The diffusion of police innovation: A case study of problem-oriented policing in England and Wales

Abstract

There is significant evidence demonstrating that when done well, problem-oriented policing is associated with meaningful reductions in crime and public safety concerns. And yet, history shows that the implementation and delivery of problem-oriented policing is challenging, and that police organisations have generally not adopted it and even when they try to it is often rejected over time. This article draws on the concept of ‘diffusion of innovation’ (Rogers, E. (2003) Diffusion of Innovations, 5th edn. New York: Free Press) to unpick aspects of the processes through which problem-oriented policing has been adopted or otherwise among police forces in England and Wales. This article shows how factors related to the nature of problem-oriented policing – notably its incompatibility with prevailing norms and values of the police service, its complexity and unobservability – have influenced its adoption. Implications are also discussed.

Background

The history of policing is filled with innovation. One such innovation is problem-oriented policing (POP), proposed by Herman Goldstein in 1979 as a framework for improving police effectiveness through paying greater attention to resolving the substantive community problems that fall within the police remit rather than responding to calls for service as they arrive. This article considers the processes through which problem-oriented policing (Goldstein, 1979, 1990) diffuses, or fails to diffuse, through police organisations.

Problem-oriented policing is a framework for improving police effectiveness. Goldstein was critical of the tendency of the police to deal with crime (and other calls for service) on an incident-by-incident basis, typically involving ‘standard’ police tactics such as rapid response and police patrols. He argued that such so-called ‘incident-driven policing’ produced only minimal short-term effects and did little to address the underlying conditions that give rise to substantive community problems. Problem-oriented policing calls for a change in traditional police work and emphasises the importance of understanding those factors that generate recurring problems and of working out how best to address them (Goldstein, 1979, 1990). Goldstein called for the adoption of a more systematic problem-oriented or problem-solving process involving the identification of police-relevant problems that are of concern to the community, analysing their sources in depth, developing tailored interventions that specifically target the factors thought to be driving selected problems and evaluating the impact of those interventions (Goldstein, 1979, 1990).

Problem-oriented policing has been a major international reform movement in policing (Weisburd and Braga, 2019). The approach is widely advocated and implemented. Successive reviews, case studies and experiments have shown that problem-oriented policing is an effective way of dealing with a wide range of crime and other issues dealt with by the police (Scott and Clarke, 2020; Hinkle et al., 2020). A recent Campbell Collaboration systematic review concluded that problem-oriented policing is associated with an overall average reduction in crime and disorder of 34% (Hinkle et al., 2020). UK police services were early adopters, with Surrey Police and the Metropolitan Police Service both experimenting with problem-oriented policing in 1982, Northumbria in 1991, Thames Valley in 1992 and West Yorkshire in 1994 (Leigh et al., 1996). Many other UK police forces have since followed. However, despite extensive evidence for and endorsement of problem-oriented policing, many studies have also found that its implementation has been highly challenging, and that commitment to the approach as an organisational strategy wanes over time (Goldstein, 2003; Knutsson and Clarke, 2006; Braga and Weisburd, 2019). Over more than 40 years then, the adoption of problem-oriented policing has been patchy, with periods of commitment and interest repeatedly followed by periods of decline. As Goldstein himself (2018: 3) put it:

I have grown accustomed to viewing successful efforts to implement POP – when carried out in all of its full dimensions – as episodic rather than systematic; as the results of relatively isolated cells of initiative, energy and competence. I view these pockets of achievement as exciting and pointing the way but sprinkled among a vast sea of police operations that remain traditional and familiar.

This article investigates the processes through which problem-oriented policing diffuses through police organisations, drawing on Rogers’ ideas about ‘diffusion of innovation’ (Rogers, 2003; Greenhalgh et al., 2004). This concept is widely used across disciplines to examine the spread (or otherwise) of innovations through social systems (Rogers, 2003). Our study takes place against the backdrop of the Problem Solving and Demand Reduction Programme (PSDRP) a £6.3 million national programme hosted by South Yorkshire Police and funded by the Home Office. The programme ran between 2017 and 2021 and aimed, among other things, to promote, facilitate and advance problem-oriented policing among police and partners in England and Wales. The programme aimed to create opportunities to reduce demand for police and partners by embedding a structured problem-oriented ethos and capability within police services (Sidebottom et al., 2020a, 2020b). The project consisted of nine interlinked workstreams that together sought to create and/or bolster the infrastructure, methodologies and resources to support a strategic shift to a more problem-oriented way of working. These workstreams included peer-support for police forces seeking to embed problem-oriented policing, the development of an online knowledge hub to facilitate experience and knowledge sharing, the creation of guidance documents on the implementation and delivery of problem-oriented policing (Sidebottom et al., 2020a, 2020b) and the re-establishment of the national Tilley Award for excellence in problem-oriented policing and its associated problem-oriented policing annual conference.

