Co-design and action learning in local public services

Abstract

The paper argues that from a new public governance and service management perspective, local public services are best conceptualised as service systems in which users co-produce and co-design; this differentiates public from private services, which have lower of trust and shared values resulting in a goods-dominant logic and are an alternative to the new public management viewpoint. Referencing new case studies from Finland and Scotland, we further argue that for local public servicesŉ co-production as an action- learning environment supports and encourages co-design: this makes local public services a special case of codesign. Analysing the two cases of co-design, we argue that since public services are subject to public scrutiny, and since design is a social activity, there exists a wider democratic footprint. Finally, we argue that co-design of local public services is best analysed from the perspective of action learning, for which we suggest an analytical framework.

Keywords

Co-design action learning local public services systems Finland Scotland

Significance and research questions

Our paper is very much about learners, but not in the formal sense of occurring within structures and systems (Osborne, 2012). The learners featured here are non-traditional in Schuetze’s (2014) social sense, often being learners in later life: active in communities and public service systems – learning and applying knowledge to improve services: non-formal learners outside of the tradition education and training systems Yarnit (2015) discusses. Osborne and Strokosch (2013) argue, co-production of public services, involving new public governances is central to improving public services by initiating new ways of working, with effectiveness trumping efficiency. Co-production is a relationship between providers and users of goods or services sharing resources (including knowledge) and legitimacy (including power) to create public value. Our paper takes another step in this argument by exploring the processes by which new public service design and delivery models can be created. Co-design envisions, plans and creates effective solutions to users/customer’s problem (Petroski, 1996) balancing form and function (Walsh, Roy, Bruce, & Potter, 1992). We present a synthetic framework drawing upon research from services management, learning, innovation studies and public management arguing that in listening and learning services co-production and co-design necessarily support each other (Kinder, 2012). We believe that co-production is a necessary precursor or co-design; our first research question asks, can co-design occur without being accompanied by co-production.

Co-production in services

Ostrom (1973) formulated the idea of co-producing local public services; later extended by Ramirez (1999) and Wikström (1996a, 1996b) to services co-design. Later work by Gronroos (2001) and Ravald and Grönroos (1996) emphasises the new relationalities accompanying co-production an idea Normann (2002) centrally locates as a defining characteristic of all services (the others being proximate consumption, intangibility and subjective experience). This service management perspective draws attention to learning and innovation processes; whilst we find Vargo and Lusch’s (2007) idea of services dominant logic useful as a metaphor, its value to practically guiding public service innovation is limited to the sphere of marketing.

While aware of debates on the roles and relationships between service users and providers (Alford, 2002; Thomas, 2012) our focus here is on learning-in-using services; we adopt Pieper’s (1997) terminology: final user (citizen, patient, client), informal user (family) and formal user (local government staff).

Our unit of analysis is the practical street-level processes of service delivery and their redesign.With Walsh et al. (1992) we understand design as combining form and function to create socially and socially acceptable products: design has an illustrious genealogy that includes craft design, user-led design (von Hippel, 1988), participative design (Bjerknes, Ehn, & Kyng, 1987); and more recently human–computer interfaces (Pascal, Thomas, & Romme, 2013), collaborative design (Parker & Parker, 2007), co-creation (Cottam & Leadbeater, 2004), creative communities of social innovation (Jegou & Manzini, 2008) and (Meroni, 2008) community centred design.

Co-design and action learning

Our interest is in the learning and implementation processes creating new service co-designs: who learns what and how? Our second research question asks is co-design best conceptualised as an exercise in action learning? To answer this we use social learning theory (Engeström & Kerouac, 2007; Illeris, 2004; Vygotsky, 1978) to construct a new theoretical framework, arguing that co-design is best conceptualised as an interactive learning environment in which both providers and users learn how better to design and deliver public services.

We view the delivery and design of local public services not as (loosely coupled) networks, but rather as closely coupled systems, since security and reliability of services to vulnerable users is paramount. Service systems necessarily include users as co-producers and forms the learning environment in which co-design occurs, as we show in case studies from Finland and Scotland. Since the legitimacy of service systems is also evaluated by wider society, we explore the democratic footprints in these two countries as it relates to service redesign: our third research question is: how important is the wider footprint of co-design for design accountability and wider legitimacy in Finland and Scotland?

