Enabling street-level work: minimal structures for customized social services

ABSTRACT

This study asks how customization (the provision of non-standardized services) in street-level practice can be enabled. Based on 300h of ethnographic fieldwork, this study demonstrates that practitioners not only need discretionary room, but also social structures, professional knowledge, ‘customization templates’, and – somewhat contradictory – boundaries that limit the endless possibilities of customization. These structures or ‘enablements’ not only enable practitioners to develop novel solutions, but also create predictability, commonality, and consistency. The analysis also explains the role of first-line managers in this practice, which can be both legitimizing and obstructing. When structures lack, customization comes with serious risks for beneficiaries and street-level practitioners.

KEYWORDS:

• Street-level bureaucracy
• responsiveness
• management-by-enabling
• ethnography
• social service provision

Introduction

A street-level practitioner is behind her computer, working on a ‘family plan’ for a child with complex needs. She has discussed the child’s previous treatments, the continuing problems, and the treatments that are contracted in their municipality with a team member, concluding that none of those treatments fit the family’s combination of issues. The team member proposed a rare treatment offered by an uncontracted provider that is believed to be suitable for the combination of needs. The practitioner tries to figure out how to proceed. She asks another team member who walks in. That team member proposes to use a personal budget, if the family can manage that. Otherwise, a subcontract through another provider is suggested as an option. Neither of them know if this is feasible or even legal. The practitioner decides to call a manager for help. (observation 2 March 2017)

In the example, a street-level practitioner attempts to help a child and its family who are experiencing multiple problems. Existing treatments, which are offered by providers that are under contract of the municipality, are considered unsuitable. Therefore, an alternative, uncontracted treatment is proposed, which can be understood as a customized solution for this family. Being responsive to citizens’ needs, to customize services where necessary, is expected from the practitioners in the example and they have been granted discretionary room to do so. The example highlights the uncertain process the practitioner goes through to actually provide such a service to the family. It is an example of the difficulty of developing customized solutions.

The quest for responsiveness in street-level theory and practice is all but new (Brodkin 2012; Lipsky 2010; Zacka 2017). Yet, more recently, we see a development towards customized services, and enhanced responsiveness, which comprises the provision of non-standardized services to fit citizens’ problemacy (e.g. Raaphorst 2021; Visser and Kruyen 2021). The development has emerged across social services where it was argued to be particularly valuable for citizens facing a combination of problems (Fuertes and Lindsay 2016; Needham 2011). It is argued that the multi-layered and interconnected character of the needs of those citizens cannot be met by services that are pre-programmed, specified in precise detail, or neatly ordered and offered in different ‘domains’ (Evans 2020a).

The aim to customize services marks a shift in thinking about how street-level practitioners implement policy and make decisions (Needham 2011; Rutz and de Bont 2020). Street-level provision was traditionally – aimed to be – characterized by its predictable, repetitive, routine character to ensure equal treatment, accountability, and efficiency. Citizens were offered standardized services based on formal entitlement to those services based on corresponding criteria (Grootegoed and Van Dijk 2012). Now, in some policy domains, the focus has shifted to providing individualized (non-standardized) services, tailored to specific circumstances and cumulation of problems. This has tangible results for practitioners as customization challenges the standardized nature of their work. To achieve tailor-made solutions, it has often been argued to increase the discretionary room of street-level practitioners, to allow them to make decisions as they see fit (Brodkin 2012; Rice 2017). Yet, as the example shows, discretionary room alone is not enough to enable practitioners to develop customized services.

This study therefore asks how customization in street-level practice is enabled? The question developed in response to the surprising observation made during fieldwork that discretionary room was not enough for developing customized solutions. The presented answer is based on the analysis of 300 hours of ethnographic fieldwork among street-level practitioners and their superiors responsible for the local provision of child and family services in the Netherlands. I abductively combined the in-depth empirical analysis with theoretical insights, combining recent studies on enabling management in street-level bureaucracy literature (Brodkin 2011; Evans 2020a; Hupe and Buffat 2014; Raaphorst 2023) with organizational theories on the minimal structures of innovative practices (i.e. Kamoche and Cunha 2001; Weick 1998).

This study proposes to use the framework of ‘minimal structures’ (Kamoche and Cunha 2001; Weick 1998) to grasp how various elements enable customization. By doing this, this article shows that practitioners not only need discretionary room to accomplish customized decisions, but also social structures that support decision-making, specific professional knowledge that helps them in making sense of the case, templates that facilitate customized decision-making, and also – importantly – boundaries that limit the seemingly endless possibilities of customization. These structures not only enable practitioners to work differently and develop novel solutions, but also create much needed predictability, commonality, and consistency for practitioners and for citizens. When structuring lacks, customization might come with serious risks for citizens, but also for the street-level practitioners involved.

Through these insights, this article contributes to street-level theories of customization or individualization¹ in particular, and responsiveness more generally, by introducing and defining the framework of minimal structures. This framework helps us nuance the emphasis on discretionary room as enablement for tailor-made services and highlight the complexities of increased discretionary room. It also contributes to these literatures by showing the opportunities of customization, as well as its dark sides. Furthermore, this study contributes to the concept of enabling management (Brodkin 2011; Hupe and Buffat 2014; Lipsky 2010) by analysing the role of managers in enabling responsive service provision. Zooming in on specific micro-interactions, the findings show the conflictive contribution managers can make.