Understanding how diffusion occurs is important. As noted above, problem-oriented policing is widely found to be a highly effective approach to policing, despite noted challenges in its delivery. It has few detractors (Eck, 2019). Consequently, better understanding of the conditions for and mechanisms through which problem-oriented policing is more likely to diffuse across police agencies matters. Informed by previous research into the diffusion of innovations more generally, this article aims to explain how problem-oriented policing has or has not diffused through police organisations in England and Wales. In so doing we advance understanding of the meanings ascribed to innovation within police work and the structures and processes that shape whether innovations are adopted, rejected or abandoned.

The article is organised as follows. We briefly review the literature on Rogers’ concept of diffusion of innovation, including examples of its application to policing. Next, we describe the methods and data used in this study. The results then follow, organised according to three key themes in the diffusion of innovation literature: ‘compatibility’, ‘complexity’ and ‘observability’. The article ends with a discussion of the main findings and their implications for police reform more generally and problem-oriented policing more specifically.

The diffusion of innovation

Diffusion scholarship has examined the processes through which innovations spread through social systems. It has explored how diffusion is shaped by attributes of the innovation itself, the features of a social system and the rate or time over which innovations are adopted or rejected (Rogers, 2003). A key contribution comes from Everett Rogers’ synthesis of innovation research, which was conducted in the early 1960s and has been updated at intervals since (Rogers, 2003). Rogers (2003) describes a five-step process whereby an individual (or other decision-making unit) comes to adopt or reject an innovation. These steps are: (a) developing knowledge about how and why an innovation works; (b) being persuaded of the value of adopting an innovation; (c) deciding whether to adopt or reject the innovation; (d) putting the innovation into practice; and (e) looking for support to confirm the benefits of adopting the innovation over time. For Rogers (2003: 172), the process of innovation diffusion is akin to an information-seeking and information-processing activity, in which an individual is motivated to reduce uncertainty about the advantages and disadvantages of adopting something new. Individuals adopt or reject innovations differently and the process will take time with people adopting at different rates. Rejection is possible at any stage of the process. The characteristics of innovations, as perceived by individuals, help to explain their differential rate of adoption (Rogers, 2003). Innovations are more likely to be adopted where they: (a) are perceived to offer the adopter advantages; (b) are compatible with existing values, past experiences and the needs of potential adopters; (c) are simple and easy to understand; (d) can be adopted on a trial basis; and (e) when the results or outcomes of innovation adoption are readily observable. In addition, Rogers notes that, ‘the patterned arrangements of the units in a system’ (2003: 24), will affect innovation diffusion and the speed thereof. For example, the views of opinion leaders, who provide information and advice about an innovation, and the social norms regarding expected behaviours within that system will affect how it diffuses.

There is a small body of research that has sought to apply Rogers’ ideas to innovations in policing. There is a precedent of using the concept of diffusion of innovation as a framework to examine reforms in policing, most notably with community policing (Morabito, 2010; Gayadeen and Phillips, 2014), CompStat/crime mapping (Weisburd et al., 2003; Weisburd and Lum, 2005) and innovation in information technology (Skogan and Hartnett, 2005). These studies tend to consider reasons for innovation diffusion across police organisations, their rates of diffusion and reasons why innovations take hold or fail to do so. For example, in their study of the diffusion of information technology across police agencies in the United States, Skogan and Hartnett (2005) attributed the rapid growth of ICT systems to the active role played by so-called ‘evangelists’ representing the host department; because it was free to access; because it primarily empowered detectives (who enjoy a privileged position in policing); and because it did not challenge the traditional mission and organisation of participating agencies.