To answer these three research questions we contribute to theory on the nature of co-production and co-design and the relationship between the two, theorising a new framework around the idea that co-design in service systems is best understood as action learning. We relate ideas on co-design to the wider democratic footprint and the legitimacy it provides.

Conceptual and framework development

Co-design and co-production in local public services

By definition, services are subjectively experienced and co-produced (Normann, 2002). As Gronroos (2007) argues, successful service design replaces transactional interaction by relational interaction. Co-producing necessarily supports some degree of learning; what Brown and Duguid (2002, p. 138) term knowing how to be in practice and Schön (1983) calls reflection in action.

Design is a creative process – a better solution to a problem; resulting in an emergence – a new combination of form and function giving greater value than the inputs and alternative solutions (Thorpe & Gamman, 2011).Figure 1 adapts Sanders and Stappers (2008) to show design framing a problem and then proceeding by creative conflict to compose a solution: teams storm before they norm (Tuckman, 1965). Solutions embody needs and constraints as Hubka and Eder (1995) argue. Knowledge flows are shown to be non-linear, social and iterative resulting in unintended outcomes.

Figure 1.

Typical design processes (adapted from Sanders and Stappers (2008)).

Though local public services are complex (Lapsley, 2009), they have three advantages for co-design (see ‘Complexity, empathy and action learning in local public services co-design’ section): (a) empathy between providers and users, (b) like all services to some degree they are co-produced and therefore provide a learning environment for co-design and (c) service context references (to some degree) a local democratic community embodying moral propositions. For example, Carroll and Rosson (2007) argue that user involvement in design is not beneficial to outcomes; it is also a moral right of users.

We agree with Greenwald and Leavitt (1984) that knowledge is a significant variable in user involvement; however, we differ by distinguishing levels of learning arising from depths of co-production, shown in Table 1. Our argument is that action learning in service delivery is an environment for learning new service designs: co-production and co-design are inextricably linked, a point we return to in our analysis. For the purposes of argument we assume that depths of user involvement synchronise across Table 1 variables, accepting that in practice this may not be so.

Table 1.

Degrees of user engagement in co-design and co-production.

Depth of user(s) involvement in codesign and coproduction	Communications	Quality of interactivity and involvement	User(s) learning footprint	User impact on service design and delivery
Passive	* Mainly from provider consulting user	* Low * User treated as customer	* Individual	* Unknown
Voice	* Two-way * Provider dominant	* Maybe deep e.g. focus groups	* Small groups set up by providers	Some feedback from providers
Participant	* Two-way * Intensive	* Full member of design / delivery team	* System reach * Equal status to providers * Danger of inequality	* Traceable * Significant * Users’ stake in the service
Champion	* User + providers justify to wider public	* User and provider work closely together	* System and environment reach * The democratic local community	* User helps justify choices to public

The deeper service users engage in co-production, the more motivated and able they are to engage in co-design. Four depths of user involvement in co-production are shown: passive, voice, participant and champion (the latter meaning vocal supporter in the local democratic community and the first two being weak co-production). Following Alford (2009) we view users as intrinsically motivated to improve their own local public services and those of others. Whilst private service providers may gather information flows from customers, they appear less likely than public services to have shared values and deep empathy and trust. Conceptualising local public services as service systems, rather than organisations or functions, envisions users ‘pulling’ service design towards greater effectiveness and personalisation. Viewing local public services users as active agents in service systems (their production and design) is critical to understanding system effectiveness and sustainability. We note with Memon and Kinder (2014b) that loosely coupled networks guiding policy are quite different from closely coupled service systems that link input transformation to service outcomes. As Osborne et al. (2014) argue service redesign in search of effectiveness is the best way to achieve sustainable services. Table 1 illustrates deeper levels of co-production, supporting new service co-designs at a service system level, in which users help eradicate resource heaviness (Chapman, 2011) by refocusing service systems towards integrative, personalised delivery.