In the following section, I discuss the current literature on street-level discretion, enabling management, and minimal structuring of innovative practices. I then present the research context of child and family services in the Netherlands along with the study’s ethnographic research methods. The findings show the various structures that enable and constrain customization and work towards a tentative structure. Finally, I discuss the implications for theory and practice as well as their limitations and conclude by suggesting avenues for future research.

Theory

The issue of responsiveness has always been at the heart of street-level theories. In his ground-breaking work Street-level Bureaucracy, Lipsky’s wrote about the need for improvisation and responsiveness to citizens to achieve policy goals (Lipsky 2010). For decades, the need for responsiveness existed in a context that was highly regulated and scripted to ensure equal treatment of citizens. Rules and regulation were in place to prevent arbitrary decisions based on racial stereotypes, ideological preferences or other moral judgements (Harrits 2019; Maynard-Moody and Musheno 2003). Street-level discretion – the behaviour of street-level workers interpreting and balancing action prescriptions to reach judgement and take action (Evans and Hupe 2020) – was perceived unauthorized, as it created unwanted variation and differences between policy design and policy delivery and led to a lack of democratic control (Bannink, Six, and van Wijk 2015).

Yet more recently, in a variety of policy and geographical contexts, an extended form of responsiveness is foregrounded as a central policy aim, challenging and partly replacing the standardized and highly regulated nature of street-level work. Street-level practitioners are encouraged to create value for citizens, customize and differentiate services to fit citizens’ needs (Durose 2011; Needham 2011; Raaphorst 2021; Visser and Kruyen 2021). Based on the insight that families are increasingly faced with needs that are interconnected and multi-layered, it is argued that services cannot be pre-programmed or specified in precise detail. They are constituted in response to and sometimes in negotiation with the beneficiaries (Evans 2020a). When services are no longer – or less – stipulated and specified ex-ante, it will have consequences for the way street-level practitioners go about their work.

Many scholars have highlighted the potential of – increased – discretionary room to enable responsive or customized decision-making at the street-level (Brodkin 2012; Evans 2020a; Lipsky 2010; Rice 2017), also defined as institutionalized discretion (Rice 2017). In these theories, discretion is seen as the site where creativity can be used to make services work (Evans 2020a; Visser and Kruyen 2021). The idea of delegating discretionary room to practitioners with special expertise and proximity to the tasks fits the managerial approach of management-by-enabling, which has been presented as an alternative to management-by-compliance or management-by-incentives (Brodkin 2011). Inspired by the early work of Elmore (1979) and Lipsky (2010), management-by-enabling has been proposed to marshal discretion in behalf of greater quality and responsiveness in service provision. It prescribes policymakers and managers to make a realistic appraisal of funding and to make and reveal political choices about what counts and what does not (Brodkin 2011). Yet, – with some exceptions – it has received little practical expression.

Yet, discretionary room does not automatically mean that street-level practitioners exercise discretion. Uncertainty or a lack of guidance can make street-level practitioners fall back into old routines or onto pre-existing rules (Bernards et al. 2021; Raaphorst 2018; Rutz and de Bont 2020). Evans (2020b, 369) has argued in this regard: ‘being free from restrictions [or negative freedom] does not mean that you can act. There is also a positive dimension to freedom, which is having the opportunity and resources to act’.

A concept that might help to better understand the resources to act is enablements, defined as ‘the range of acts that enable bureaucrats to fulfil their tasks’ (Hupe and Buffat 2014). Raaphorst (2023) has further conceptualized enablements by foregrounding the ways in which practitioners experience enablement and their perspective of what constitutes enablement, a conceptualization I follow in this research. Here, we can learn from insights of Practice-theoretical approaches to street-level work, that have shown that practitioners do not just follow rules, or ‘the room therein’. Rules set constraints and suggest possibilities, but they cannot determine behaviour. In addition to rules, practitioners are also bound by and possibly change professional and practical knowledge and experience, norms, social interactions, and technologies, all the while dealing with specific time, budget, information of the situation and responding to the citizens’ particular request (Maynard‐Moody and Musheno 2012; Wagenaar 2020). Managers are part of this practice (Visser 2024), and can support responsible use of discretionary room by shaping tacit convictions and perceptions (Keulemans and Groeneveld 2020), activating professional knowledge and creating supportive communities (Møller and Howard Grøn 2023), and facilitating a psychologically safe team environment (Bernards, Schmidt, and Groeneveld 2023).

The concept of enablements helps us to understand how the exercise of discretion is intimately tied with structuring and demarcating street-level work. There is a risk that insufficient attention is paid to what needs to accompany institutionalized discretion and at other times, rules are withdrawn without substituting them for something else, creating an institutional void (Durose 2011; Rutz and de Bont 2020). In this regard, Evans (2020b) has argued: ‘There has to be a structure, but it needs to be loose and flexible to allow for innovation’.