The current study

Data and methods

The findings reported here are taken from a larger process evaluation of the PSDRP (Sidebottom et al., 2020b). As part of the process evaluation, 26 semi-structured interviews were undertaken with: (a) 19 individuals purposefully selected for their involvement in conceiving, designing and/or delivering the PSDRP; and (b) an additional seven individuals who were identified through snowball sampling as people actively involved in problem-oriented policing in England and Wales, but not directly involved in the creation or delivery of the PSDRP. These key informants were approached via email by a member of the research team inviting them to participate in an interview. Once consent had been obtained interviews took place either face-to-face in office premises or by telephone and all were digitally recorded and fully transcribed. The interviews explored participants’ perspectives on and experiences of the development and implementation of the PSDRP. As well as providing information on the obstacles, enablers and outputs associated with the PSDRP, the interviews also provided a rich source of information on the diffusion of problem-oriented policing in England and Wales in the past and similarly challenges and opportunities regarding the diffusion of problem-oriented policing currently. As indicated above, these data are interrogated here through the lens of diffusion of innovation in an effort to unpick perceptions of the features of both problem-oriented policing and police organisations that seem to influence the adoption and rejection of problem-oriented policing. In what follows we organise participants’ reflections under three themes central to Rogers’ idea of innovation diffusion: ‘compatibility’, ‘complexity’ and ‘observability’ (Rogers, 2003; Greenhalgh et al., 2004).

Results

Compatibility

Research shows that the greater the compatibility between an innovation and the prevailing organisational landscape the greater the chances of its adoption (Rogers, 2003; Greenhalgh et al., 2004). In relation to problem-oriented policing, the accounts of participants interviewed here drew attention to several perceived incompatibilities, which were thought to act as a barrier to the adoption of problem-oriented policing. One such perceived incompatibility was cultural. Participants repeatedly described how the (perceived) core task for the police service has been (and continues to be) responding to calls for service and enforcing the criminal law, an operating model that is at odds with and may reduce the willingness to embrace problem-oriented policing (see also Leigh et al., 1998; Read and Tilley, 2000; Goldstein, 2003; Townsley et al., 2003; Braga, 2002). One participant explained that:

Some of the challenges, it’s a cultural change. I think it would be fair to say that organisationally a lot of police are comfortable in the idea of responding. You know, the police as an organisation, people often tend to describe them as very good at dealing with the here and now.

Changing organisational culture to enable the introduction of new ways of working is a formidable task. In the context of innovation diffusion, explicit support for an innovation from organisational leaders is widely seen as one important step in encouraging staff to break away from dominant ways of thinking and working and to embrace new methods (Greenhalgh et al., 2004). Similarly, the demonstrated attitudes to and behaviours of police leaders were seen by interview participants as fundamentally important in influencing whether problem-oriented policing takes hold in police organisations and whether it endures over time, as has been suggested in previous studies (see also Leigh et al., 1998; Read and Tilley, 2000; Scott, 2000, 2006; Goldstein, 2003). Indeed, in police forces in which problem-oriented policing is more apparent, several interviewees attributed this to senior leaders who provided a clear vision and set of expectations and were able thereby to engender cultural change. As one participant put it:

Leadership is the catalyst for that which needs to follow, and what needs to follow of course is structure and processes and a performance management regime that oxygenates the notion of problem-solving. But as with many things self-evidently, leadership is the thing that kicks that off. If you don’t believe in it, it won’t happen.

Those interviewed provided several examples of how police leaders have acted as a ‘catalyst’ for problem-oriented policing. For example, some spoke of leaders who have been able to ‘energise people, to encourage them to act in a particular way’. Others talked of leaders inspiring others to work in a more problem-oriented fashion. As this interviewee put it: ‘with [named senior leader] driving it, with the energy, and then having an underlying plan, that's basically as good as it gets because the whole organisation was being driven towards this fundamental goal with embedding problem-oriented policing’. However, such passion for problem-oriented policing among contemporary police leaders was considered uncommon. Instead, several participants noted that many police leaders embraced a different and more traditional vision of policing – oriented toward enforcement of the criminal law. One speculated that some chief constables think ‘[problem-oriented policing] It's all very fluffy. Problem-solving, what the hell's all that about? Doesn’t work for me. We’re going to do away with all that. We’re going to put teams together, we’re going to tackle the bad guys’. Negative attitudes towards such reform among leaders may lead to inadequate resourcing, a fragmentation of reform efforts and convey a sense to rank-and-file officers that police reform was either not important or it had already been completed (Chan, 2007). These messages help frame and define the meaning of reforms and they provide the cues and labels that rank-and-file officers could draw upon for making sense of organisational changes (Chan, 2007).