Complexity, empathy and action learning in local public services co-design

Local public services have advantages in design and delivery.Kinder et al. (2014) argue: close empathy between providers and users brings what Koskinen, Battarbee, and Mattelmäki (2003) term empathetic design.Empathy here is a high degree of interactivity, trust and shared values between service users and providers arising from mutual attachment and shared destiny emanating van Rijn et al. (2011) suggest from face-to-face interactions. Empathy allows experimentation and shared experiential learning using techniques such as service blueprinting (Radnor, Osborne, Kinder, & Baranova, 2014), service walk-throughs (Carroll, 1995), subjective user feelings evaluation at touch points (Richardson, 2010), the usability quadrant (Kinder, 2002) and value mapping (Katzan, 2011).

From this perspective, co-design is a learning process (Ehn, 1988), trading off as Chesbrough (2006) argues openness (to new ideas) and closedness (purposive focus on solutions). Learning environments Dindler (2010) terms fictional space, Ogle (2007) ideas space and Iser (1993) free play: trustful space for framing problems and solutions. Our second research question asks if this theorisation is found in practice.

Action learning

Action learning (Pedlar, 1983; Raelin, 2000) supports reflection by co-producers on practice; purposefully improving service practice.Originating in Vygotsky’s (1978) idea of learning in zones of proximate development: learners making sense of events/artefacts in the light of previously learned experiences and frameworks.Action learning in Engeström’s (1987, 1999, 2007) terms invites service users and providers to enter expansive cycles of learning, digging deeper together into what services deliver and to design improvements. Mattelmäki (2008) notes how probing tools in design stimulate expansive cycles. Wood, Bruner, and Ross’s (1976) idea of scaffolding is important, since professionals bring into the learning environment specialist knowledge that helps influence decisions (standards and regulations). As Sanders (2001) points out, action learning invites all participants to alter mindsets and envision new ways of framing problems and solutions and results in action (Argyris’ (1977) double-loop learning) inviting news ways of learning – Habermas’ (1971) emancipatory learning.

Action learners are motivated to do the best job possible for the sake of it (Sennett, 2008); far removed from passive ‘knowledge management’, unthinking transfers of supposed best practice, or imputing cognitive ability to inanimate learning organisations. Exchanging ideas between service providers and users in collaborative endeavour (Brereton, 1996) creates a social learning process in which cognition is situated, distributed and enculturated; then embedded in a new co-designed service (Svihla, 2010). Values are negotiated because (LaDantec, 2010) co-design presumes emotional attachment to learning a better service design; this is why empathy and new metaphors are important.

Collective processes and output

Teams of users, informal users and providers share learning (distribute cognitions) then test the team’s solution socially (acceptability, viability, cost, etc.): collective emergence to use Csikszentmihalyi and Sawyer’s (1995) term. Communications with the co-design team may be pictorial (Oxman, 2002), or Shaw (2010) schematic suggests (model, flow diagram, standard) and invariably includes negotiating terms, metaphors and language. These are not simply technical processes; rather they are beset by vagueness and ambiguity (Glock, 2009). Distributed cognitions create the wisdom of crowds enhanced by iterations, learning by trying and learning by unlearning (Jordan & Henderson, 1995). Digital divide issues are always important and public service co-design process should be careful not to exclude contributions (Paulini, Murty, & Maher, 2013) or sideline emotional attachments (Griffiths & Williamson, 2011), or normative questions (Noffke, 1997) over multiple time cycles (Lemke, 2000).

Framework

Co-design by providers and users of local public services systems is best conceptualised as action learning in a learning environment enabled by rich depths of co-production, characterised by individual and distributed cognitions and referencing the whole service system and wider democratic footprint: Figure 2 illustrates our synthetic framework for co-design.

Figure 2.

Co-design as active learning (adapted from Illeris (2004)).

We adapt Illeris’ (2004) action learning framework in which emotional attachment and the subjective meaning of learning to individuals are important. Inside the top (downward) triangle, individual learning occurs. Learning is shaped and motivated by the bottom (upwards) triangle: the context in which the learning occurs. Co-design is a team-learning environment (bottom triangle), distributing learning from individual (users or providers) cognitions (the top downward triangle): what Lemke (2000) terms a micro social grouping in which identity and ways of working are renegotiated. Note that problem and new service design at the centre of the diagram is the artefact around which expansive cycles of learning occur iterating between (cognitive) individuals and the design team. Individual identity and contribution to the new solution iterates between a situated learning content and learning dynamics (values and empathy with users being especially important) and are then fed into the team effort. The team references a technical-organisational learning environment (shaping the sense-making of ideas and suggestions) iterating with their collective socio-cultural environment (professional bodies and accountability or democratic footprint). Team and individuals become a learning environment of re-design ideas creating the new solution, referenced against the values and commitments of individuals and the social setting: effective redesign asks individuals to recompose their own identity performing new activities, nuancing their values, unlearning old ways of working.