To further develop our understanding of flexible structures and enablements for customized decision-making, I turn to organizational theories that have studied how to manage organizational innovative or improvisational activities (Kamoche and Cunha 2001; Weick 1998). Taking inspiration from jazz improvisation, studies have argued that organizations should not simply substitute planning and control for autonomy and discretion to cultivate novel solutions (Cunha, Miner, and Antonacopoulou 2015; Vera and Crossan 2004). Rather, seemingly paradoxically, structure, prediction and control need to be balanced and combined with autonomy and discretion. Weick (1998, 551) has argued that it is: ‘a mixture of the precomposed and the spontaneous, just as organizational action mixes together some proportion of control with innovation, exploitation with exploration, routine with nonroutine, automatic with controlled’.

The concept of minimal structures was developed to refer to such enabling structures (Kamoche and Cunha 2001; Morgan 1986; Weick 1998). The principle of minimum was defined as a reaction to bureaucratic principles in which organizational arrangements were defined as clearly and precise as possible and rules were added whenever new situations or hiatus appeared (Morgan 1986). Minimal referred to the need to have several structures that guide the process of decision-making instead of an endless, ever-growing number of predefined rules.

The concept of minimal structures offers us two interesting insights. First, it argues that innovative activities need a well-defined backdrop of rules and roles, a framework of commonality and certain routines to invent and coordinate ideas (Bastien and Hostager 1988). This way, a level of predictability and standardization can be created which both enables but also positively constrains discretion and learning to a certain extent (Kamoche and Cunha 2001). Theories that have focussed on minimal structures go beyond a dichotomous understanding of discretionary room versus structure by showing how discretionary room and rules can go together to enable innovative behaviour. Structures that enable this behaviour differ from traditional bureaucratic structures.

Second, the body of literature shows that what enables improvisation are ‘technical structures’ and ‘social structures’. Technical structures are composed, amongst other things, of rules and knowledge. In jazz improvisation, the social structures consist of relatively informal behavioural norms, specifying who has the authority to decide during improvisation, and how to communicate through codes during performances (Kamoche and Cunha 2001). Together, it is argued, these structures constrain invention, diminish uncertainty, and enable consistency yielding an apparent structured (and sonically appealing) result. These insights have largely guided the analysis of this study.

Research setting, methods, and data

To explore how customization in street-level practice is enabled, I have conducted ethnographic research. This approach is well-suited to illuminate real-time work activities and their relationship with the organizing process (Ybema et al. 2009). I have conducted fieldwork in a local child welfare organization as well as a medium-sized municipality in the Netherlands. Since major reforms in 2015, one of the policy priorities of child and family services is the provision of customized services, making this a suitable case.

The aim to customize services has changed the way street-level work is organized. Previously, decisions on the allocation of services were made based on an external assessment of the case, executed by a separate national agency. This led to a ‘care indication’, a formal and fixed diagnosis based on very detailed requirements. The indication in turn created a citizen’s entitlement to a specific type of treatment. The work of the child and family practitioners had thus been characterized by being dependent on an external assessor, leading to a pre-defined outcome that was rather fixed in nature.

The national reforms had stipulated the dismantlement of the formal assessment procedure that determined which form of care a child was entitled to, as well as scrapping the care indication as a requirement to allocate services. This granted much autonomy to local authorities to organize the allocation of child and family services. This autonomy was confined to ‘the voluntary domain’, where children and their families have the choice to participate in services. In the mandatory domain, practitioners were still bound by stricter allocation procedures, for example judge-mandated outplacements of children or child probation services.

At the local level, municipalities dealt in a variety of ways with this decentralized responsibility to organize street-level work. Some municipalities decided to grant the final decision-making authority to street-level managers who always had to check these decisions. Others kept extensive policy rules in place, but referred to hardship clauses to create the possibility for customization. Therewith, in essence, delegating the responsibility to develop customized services to the practitioners. In the case under study, however, discretionary room was further institutionalized at the local level.

To provide formal access and to deliver primary educational services, the municipality commissioned and subsidized a (semi-)private non-profit organization, the CJG (Dutch acronym for Centre for Child and Family). Allocation decisions (and corresponding expenditures) could be made without the approval of the municipality. The CJG had entrusted this responsibility to its – relatively – autonomous teams of street-level child and family practitioners (called practitioners henceforth).

Within the CJG, approximately 60 practitioners are responsible for judging requests from either parents or other providers and allocate them fitting services. The practitioners work in teams, which function autonomously. Within the CJG there are seven teams of practitioners, usually comprised of three to nine coaches. The practitioners have a professional degree, most in pedagogical science or social care. Some have a university degree in psychology or pedagogical sciences. Six managers support these practitioners. In addition, they give account to the municipality and develop policies collaboratively.

Data were collected during approximately 300 hours of observation between December 2016 and December 2017. Sixteen practitioners were observed during the course of their regular work day (Spradley 1997). I also followed them in their encounters with managers or policymakers. The observations were supplemented with dozens of daily informal conversations with those involved to clarify their actions. During observation and informal conversation, detailed fieldnotes were made and speech quotations written down as much as possible (Emerson, Fretz, and Shaw 1995). As most work was desk work, most notes were directly made on a laptop. The notes were later further elaborated and for the purpose of this article some were translated into English. Where available, I gathered and analysed documents that either informed these interactions, such as new policy plans, and that performed interactions digitally, such as emails.