When supportive leaders move on, which they inevitably do, commitment to the vision of a problem-oriented organisation may shift. One participant suggested that at a senior level there were about ‘maybe five to ten’ police leaders across England and Wales who were strongly associated with problem-oriented policing but cautioned that ‘the issue is that those leaders are going to leave at some point in the next couple of years and it's just how important does the next leader think problem-solving is’. Another participant put it this way:

I know a lot of forces, when they get new leaders, a new leader obviously wants to put their stamp on something and say, ‘This is my vision for the organisation’, and the vision might be great but people might have already been doing really positive work that then either doesn’t continue or sort of falls by the wayside a little bit.

Participants described how trying to persuade senior officers that problem-oriented policing was ‘business for everybody […] was a real uphill struggle’. Changing the culture of police leaders was described by participants as difficult because ‘most of these people have got a long history of policing and that way of policing is all they’ve ever known’. Indeed, in relation to the operation of the PSDRP, this participant concluded that, ‘It is our biggest challenge as a programme team, and it was certainly my biggest challenge’.

That problem-oriented policing has not been prioritised by many officers was also explained in terms of aspects of the police role and the organisation of police work. Participants drew attention to the pressure of routine work, which made it difficult for police officers to find time for problem-oriented policing. As one participant told us: ‘I think we are competing in a very noisy environment’. Another said they thought that: ‘the issue with this discipline is that it needs constant support and we need to constantly reiterate that this is high on the agenda, and it's difficult to do that in a world of conflicting responsibilities’. Competing demands on time can more broadly be understood in terms of ‘a struggle between a lack of resources, and having to spread fewer officers wider’. Indeed, in discussing how senior officers may well see some value in problem-oriented policing they noted that they still have to make choices about what to prioritise in the context of limited resources. One participant told us that: ‘I think it's a case of they’re having to balance … having enough officers on the ground to cover the day to day calls, and giving officers that time that they need to properly problem solve’. This balance needs to be understood in the context of cuts to UK police budgets over the last ten years (Home Affairs Select Committee, 2018; Higgins, 2018). Budget cuts have led to police leaders making hard choices about what forms of police work to prioritise. The outcome has been to give precedence to response policing and downgrade neighbourhood policing (where attempts to introduce problem-solving have tended to concentrate) as an organisational strategy (see also Higgins, 2018). As one participant noted: ‘the austerity (is) making it much harder for forces to put resourcing to neighbourhood policing, I think that's where we started to see things being dismantled’. Because problem-oriented policing tends to be found in neighbourhood policing teams, the outcome in many forces suggests that ‘capacity within the organisation had been lost’. Indeed, participants spoke of cuts to police funding as a key reason why commitment to problem-oriented policing had waned in England and Wales over the past decade.

Complexity

Innovations that are perceived by key players to be simple to use are more easily adopted (Askarany, 2003; Rogers, 2003; Greenhalgh et al., 2004). Those interviewed, however, did not perceive the adoption of problem-oriented policing to be simple. Likewise, ‘if an innovation in an organizational setting has few response barriers that must be overcome, it will be assimilated more easily and so interventions to reduce the number and extent of such response barriers improve the chances of successful adoption’ (Greenhalgh et al., 2004: 596). The diffusion of innovations is seen as a highly organic process through which an organisation adapts to the innovation and the innovation is adapted to the organisation (Greenhalgh et al., 2004). This adaptation may not come easily (Greenhalgh et al., 2004). Indeed, consistent with previous research on problem-oriented policing (see Goldstein, 2003; Leigh et al., 1998; Read and Tilley, 2000; Scott, 2000, 2006) participants drew attention to the need for police organisations to adapt their processes in order for the approach to diffuse through them. Such adaptation included building an infrastructure (such as developing analytical capacity, knowledge and experience); incorporating it into police performance management systems and personal development and promotion processes; establishing systems that reward and recognise it; and creating effective methods of supervision and accountability. In this vein, participants drew attention to how problem-oriented policing was often perceived to be highly complex to deliver, and required a great deal of maintenance and attention over time. As one participant told us:

You have to have very strong implementation planning because it's like you’re delivering a project, like we said, good communication strategies, operational plans, operational delivery, all sorts of different things that take place, across the organisation, at different levels […] and those ducks have to be all in a row.