Four key themes emerge from our framework, which we will use to structure our data presentation and analysis. First, the importance of heritage and context in framing the co-design project, which influences the footprint of the service system, its governances and the scope of accountability and empathetic shared destiny. Second, learning processes: the ease with which learning is distributed with the team and then to stakeholders including the processes by which innovation decisions are taken and the preparedness of stakeholders to implement learning. Co-design output is our third theme: how is the new co-designed service represented and tested including in this experimentation, playing with new ideas, testing, mental modelling, trialling and service walk-throughs. Our final theme is implementation: the preparedness of users and providers for double-loop learning, including ‘unlearning’ and adopting new ways of working.

Method

Our three research questions are as follows: can co-design occur without being accompanied by co-production, is co-design best conceptualised as an exercise in action learning and how important is the wider footprint of co-design for its legitimacy?

Our research design is to test our framework by gathering data guided by the four themes identified in ‘Framework’ section and use thematic cross-case analysis to answer our research questions. This then is an interpretive inquiry (Rabinow & Sullivan, 1985) what Yanow (1999) terms meaning making. Following Bhaskar (1986) our research is grounded in situated decision-making and events; what Hajer and Wagenaar (2003) call clumsy space. Our commentary is normative (Palumbo, 1992) and validity rests on usefulness rather than the truth: data honestly gathered data, honestly interpreted, respecting alternative interpretations (Angen, 2000). Our research design tests the usefulness of our framework against two cases, searching for validity and invalidity and triangulating with previous research (Eisenhardt, 1989).

We chose two cases of integrating local health and social care services co-designing a new service system. Since we explore the local democratic aspect of co-design and wanted to contrast Finland and Scotland we choose one case from each country. Our Scottish case whilst based on longitudinal research was updated with four new interviews. Cases were chosen for their richness and typicality (Yin, 1984); they are characterised by non-linearity, prominently featuring users and a service systems footprint crossing organisational boundaries to integrate social care and health services. In both cases, new service designs are in part the result of co-produced service delivery.

Dr Laitinen gathered data on the social and healthcare service systems of the City of Helsinki between 2010 and 2014, conducting 10 new (consented) semi-structured interviews for our case study, following up detailed questions by email and reviewing case documents such as plans, reports, reviews, assessments, and final reports and evaluations. Dr Kinder began gathering data on West Lothian social and healthcare service systems in 1998 and has since tracked the case as longitudinal research, publishing several papers on aspects of the services. This research includes recent document analysis and four new interviews (Senior Manager, Social Work middle manager and two users involved from the beginning of the process).

In our case the analytical structure is derived from the literature review and framework (‘Conceptual and framework development’ section).Validity and rigour come from interpreting the facts empathically (Yanow, 2000, 2003), whilst cross-referencing previous research. A danger, even with theoretical sampling is deterministically portraying events. We protect against this by careful coding, triangulating with previous theorisations (Eisenhardt, 1989, p. 103) and using cross-case and in-case analysis (Bryman & Bell, 2007). Validity here is usefulness rather than generalisable truth: our conclusions are contextually bounded (see Llewelyn, 2003); however, we consider that the framework we offer may be of use in analysing other co-design processes.

Data

Scottish and Finnish co-design footprints

Winschiers-Theophilus, Bidwell, and Blake (2012) show that the social factors referenced in design and communications vary across cultures. Björgvinsson, Ehn, and Hillgren (2012) and Ehn (1993) claim that Nordic design processes are especially democratic. If so, then as Suchman (1993) suggests, wider participative design makes explicit the critical, and inevitable, presence of values in the system development process. Do local public services design footprints vary between Finland and Scotland?

