis enabled us to identify emerging patterns, differences and commonalities in regard to the participants’ work identity tensions. However, our research design is limited in that it is unable to provide longitudinal, within-person insights regarding the role identity changes that individual ITPMs may experience over time. Future research could extend our work by undertaking such an investigation. Especially if such a study was conducted during an organisation’s transition from traditional project management to either fully-agile or hybrid ISD, it might provide additional insights into individual role identity tensions and the activities required to resolve them. Future research might, for instance, uncover how ITPMs combine multiple identity work activities when adapting to their new role over longer periods. Another potential avenue for future research is to explore the determinants of the intensity of role identity tensions by assessing whether identified contextual factors, such as agile literacy, or additional unexplored factors like organisational tenure, influence the extent to which individuals experience these tensions. Altogether, it appears promising to supplement the presented work with longitudinal data to further understand how our interviewees’ role identity develops and changes over time.

Third, we want to explicitly recognise the constraints on the generalisability of our findings. Our interviews were centred on ITPMs within a specific context – an agile ISD project embedded in a traditional organisation. Consequently, it is essential to acknowledge that the applicability of our results to other settings may be limited. For example, the interviewees in our sample work in organisations that exclusively use Scrum as their agile framework of choice. Although Scrum is a widely used approach, we recognise that our findings may not generalise to other approaches, such as Extreme Programming (XP). Future research could investigate if other forms of agile support the same relationships with the ITPM and the traditional organisation as we found with Scrum. In addition, our sample is skewed towards larger companies, especially those in the financial services industry. While we believe this bias emerged naturally with the particular scenario we were looking for, we encourage researchers to also examine different contexts such as small- to medium-sized companies where, for example, top management is closer to the ITPMs and the teams, such that the tensions and the identity work might appear different.

Finally, while our primary focus was on theory development, we recognise that the theoretical model we have crafted in this study warrants empirical scrutiny for its validity and practical applicability. A crucial next phase thus involves gathering quantitative or supplementary qualitative data to validate and demonstrate the model’s utility for both researchers and practitioners. On this basis, more studies can explore the intricate consequences of identity tensions on vital aspects of the agile ISD process (e.g., team performance, individuals’ perceived work-related stress, turnover intentions). Although it appears reasonable to assume these consequences will mostly have adverse effects – and thus, the tensions should be avoided as much as possible – it is possible that identity tensions may also offer beneficial opportunities for agile ISD teams. Provided that they are resolved effectively and successfully, role identity tensions might ultimately lead to an improved team setting and a shared understanding of each team member’s role. For that to happen, it is imperative that each individual’s identity work activities are aligned with agile principles and do not negatively affect other agile ISD team members. To understand this, further research is needed, for which our work may provide a solid foundation. Furthermore, although we identified specific identity work activities of ITPMs, our findings do not shed light on their impact on ISD project success. Therefore, we recommend further research to explore how the different identity work activities of ITPMs influence the overall success of ISD projects, aiming to elucidate the underlying mechanisms and resultant outcomes on project performance and effectiveness. In addition, we strongly encourage future research on the contextual factors we identified. For instance, a more comprehensive examination of ITPMs’ agile literacy could unveil the thresholds at which ITPMs exhibit a greater inclination to embrace the new agile environment, as opposed to maintaining familiar traditional project management practices. This distinction is crucial, as reliance on traditional practices may pose challenges for the agile ISD team, potentially hindering project performance and team dynamics. Such an analysis could further yield valuable insights into the necessity of tailored training programmes designed specifically for ITPMs. Apart from the identified contextual factors, future research could explore additional potential factors that may impact ITPMs’ identity tensions or work activities, which were not captured in this study. For instance, investigating whether organisations intentionally position former ITPMs as boundary spanners in the liminal space between agile and non-agile parts of the organisation, or whether this occurs incidentally, could be crucial for understanding the specific tensions they experience and their responses to these tensions. Overall, delving into the contextual factors opens avenues for a deeper understanding of the inherent challenges and impediments encountered in a hybrid project setting, cutting across individual, project, and organisational dimensions.

6. Conclusion

Agile ISD teams operating in traditional, plan-based organisations can have unintended effects on ITPMs, who are often stuck between the diverging demands of these two contradictory governance structures. In this paper, we crafted a theoretical model that reveals how ITPMs navigate work-related role identity tensions through role-based identity work activities, while accounting for the relevance of contextual factors. We thereby contribute to an enhanced understanding of this important boundary actor in agile ISD and, more broadly, towards clarifying the individual-level challenges in agile ISD and how to cope with them. We hope our findings will open up avenues for further research into the boundary work of ITPMs and other stakeholders that are crucial for the success of agile ISD projects and digital transformation more broadly.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Authors elect to not share data due to privacy and confidentiality restriction.

Notes

1. We acknowledge the anonymous reviewer who encouraged us to further examine the importance of contextual factors.

Appendix

A.1. Details on the interviewees

Table

ID	Age	Work experience	Organisational tenure	Experience with agile	Industry
ITPM 1	45	10	8	7	Life Sciences
ITPM 2	50	29	20	7	Financial Services
ITPM 3	40	22	20	3	Financial Services
ITPM 4	45	15	10	7	Life Sciences
ITPM 5	50	30	25	5	Financial Services
ITPM 6	28	5	2	5	Life Sciences
ITPM 7	45	16	16	2	Financial Services
ITPM 8	33	5	1	5	Life Sciences
ITPM 9	33	10,5	7	7	Life Sciences
ITPM 10	28	13	7.5	5	Life Sciences
ITPM 11	30	5	4	5	Life Sciences
ITPM 12	50	25	25	2,5	Financial Services
ITPM 13	40	14	4	13	Financial Services
ITPM 14	30	5	4	2	Financial Services
ITPM 15	38	12	5	2	Financial Services
ITPM 16	40	20	20	4	Financial Services
ITPM 17	31	5,5	2	2	Financial Services
ITPM 18	43	17	15,5	4	Industrial Goods
ITPM 19	42	15	6	5	Building Services
ITPM 20	42	18	2	2	Financial Services
ITPM 21	47	24	4	4	Aviation
ITPM 22	49	24	4	6	Financial Services
ITPM 23	53	28	n.a., freelancer	5	Financial Services
ITPM 24	73	51	35	12	n.a.
ITPM 25	41	22	2	4–5	Financial Services
ITPM 26	51	>20	1	7	Financial Services

A.2 – Interview guideline

• Can you please briefly introduce yourself and elaborate on your professional resume?

• Can you briefly tell me about your methods in classic as well as in agile project management and explain them?

• What makes the special difference for you? What do you personally think of agile project management and why?

• How would you describe your role in an agile software development?

• What is your scope in an agile context? What tasks does the team itself take on? What tasks do you take on as a team leader?

• How has the change altered the scope of duties?

• Can you compare the role as project manager in classic project management with that in agile project management?

• What is the relationship between you and the Scrum Master?*

• What is the relationship between you and the Product Owner?*

• What is the relationship between you and the Scrum development team?*

• How does the agile team view you as the project manager of a “self-managing team”?

• Were there any misunderstandings between the roles?

  1. Can you describe them and how was that for you?

  2. Discuss conflicts

• From your perspective, what were the key challenges that arose? When and where did these occur? How were these challenges resolved?

• How relevant do you see the position of project manager in agile project management?

• Would you still like to mention and elaborate on important aspects that have fallen short in the interview so far?

* Questions added when adapting the interview guideline