Participants consistently stressed that more needs to be done to achieve these conditions in order to embed and advance a problem-oriented approach. Participants shared the view ‘that some sort of concerted effort is needed nationally in order to re-energise problem-solving across England and Wales’ and to develop ‘a central capability to actually go out and implement strategy’ and to ‘to give us some more support and expertise around how could we do that’.

A major theme in participants' accounts was that there has been a lack of understanding of what thorough problem-oriented policing involves among officers and staff. Indeed, this was seen by participants as a primary reason why the approach has not diffused throughout police organisations (see also Goldstein, 1990; Read and Tilley, 2000; Williams 2003; Bullock et al., 2006). As one participant put it: ‘I would suggest the majority of forces still don’t understand it [problem-oriented policing] enough to be outside the level of just ticking a box’. For another participant: ‘What we were acutely aware of […] is that understanding of problem-solving, both the methodology of it and how to apply it in practice was hit and miss at best’. Participants similarly drew attention to how ‘Forces haven’t been providing much around problem-solving training for a long time’.

It was also suggested by participants that there has been inconsistency across police forces in respect to training in problem-oriented policing. One participant remarked: ‘It was a mixed landscape. It was a little bit of everything from everybody with no consistency and there was no real format to it’. And by extension: ‘When you’re trying to guide forces as to what training they have, what we wanted to make sure was that there were different levels of training depending on your role and responsibilities because I think that's quite important’. That said, training was not viewed to be enough on its own to support the diffusion of the approach:

Training does not mean that you are problem solvers. There is a lot more to it.

Instead, participants discussed how important it is to ensure that officers and staff understand the benefits of problem-oriented policing and that it is embedded in wider strategic plans. As one participant told us: ‘It's around really understanding, “Do they get this”, as in the wider crime prevention approach, and then do they understand the benefits of problem-solving and how you have to really pull together a strategic plan around how you implement it of which training is one part?’

The research literature on the diffusion of innovation has shown that the complexity of an innovation can be reduced by practical experience and demonstration (Greenhalgh et al., 2004). In respect of this, the identification of ‘champions’ – key individuals within social networks – has been identified as central to the diffusion of innovations within organisations (Greenhalgh et al., 2004; Rogers, 2003). Champions have been much evident in previous studies that have documented the adoption of problem-oriented policing (Laycock and Webb, 2003). It follows that the presence of knowledgeable individuals who could promote problem-oriented policing, generate interest and provide support, were seen to facilitate adoption. ‘Semi-professional advisors within forces’, as one participant put it, or ‘tactical advisors to give that advice and guidance and push things forward’. Such change agents clearly exist; however, their position can be insecure within the organisation as ‘people move posts far too quickly’.

Observability

Reviews have further demonstrated that where the benefits of an innovation can be immediately and easily observed, they are more likely to be adopted (Askarany, 2003; Rogers, 2003; Greenhalgh et al., 2004). As described in the introduction, there is now a very large body of research evidence, in the form of case studies, experiments and systematic reviews, which demonstrates that problem-oriented policing can be effective at resolving a wide range of crime and public safety issues. Research evidence of effectiveness is also complemented by a wide range of guidance documents and resources designed to assist problem-oriented ways of working. Moreover, much of this material can be easily located in a dedicated, practitioner-oriented online resource: the ASU Center for Problem-Oriented Policing (www.popcenter.org). Despite this, participants in this study repeatedly drew attention to how they believe the benefits of problem-oriented policing have not always been immediately observable to police personnel and, at the individual project or police force level, that it often takes time to implement problem-oriented policing and to achieve results. This was seen as one reason for low momentum in the uptake of problem-oriented policing, as well as one reason why it has often withered over time. One participant argued that: ‘forces get discouraged when they’re not seeing quick wins, the pressures are to bin it and do something else, and this [problem-solving] is a longie: it takes a long time’. That the benefits of problem-oriented policing were not immediately observable, led participants to draw attention to how it can be difficult for senior officers to take a ‘leap of faith’ and ‘give you that leeway’ and invest in problem-oriented policing. Following on from this, participants also noted that it can be difficult to justify the introduction (and resourcing) of problem-oriented policing because benefits may be hard to measure and quantify. As one participant stated:

You can’t measure a lot of it. You’re trying to find proxy ways of showing the value of doing something proactive rather than falling into the trap of, well, we’re just looking at how many arrests we’ve got or how many incidents we’ve got.