Nordic institutions have deep democratic traditions (Iversen, Halskov, Leong (2012) based on a shared absence of feudal structures; allemansrätt – everyman’s right to commons; industrial democracy (Bansler, 1989; Bjerknes et al., 1987) and teamworking (Petersen, 2004). Mouffe (2000) argues that liberal agonistic pluralism limits sovereignty in Nordic societies supported by a Lutheran tradition of self-study and improvement (see Elster, 1998 on deliberative democracy and its social architecture).

Scotland too has long democratic traditions in the Celtic clans, humanist Scottish Enlightenment, Reformation, working class trade union/socialist traditions and high voluntary organisation participation rate. The Scottish Government encourages co-production of services. Yet Bort, McAlpine, and Morgan (2012) describe Scottish local democracy, as below the national level, Scotland is the least democratic country in the European Union.

McIntosh et al. (1999) state: It could be said that Scotland today simply does not have a system of local government in the sense in which many other countries still do. The Confederation of Scottish Local Authorities recently said that Scotland is one of the most centralised countries in Europe. Our purpose is not to compare Scottish democracy with other parts of the UK or EU, rather to contrast wider service accountability with Finland.

Table 2 illustrates some of these differences. Scottish Councils have narrower powers than Finland, less control over income, are a larger size with fewer Councillors and face a centralising trajectory as the Parliament increasingly regulates education and social Work.

Table 2.

Comparison of local government in Finland and Scotland.

Population	5,500,000	5,200,000
Local government employees	425,000	246,400
Elected community politician to citizen ratio	1 in 500	1 in 4,270
Numbers standing for election (ratio to population)	1 in 140	1 in 2,071
Local election turnout	61%	32%
Average Council population	15,960	163,200
Average size (sq. km)	1,006	2,461
Local public expenditure as % of national GDP	20.10%	10.70%

Note: Scotland has 246,100 local government employees; additionally there are 157,400 Scottish NHS staff.

Bodker and Iversen (2002) suggest Finnish people expect their values to be embedded in local service design; Carroll and Rosson (2007) describe this as a moral proposition. In summary, the Nordic democratic tradition suggests a wider social footprint for co-design of local public service processes than Scottish local government: is this reflected in practice – our third research question?

Finnish case

Overview: Problem and solution

New practices supporting independent living for the older people were developed in the Lauttasaari district of Helsinki under the Customer-oriented Service Network project established by the Social department in 2010–13 to develop an integrated service centre for older people involving the Council, users and third sector. Target users were the one hundred over-65s, receiving informal care in their own homes from home care support services. Using personal care planning and budgeting to personalise services data, needs assessments and care plans were integrated across the service system. Lauttasaari was chosen for the co-design pilot because of the close working relationships health and social work professionals have with service users. One older user commented, We like to do what we can for ourselves and not depend entirely on the Council – its keeps us young.

Context

Service users and providers, supported by a Service Manager and a local university, carefully mapped care needs, services available and how these might be better joined up to meet individual needs. Older people and voluntary organisations were deeply involved at every stage in planning the new service system and revisiting designs as a result of learning from co-production. Although health and social care services are both provided by the City of Helsinki Council and professionals work closely on the ground, they are in separate Departments of the Council, as one senior clinician stated in the past its like we worked for different organisations.

User/provider learning potential and its relation to service system

It took time for the various agents to gel and trust each other, an occupational therapist commented, at first this team was hell, but then we got going. Eventually they decided to use graphic representations of services to plan the new designs including video clips, real cases, home carer customer journey maps and client pathways focusing pragmatically on the new system rather than defending the old.

Actually one representative from the City’s department said that we have been saying that we are customer centric, but everything had been done from the organization’s point of view, but now that was the first time to design services from the customer’s perspective. (Interviewee)

Another commented, we learned so much from just listening to older people, though of course, some have more contribution than others. Only after lengthy discussions and service walk-throughs were gaps in understanding and attachment to old systems overcome. This was especially so between the diverse sets of professionals in the team a process greatly helped by the user-centred focus of the project. Eventually, all interviewees agreed the project came to centre on users not individual services or organisations. One social worker said putting clients at the centre is what made a difference to thinking about how we organise services. Another said, once we started seriously looking at any ideas from anyone, we really got going.