In the same period, I conducted semi-structured interviews (all between one and two hours) with practitioners, their managers, and municipal policymakers. Fifteen interviews were conducted with practitioners, focussing on the complexities of customized service provision. Furthermore, 27 interviews were conducted with managers and policymakers about the local room for customization and their response to actual activities thereof. Interviews were recorded and transcribed verbatim.

The analysis of the data was done abductively, in an iterative cycle between data and theory. In this process, I engaged in multiple reflection sessions with peers to discuss, question, aggregate or rename codes and categories, their cues and definitions, leading to refined analysis. The first step of the analysis was to order the data into the performances of customization – including detailed narratives of sets of actions and related explanations. In a second step, I started a round of ‘open’ coding to identify elements that supported the exercise of customization and elements that hampered it, using mostly informant-centric terms and codes. This resulted in codes such as ‘wrong person for discussion’, ‘knowledge of social map’, and ‘reversal of decision’. Subsequently, I explored theories mostly in organizational sciences about enabling creativity and novelty and found the framework of minimal structures. Several rounds of coding led to the grouping and regrouping of codes into larger themes such as ‘routines for deliberation’, ‘customization template’, and ‘role of managers’. Finally, I categorized the codes in terms of different types of minimal structures. However, after a first round of reviews, I found these types to be too confining, using the theories of minimal structures more loosely and returning to the previously identified themes.

Findings

The findings include four sections. I start to explain the various measures through which discretion is institutionalized in the case under study and how this is experienced by the practitioners. I continue to unpack how four structures enable the development of customized services. This is followed by analysis of structures that are missing or under development and the risks that unfold. Finally, I work towards a tentative structure to enable the customization of social services.

Customization and street-level discretion

The aim to customize services within the CJG, and more generally within child and family services in the Netherlands, entailed a rather extensive understanding of working responsively. Street-level practitioners were encouraged to develop non-standard solutions. This meant that in cases where standardized services were expected to be insufficient to help the child or family in need, practitioners were expected to develop alternative or new solutions. In practice, this resulted in four forms of customized services. First, practitioners combined services from different providers. Second, they allocated families to service providers that were not under contract of the municipality. For example, providers that offered a very specific or specialized service. Third, practitioners developed or offered new services that were not yet offered by other providers. Or fourth, practitioners broke rules to allocate services for families, for example when these families did not meet specific criteria and therefore would not have received help.

As explained in the methods section, a combination of formal, legal measures at the national and local level had granted practitioners discretionary room to develop customized decisions. The practitioners often referred to the importance of this granted discretion, both in conversation with me, and among each other. One practitioner exclaimed: ‘It is the world upside-down. Whereas before, I had to almost get on my knees and beg [for a care indication], now I can decide myself’ (interview IV). After years of being confined by bureaucratic procedures and external assessors, practitioners were content in making their own judgements in cases and helping the families in ways they saw fit: ‘You can sometimes just decide something. When you think; well, this is best for the parents, I’ll allocate this’ (interview V). The mandate was perceived as empowering and gave a feeling of ‘being taken seriously’ (interview I). As a result, the possibilities for practitioners were – seemingly – endless, both in terms of the decisions they could make and the process they could follow to get there.

Enabling structures, enabled practitioners

Within this newly institutionalized and organized room, practitioners started to develop customized solutions. These were ongoing, social, at times subtle or swift processes that would sometimes take up to weeks or even months. The solutions were open to re-assessment or adjustment when circumstances changed. Based on observations of these processes and on interviews, in this section we can start to discern what helped practitioners to customize services. The granted discretionary room was very important, but in itself it was not enough to develop the solutions. The analysis shows how professional knowledge, multidisciplinary teamwork, customization templates, and a supportive role by the managers enable the practitioners in different ways. To identify these structures, I have analysed cases where practitioners have been able to accomplish customized solutions. Below, I explain the four enablements, illustrated with examples from the fieldnotes and interview quotes. Although they are presented separately, the structures were mobilized in combination. The practitioners almost always drew on the multidisciplinary team, and mostly also on their professional knowledge. Less frequently they used customization templates, although in the cases they used the template, it did play an important role. Support by managers played more on the background, as in many cases, practitioners did not directly involve managers in the decisions, but did they did feel they were trusted to make the decisions.

Professional knowledge, including specialized, and abstract scientific knowledge was mobilized for customization practices. In this specific case of child and family services, pedagogical theories about child development and theories about treatments helped the practitioners to define possible solutions beyond existing contracted services. Furthermore, professional guidelines provided a structure to identify and really understand families’ needs. Throughout their everyday work, CJG practitioners mobilized this professional knowledge. For example, when searching fitting solutions for a child with autism, the practitioners discussed the latest insights on autism (observation 23 January 2017). When a practitioner ‘got stuck’ with a family and was not making any progress, among peers she discussed a new treatment that could help (but was not yet offered by any contracted providers) (observation 27 January 2017). The organization and national policy stimulated the professionalization of the practitioners. Practitioners were called ‘T-generalists’, referring to a broad base of general knowledge that was shared amongst the practitioners (the horizontal line of the T), and specific specialist expertise that individual practitioners developed and kept up-to-date (the vertical line of the T). This professional knowledge helped the practitioners to make sense of cases given the lack of structure that resulted from the lack of eligibility criteria. It also enabled them to identify possible treatments beyond standardized, contracted services.