This aligns with Rogers’ comments on the difficulty of disseminating preventive innovations of the kind promoted in problem-oriented policing:

One type of innovation has a particularly slow rate of adoption because individuals have difficulty in perceiving its relative advantage. A preventive innovation is a new idea that an individual adopts now in order to lower the probability of some future unwanted event. The desired consequence is distant in time and so the relative advantage of a preventive innovation is a delayed reward. In contrast an incremental (that is, not-preventive) innovation provides a desired outcome in the near-term future. (Rogers, 2003: 234, italics in original)

Under the theme of observability, participants drew attention to the importance of being able to demonstrate the impact of problem-solving, and to communicate the value of adopting a problem-oriented approach. As one put it: ‘We knew that, in order for (problem-oriented policing) to run and be sustainable […] there needed to be some viable work that comes from it or some evidence to show senior officers that actually it makes a difference’. Participants described how being able to produce evidence on the impact of problem-oriented policing was especially important when considering its extension beyond neighbourhood (or community) policing. As one participant stated:

I think that's one of the tricks of the problem-solving and demand reduction work; we’ve got to find those hooks for people to realise that it is equally applicable to solving the CT [counter terrorism] issue or a serious organised crime issue, as it is to an anti-social behaviour issue on a problem estate.

Case studies were often seen as important in demonstrating the potential for problem-oriented policing: ‘if we did get some case studies, or examples where problem-solving had been used, as in preventative work in fraud and cyber, that would really help’. More generally, ensuring that officers saw the relevance of problem-oriented policing across the wide range of police activity was viewed by participants to be essential:

you’ve got to really try and negotiate your way around trying to make them interested and actually there's lots of things for their own department. So we tailor the training that we give them to their departments and how it can actually work for them.

Conclusion

We started this article with the observation that the history of policing is filled with innovation: original and disruptive ideas intended to bring about improvements in both the function and outcomes of policing. One such innovation is problem-oriented policing, first proposed by Herman Goldstein in 1979. It is clear that when done well, problem-oriented policing is associated with meaningful reductions in crime and public safety concerns. And yet, history shows that the implementation and delivery of problem-oriented policing has proven challenging. Serial weaknesses have been identified. As summarised by Sidebottom et al. (2020a: 4), ‘despite extensive evidence for and endorsement of POP, it has not become the modus operandi of British policing’.

This article has drawn on the concept of the diffusion of innovation to unpick aspects of the processes through which problem-oriented policing has been adopted or otherwise among police forces in England and Wales. According to Rogers, the diffusion of innovation requires increasing awareness of the innovation, decisions to adopt it, and its initial and sustained use throughout a social system (Rogers, 2003). However, the adoption of an innovation may be more akin to ‘an organic and often rather messy model of assimilation in which the organization moved back and forth between initiation, development, and implementation, variously punctuated by shocks, setbacks, and surprises’ (Greenhalgh et al., 2004: 601). Likewise, Rogers (2003) draws attention to how the process of innovation adoption is not always straightforward, and innovations may be discontinued at any time. Indeed, the adoption of an innovation can be long and complex (Rogers, 2003).

The starting point of this article was, as indicated above, that while problem-oriented policing has been an important reform movement in policing for the past 40 years, police organisations have generally not adopted it and even when they try to it is often rejected over time. This article focusses on how factors related to the nature of problem-oriented policing – notably its incompatibility with prevailing norms and values, its complexity and unobservability – have influenced its adoption. We have unpicked three major themes that have influenced whether problem-oriented policing diffuses though police organisations.

First, the level of compatibility between the innovation and dominant modes of police work. Problem-oriented policing emerged following a critique of the then prevailing approach to policing, which Goldstein viewed as generic, ineffective and too reliant on response and enforcement. To some extent these practices still (and will always) remain, and so problem-oriented policing stands in contrast in terms of style, but also in terms of what officers have traditionally valued (the pursuit of criminal suspects and enforcement of the criminal law with an emphasis on action and fast-time results). To underscore this point, where innovations in policing reinforce and support the traditional orientation of police work, such as CompStat or crime mapping, they may diffuse more easily (Weisburd et al., 2003; Skogan and Harnett, 2005). The adoption of problem-oriented policing – certainly outside neighbourhood (community) teams – has not been seen as supporting those core organisational needs. Instead, it represents a major change to the dominant approach to police work. In turn, problem-oriented policing has not been prioritised – especially in a context in which there are competing demands on time and limited resources.