Organizing the practitioners in multidisciplinary teams created and structured the deliberation that preceded developing customized services at CJG. The practitioners, informally and ad hoc, or more formal and planned, discussed cases with team members. Deliberation with peers helped practitioners to talk through different aspects of the case, develop a somewhat coherent and complete understanding of the case and make sense whether a customized solution was necessary. In addition, discussing cases with others in their team, specifically those with other expertise, helped practitioners to acquire additional perspectives that helped them to explore and develop alternative solutions. Lastly, customization challenged the practitioners to make far-reaching decisions, both in terms of treatments for the families and in terms of budget and legal contracts, often without clear rules or even breaking rules. Working in a team of supported practitioners to deal with the increased uncertainties as a result of developing customized solutions. The different purposes can be read in a practitioner’s comment: I always consult with someone from the team, or with a superior when I think it is necessary; is this the right move, is it responsible, does it fit the trajectory? I just want my decision to be substantiated, to be correct. Of course, you never know if it will work, but you should at least have properly thought about it (interview II). The practitioners made active use of the team structure in their everyday work in which the collective sense-making, the exchange of additional perspectives, and the emotional support enabled the practitioners to make decisions within the wide range of possibilities that the aim of customization had created.

Developing customized solutions was further facilitated by what I propose to call customization templates, inspired by the templates in jazz that provide an existing pattern upon which to improvise (Kamoche and Cunha 2001). The templates structured, sometimes created a mandate, and in a way simplified the development of customized solutions. At the CJG, there were various customization templates. For example, the CJG had developed a mobile application for an overview of contracted providers and services. Through this app, providers could easily communicate waiting lists, the status of annual budgets, new services on offer, or new agreements with subcontractors. Practitioners used this knowledge of ‘the social map’, including their own or peers’ experiences with them, to define whether an existing service would fit the families’ needs or an alternative, customized solution was necessary. Furthermore, the CJG developed a ‘light’ contract document. This provided a format to draw up contracts with uncontracted providers for customized services.

A final enablement of customization was the supportive role of managers for the development of customized solutions. Managers often expressed their trust in the practitioners using their discretionary room to make decisions. A manager explained that the guiding principle was that practitioners: ‘should decide what is best, based on their expertise’ (interview XXIII). This was communicated in interaction between managers and policymakers, for example in policy meetings in which managers and practitioners met: ‘The style remains the same; professionals know best’ (observation 22 November 2017). A manager further explains the managerial role in propagating this principle: ‘When [practitioners] worry, they sometimes ask us what they need to do. On such ‘bad moments’, they long back to the strict assessment procedures. But we keep them on track: you are making decisions as a professional, based on your professional expertise. You can do this’ (interview XXIX). Often when practitioners proposed customized solutions, their judgement was not questioned. In an interview, a practitioner was asked whether she had to give ample explanation or account when she contacted the management about a customized solution: ‘No, I told the story, I knew the child protection officer who requested the allocation, and I knew the provider because I have sent children there before. So, it was not necessary at that time no. So, I did not have to give more account for that no’. (interview VI). This support was often felt by the practitioners and perceived as enabling: ‘I like to solve issues or tensions in our work. It might not always be the right solution, but I feel the organization values our efforts’ (interview I). The practitioners felt they had the space to make decisions, they felt taken seriously, which stimulated their sense of ownership and the trust in their own capabilities.

Missing or developing structures and risky business

Notwithstanding the institutionalized discretionary room and the cases where practitioners developed customized solutions, there were also cases where practitioners did not succeed to accomplish customization in cases where they did feel it was necessary. The practitioners explained that they did not always ‘dare’ to make decisions (observation 21 February 2017), that they fear unequal treatment (interview VII), and that they could make the decision, but not bear the responsibility (interview V). Analysis shows that the development of customized solutions sometimes failed, either because practitioners felt they lacked a clear mandate, because they worried about unequal treatment, or because there were too many options and their work and possible solutions needed to be limited. Looking closer at the organization of their work, we can start to discern certain structures that were under development at the CJG and structures that were missing. These missing or underdeveloped structures in some way or another inhibited or frustrated the process of developing customized services, a process that came with risks.

The first enabling structure that was set up, but not completely developed, was routine for deliberation and escalation. Beyond their own team, the practitioners also deliberated with peers from other teams, specialists, managers or policymakers to develop customized solutions and make the – sometimes difficult – decisions to allocate uncontracted services. Furthermore, some cases formally required a consultation with an internal coach or expertfor example, when there was a suspicion of violence in the family. In others, practitioners preferred to discuss the case with either an expert, a manager, or a policymakerfor example, when decisions involved high costs, conflicts with parents or professionals from other organizations, or very complex problemacy. The CJG had developed various routes that defined whom to talk to and how in these cases. The developed routine for deliberation and escalation was vital to develop customized services. Yet, it was sometimes unclear or incomplete. In these cases, practitioners spend much time finding the right persons that could help them with their case. The following fieldnote excerpt exemplifies this.

A practitioner is sitting behind her desk, working on the allocation of an uncontracted services for a child (a customized solution). A team member has suggested a solution for contracting, but the practitioner is not sure how to proceed. She starts contacting people to find out; not the municipality, she never calls the municipality she says. She first calls the team member who had suggested it in the first place. The team member replies that she doesn’t know how to do it either. Then the practitioner contacts a colleague with much knowledge of contracting, but that colleague does not answer the phone. Then, she calls a colleague from another team who has been involved in the case. That colleague answers the phone, but isn’t sure either. The practitioner looks at me and says she wants to do it right. She decides to call the service provider that she wants to contract in the hopes they will know the contracting rules of the municipality. (observation March 14, 2017)

As a result, the process was slowed down or even inhibited. Practitioners felt frustrated, sometimes desperate during their search and sometimes received contradictory advice from different people that they had contacted along the way. This increased their already large workload. It also had consequences for families as they had to wait longer to be allocated services or received contradictory information.

A second structure that remained unclear at CJG was a division and demarcation of responsibility and accountability. The CJG attempted to instil collective responsibility for decisions. Results from monitoring were discussed on team level to stimulate team learning, rather than scrutiny. In case of formal complaints against decisions, managers prepared and attended the appeal session to take the heat of the practitioners. Yet, practitioners felt uncertain, partly because specific checks and balances were missing. For example, a practitioner missed someone checking complex calculations she made for a personal budget:

I am responsible for my own allocations, even for personal budgets, in the sense of the one I wrote there uh … It is my decision and if I make a calculation mistake, I have done it too. The decision is not checked […] I have indicated this when we just started [with this way of working] by saying that I think that is a major responsibility. Because a comma wrong or a calculation error made, hey a zero too much or too little, has major consequences. (interview V)

The uncertainty about individual responsibility was partly instigated by superiors, who tended to blame individuals for not achieving sufficient customization of services, as illustrated by a comment made by one of the policymakers: ‘It is about customization. That is sometimes very difficult for the professionals. They already pity the child and then they must be able to cope with the pressure of the parent’ (interview XXII). As a result of perceived individual responsibility, practitioners were at times hesitant to make a customized decision. The consequences of possible mistakes loomed strongly over their heads. The practitioners also feared serious financial consequences for their organizations in case of wrongful allocations.

The practitioners were missing a clear definition and demarcation of practitioners’ tasks and priorities. The practitioners sometimes felt there were too many options to choose from and they needed some shared boundaries. In addition, they were still short in time and resources, making it necessary to make choices. Yet, at the CJG, the task of helping citizens with a solution that fits their needs at times became so indefinite that practitioners did not have the time or shared priorities to do so. The overall motto of the CJG was: ‘We don’t say no’, meaning the organization’s goal is to help every family that requests help in one way or another. Several of the practitioners’ explained that they doubted whether they had the capacity to do so: ‘What I notice more and more […] that we as CJG are actually very – maybe too – broad, we accept almost all types of cases and issues’ (interview I). And: ‘So, for us it sometimes means that we are ‘searching’. Because yesterday one of the coaches said; we accept everything as CJG! […] Then you can ask yourself, yes, so we can do everything, but is that really true? Can we … ?’ (interview III).

The lack of clear priorities and a demarcation of their work sometimes led practitioners to define boundaries for their tasks themselves, for example by deciding not to perform certain tasks or by interpreting goals in a way as to transfer tasks to other providers. Furthermore, in response to the unclarity, practitioners sometimes improvised in a resistive manner. With the argument that policymakers or managers had to take a stance on a certain topic, the CJG practitioners gave families what they requested, but for a short period of time (several observations January – May 2017). This way, they could answer the request within a reasonable timeframe, but still put pressure on their superiors to provide the necessary structures. A practitioner explained: ‘So the boy, I allocated extended services for him for three months. We do not leave clients in the cold so we allocate it anyway. But this is not desirable, we need to create clarity on this’ (interview IV). Although practitioners do focus on still providing services in these cases, an allocation for a shorter period of time is detrimental for the families, for whom the continuous contact with and need to request services is a burden.

Finally, something that thwarted the development of customized decisions was the interference of the managers and municipal actors in individual cases. At times, managers and policymakers reversed decisions made by practitioners. These actions undermined the supportive role by managers and further blurred the unclear division of responsibility. A policymaker explains: I occasionally check an allocation, […] and sometimes I correct it. The other day I called [name] to let her know that I did not agree with her assessment, that the service was too expensive, that she made the wrong assessment of the situation as the girl to which services were allocated was likely to improve due to the treatment (interview XXII).

For the practitioners, reversals of customized decisions felt like a challenge to their professional autonomy. This in turn impacted the practitioners’ capacity and willingness to make customized decisions. For example, a practitioner commented a manager’s critique to a team member by stating: ‘She does not know the case. She has not assessed the case like I did’ (observation 21 December 2016). The fact that decisions could be challenged created a fear of making mistakes. Occasionally, specifically after incidents in which confrontation had occurred, practitioners did not feel confident in making decisions on customized services themselves and expressed worries on the consequences of ‘doing it wrong’. Finally, it created frustration among the practitioners. A practitioner explained: ‘But if you then start reversing our decisions, I get annoyed’ (interview V). These feelings can result in practitioners’ inclination to not use their discretionary room, instead increasingly requesting confirmation and permission from superiors.