Second, the degree of complexity associated with the innovation. Problem-oriented policing has been viewed as complex to deliver because it needs new structures to support its operation and requires knowledge and skills that officers do not routinely have. Indeed, problem-oriented policing might be described as an example of what Rogers (2003) terms a ‘radical innovation’ that represents a new paradigm and is characterised by a high degree of uncertainty. Although there is some evidence of police organisations adapting their infrastructure to accommodate and support problem-oriented policing, the infrastructure needed to support problem-oriented policing is not always present or is in its infancy, or commitment to sustaining it erodes over time.

Third, the degree to which the benefits of innovation are observable. Although there are clear benefits to adopting problem-oriented policing at the project level, they may not be readily observable at an organisational level. At the organisational level, implementing problem-oriented policing takes time and any overall benefits that accrue from its adoption may become apparent only over the long term. Compare this with arrest: although arresting an individual might not lead to a long-term outcome it at least produces a short-term immediate output. In turn, the ‘relative advantage’ – the extent to which an innovation is viewed as better than the mode of practice it replaced – of adopting problem-oriented policing as an organisational strategy has been insufficiently apparent because officers have not been persuaded of the value of adopting the approach. The relative invisibility of the pay-off from problem-oriented policing has both impeded its adoption in the first place and also helps to explain why it has frequently been abandoned over time.

Communication about an innovation is important if it is to diffuse through an organisation. The adoption or rejection of innovations is shaped by cyclical processes of generating knowledge about an innovation, being persuaded of its value and receiving positive confirmation of any decision to innovate over time (Rogers, 2003). For problem-oriented policing to be adopted, officers need to be aware of it, they need to know how to do it and they need to be persuaded that it will add value to their work. This may not be an easy task. We have seen that the patronage of key staff is important in this process of communication. The patronage of leaders (at different levels of the organisation) communicates important messages about the desired organisational direction and expectations of police personnel. It communicates that the innovation is valued and reinforces any decisions to engage with the innovation over time. Line managers play a similarly important role by reinforcing messages regarding the value of delivering problem-oriented policing. In this sense, managers ‘become institutional entrepreneurs—taking on the difficult task of changing taken-for-granted cultural beliefs and values to help ensure that the innovation is perceived as legitimate inside and outside of the organization’ (Willis and Mastrofski, 2011: 325). Likewise, the presence of highly committed ‘champions’ and figureheads who are able to effectively communicate the virtues of problem-oriented policing and inspire officers and staff to implement and sustain the approach facilitates implementation (see also Read and Tilley, 2000; Kirby and Reed, 2004; Bullock et al., 2006). The latter point resonates strongly with diffusion of innovation theory, which predicts that champions will be important in persuading would-be adopters of the value of innovating (Rogers, 2003). The provision of training and guidance, disseminating good practice and rewarding problem-oriented policing also likely play an important role in this process of communication. For all are important in raising knowledge, persuading officers of the value of an innovation and generating commitment.

In conclusion, many studies have shown that the adoption of problem-oriented policing can bring about reductions in both crime and other demands on police organisations (Scott and Clarke, 2020; Hinkle et al., 2020). However, getting a new idea adopted, even when it has clear advantages, can be difficult, as has been the case for problem-oriented policing. The diffusion of innovations is not an instantaneous act, rather it is a process that occurs over time and consists of a series of actions, with rejection possible at any stage of the process (Rogers, 2003). Where police organisations adapt to facilitate the delivery of problem-oriented policing, the innovation may be abandoned over time owing to changes in internal and external conditions and constraints. In complex systems it is easy to see how innovative practices might fail. If they are to succeed, attention to implementation must be constant and requires patience, vigilance and time. The adoption of problem-oriented policing can only result from concerted efforts made by police leaders to build and maintain a culture and infrastructure to sustain it. This involves providing training and ongoing professional development; supportive management and supervision; analytic support; and embedding the approach into promotion arrangements. This is no easy task, but one the police service will need to engage in if it wishes to reap the benefits that problem-oriented policing can bring to police organisations and the citizens they serve.