Towards a tentative structure

The experiences and perceptions of CJG demonstrate the difficulty of developing customized services, even if discretionary room is granted. Institutionalized discretion empowers the practitioners, but also creates the perception that there are too many possibilities and too little structure, that the weight of responsibility is too heavy, and they fear a lack of consistency and common ground. Seemingly contradictory, the practitioners ask for discretionary room, but they also ask for structure and clear guidelines. At the same time, we see an organization that is committed to professionalization and developing professional knowledge, independent teams, and deliberation and teamwork, but that is also reluctant to provide more structure out of fear to return to pre-defined rules and procedures.

Based on these observations, also of the risks and ‘failed’ attempts to customize services, we can start to identify a structure of eight enablements for making customized decisions: institutionalized discretionary room, social structures including multidisciplinary teams, routines for deliberation and escalation, and supportive role of the management, and more ‘technical’ structures (Kamoche and Cunha 2001) such as professional knowledge, customization templates, the division and demarcation of responsibility and accountability, and the definition and demarcation of practitioners’ tasks and priorities.

Social structures entail multidisciplinary teams, deliberative routines, and supportive management. These enable customization by defining who helps to make the decision and when (Rutz et al. 2017; Visser and van Hulst 2024). Social structuring helps to include alternative perspectives and see alternative possibilities (Rutz et al. 2017; Visser and Kruyen 2021), it helps workers to make far-reaching decisions and acquire a mandate (Visser and van Hulst 2024), it provides emotional support (Møller 2021) and it creates a kind of horizontal accountability mechanism to filter judgements (Zacka 2017). At the same time, and not unimportant in the perception of practitioners is the fact that social practices can increase the consistency (Rutz et al. 2017).

Technical structures revolve on the one hand around mobilizing knowledge. Confirming insights from Evans (2020b) and Møller (2022), we see how professional knowledge, which is both specialized and abstract scientific, is used as a resource to make sound decisions. For customization, practitioners mobilize knowledge to make sense of a case or to define alternative services. The knowledge enables them to define what the problem is and how the practitioners can help. It creates a common ground on which the decisions are built. Customization templates support improvisation, in this case choosing a provider of defining a contractual arrangement. In jazz, a song, chorus or riff provides particular patterns to base improvisation on. Similarly, here, an app or contract format structures, but does not confine discretionary room, as it still left options open and practitioners also continued to use those other options. The templates can be understood as a form of easily accessible knowledge, purposefully made explicit. The base of professional knowledge of child and family services is extensive and is continuously developed, as is the knowledge-base of jazz musicians (Kamoche and Cunha 2001). An important difference in the case of social service provision is that whereas the knowledge base in jazz improvisation is mostly tacit, in this study we see knowledge being discussed and mobilized explicitly (Møller 2022).

On the other hand, technical structures also define appropriate levels of responsibilities and priorities. Practitioners express a need for boundaries to confine the endless possibilities and therewith endless amount of work when customizing services. Boundaries define what practitioners do not have to do which enables them to focus on the remaining tasks and possibilities. Technical structures create the necessary certainty to make decisions, while leaving wide zones of manoeuvre for the process and the outcome of decision-making (Kamoche and Cunha 2001).

Discussion and conclusion

The recent developments to create value for citizens and customize and differentiate services raise the question how customization can be enabled. Based on 300 hours of participant observation of the provision of child and family services, this study unpacks the minimal structures that enable street-level practitioners to provide customized services. This study has shown that practitioners use and feel empowered by granted discretionary room that gives them the necessary flexibility to help families in need. Yet importantly, they also draw on a variety of structures that help them in the process of developing customized services, an at times lengthy process that is characterized by a large number of possible solutions and steps to get there. Minimal structures help shape and confine the many possibilities, explore alternative solutions, deal with uncertainty, create common ground and consistency, help prioritize and share responsibility and accountability. With enabling structures in place, practitioners customize upon professional knowledge and customization templates, they mobilize and are supported by social structures such as multidisciplinary teams, routines for deliberation and escalation and a supportive role of managers, and are confined by clear tasks and priorities and a division of responsibility and accountability.

With these insights, this study contributes to recent insights on tailor-made services in street-level work (i.e. Rice 2017; Visser and Kruyen 2021), and to our understanding of responsiveness in street-level work more generally (i.e. Lipsky 2010; Zacka 2017), as it nuances our focus on the enabling role of discretionary room (Freier and Senghaas 2022) and further develops our knowledge of how responsive practices take shape and how they can be enabled (Bakkeli and Breit 2022; Fuertes and Lindsay 2016; Toerien et al. 2013). The concept and elaboration of minimal structures offers a framework to analyse enablements in this specific context of customized services. This study also offers an empirical analysis that unravels the complex relationship between freedom and control, which has often been presented as opposites (Evans and Hupe 2020). I show how discretionary room and a variety of guiding structures are layered. The freedom to innovate is enabled by defined roles, commonality and routines (Visser and van Hulst 2024). The structures create a degree of predictability and standardization for the practitioners, functioning as guiding principles for the process of decision-making, while the outcome remains open. The principles are not mindlessly followed by the practitioners, but structure their agency while they pragmatically rework them in new situations (Maynard‐Moody and Musheno 2012; Wagenaar 2020).

This study also contributes to existing theories on customized, responsive service provision by empirically substantiating possible problematic and negative consequences of the aim to customize services (Borghi and Van Berkel 2007; Needham 2011; Skjold and Grimastad Lundberg 2022). Practitioners experience problems and frustration related to the prioritization of tasks and maintenance of boundaries and a heavy burden of responsibility and accountability. When not properly structured, the development towards more responsive services can imply a shift of responsibility to street-level practitioners, possibly resulting in overwork or even burn-out. Possible risks for beneficiaries are related to the long processes of finding solutions when routines lack, to unequal treatment or even not getting what is needed because of disagreements between practitioners and superiors. Customized service provision with adverse working conditions thus seems unsustainable. The structures seek to address some of these issues but will in themselves not be a one-size solution to all of them. They require continuous attention and discussion.

Furthermore, this study contributes to theories of enabling management (Brodkin 2011; Lipsky 2010), by deepening our understanding of the role of managers in enabling responsive service provision. Our findings confirm that managers can support responsive decision-making and professional uses of discretion (Bernards, Schmidt, and Groeneveld 2023; Keulemans and Groeneveld 2020; Møller and Howard Grøn 2023), by explicitly confirming practitioners’ mandate and expressing trust in their decisions made. However, managers need to walk the talk and enact that trust through their everyday actions. Small managerial actions can undermine the – feeling of – professional autonomy and supportive atmosphere, possibly resulting in a reluctance among practitioners to exercise their discretion. There is thus for managers a fine line between legitimizing and hindering customized decision-making. Furthermore, the minimal structures are dynamic, challenged, and provisional (Kamoche and Cunha 2001). They might become unproductive, develop in response to new lessons, or new problems that occur and thus need constant attention. Managers need to be attentive to developments and play a role in the deliberation, coordination, and adjustment of the structures.

Limitations and suggestions for further research

This ethnographic case study of child and family services in the Netherlands has allowed observation of how practitioners customize services for families and the structures that helped them to do so. Through participant-observation over an extended period of time, combined with interviews and document analysis, I could explore the everyday dynamics at play. This study and its design, however, has a number of limitations. I conducted over 300 hours of observation, but all these observations occurred within one case, within a specific type of street-level work, and within specific local and national contexts. The findings can therefore be insightful for other cases, but their transferability depends on similarities and differences within other research settings (Lincoln and Guba 1985).

As the development towards customization of public services seems to increase across contexts, I believe the findings have relevance well beyond this case, yet how exactly they play a role in different settings should be subject to further research. I therefore encourage additional, comparative research in other local settings, other domains, and other countries to expand our knowledge on how to enable enhanced responsiveness. The framework of minimal structuring could serve as an analytical tool to further explore this.

In addition, further research is needed that more closely examines the various practices of street-level managers in a context of street-level customization and how they balance conflicting demands. Other research could zoom in on the interactions between street-level practitioners and managers when engaged in customized decision-making. By offering the framework of minimal structures, as well as an empirically grounded analysis of enablements and risks, this study provides a basis to further explore customized services for both research and practice.

Supplementary Material

Supplemental data for this article can be accessed at https://doi.org/10.1080/14719037.2024.2381067

Disclosure statement

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

Additional information

Notes on contributors

E. Lianne Visser

E. Lianne Visser works as an assistant professor at the Institute of Public Administration at Leiden University in the Netherlands. Her research explores customization in public service provision and governance. She uses ethnographic methods and focuses on actor’s everyday practices.

Notes

1. Including studies on similar developments such as personalization and pro-social rule-breaking (Bernards, Schmidt, and Groeneveld 2023; Borghi and Van Berkel 2007; Fuertes and Lindsay 2016; Rice 2017).

Appendix 1.

Declaration on transparency, research integrity and research ethics

This study involves human subjects. The study was planned and carried out in 2016. At that point, the home institution of the author did not require formal ethical approval of this type of research. This appendix provides information about how the study was carried out in order to increase transparency.

• I have followed general principles for human subjects’ research when planning and carrying out the project, such as national guidelines on research ethics as well as the normative standards of our research community.

• To reduce the risk of causing harm to research participants, I have thought carefully about how data was generated (the topics and formulations of questions in interviews, briefing and debriefing of interview participants, etc.), as well as how I could secure the confidentiality and anonymity of participants.

• All participants have been introduced to the research topic as well as how the data would be used and stored and have subsequently given their consent to participate. Participants have been informed that they could withdraw their consent at any point in time.

• The data has been stored on a secure network drive in compliance with the European Union’s General Data Protection Regulation (GDPR).

• All data that is presented in the article has been anonymized by changing the names of the participants and ensuring that no locations, organizational or other details are provided that would make participants identifiable.

• Participants did not receive monetary compensation for participating in the study.