Service problem and feasible solutions

We just dug deeper and deeper until we had solution, one social worker remarked, we tossed around ideas and sometimes investigated dead-ends, just to get our ideas straight. The project ended 2013 and final reports and evaluations have been already published. According to the documentation and interviewees, the customers, older people who joined the program were very satisfied. They have more choice and they are more active in the community; they feel more secure at home and dignity. The use of personal budgeting and the new role of a Service Manager as a mediator are working well. At a system level, the various service providers and stakeholders have requested the City of Helsinki to continue leading innovation.

Implementation: Ways of working, identities and evolving service system

In consultation with the team, City representatives decided that in spring 2014 the new service should be self-acting and independent, including maintaining user involvement in cross-disciplinary team to drive further innovations. According to the interviewees the ethos of co-designing and customer participation has been challenging for some small private companies, which had been seeking new business opportunities.

The City of Helsinki has now decided to offer co-design of service systems for older people as an option throughout the city, supported by the Department of Social Services and Health Care. This is a major innovation in the Finnish context, was the subject of a great deal of debate, which involved users from Lauttasaari promoting the new system to users throughout the city.

Scottish case

Overview: Problem and solution

West Lothian Council (Scotland) concerned at the ethics, quality and cost of its residential care homes in 1998 decided to replace them with 200 newly built, hub-and-spoke designed independent living homes, equipped with alert, alarm and assistive technologies. Building upon a history of cooperation between primary healthcare and social work professionals, the Council and NHS Trust agreed to set up a Community Health Care Partnership, with a devolved budget and responsibility for integrating primary healthcare and social services. Over the last 16 years, the footprint of this new service system has widened to include other local services (Community Education, Police), a R&D partnership with an assistive technology developer and grown into a popular service system supporting 8000 older people living independently in addition and integrated with healthcare services, reducing for example bed-blocking and using shared assessments and care plans.

Context

Initially framed as the physical replacement of residential homes with smart housing, the project soon recognised that the smartness was an integrated service system. The Community Health Car Partnership’s (CHCP’s) original messy governances persist: senior management report to a CHCP Board, Health Board and the local Council. The scope of accountability has widened as new support systems and technologies have been introduced.

User/provider learning potential and its relation to service system

The CHCP held hands-on training for users and their informal carers, learning to use and reconfigure (as needs changed) the assistive technologies. All welcomed single point of contact via voice alerts into the emerging service system. As more people joined the system and participated in co-producing their own care, suggested service improvements flowed and continue to flow into system design; a social worker commented, we played with ideas, explored alternatives just for the fun of it and learned so much from older people themselves: at times we were teachers and at other times pupils.

Formal staff training included IT and communications skills and management development courses in addition to on-going professional training. Several staff teams were upskilled, doing tasks previously performed by high-level professionals.

Learning in informal settings included cross-disciplinary teams exploring areas such as joint assessments, online database sharing and eradicating bed-blocking. Similar teams explored the integration of HR, training and budgetary systems. Users are represented in informal communities of practice and all Boards bringing learning from co-producing practice into service designs. In recent interviews a social worker describes thinking about a new way of classifying home aids and adaptations to better suit acute hospital discharges; this idea was circulated, discussed and within two months implemented.

Individual doctors, social workers and others were motivated to roll-out the new system seeing its advantages and having enjoyed strong empathy with each other and users; in some case following debate and initial opposition. Learning by doing led to numerous small changes that steadily integrated the service system, for example the GP and social work access to care package menus. Learning was rapidly distributed via online forums.

Service problem and feasible solutions

Council and NHS Trust offered the CHCP support by redesigning their own systems, one early example was joint discharge arrangements, other now include preventative health monitoring. Over time, more of the staff became co-located in places suitable to meet user needs including physical one-stop shops. Service system software and protocols now presume system-wide reach.

Implementation: Ways of working, identities and evolving service system

Over the last 16 years new staff were attracted by the integrated service system; staff are positive; doctors for example, closely work with social work on discharge and home care packages. Initial implementation was costly: IT equipment, training and professionals’ time spend by increasing trust, negotiated shared language and meaning, and adjusting work roles and relationships. Council and Trust stood back from day-to-day decisions in the service system; made easier as accolades accumulated. This support further encouraged and supported staff/user groups to generate ideas. Informal service users were involved in the physical and service redesign from the beginning: many vociferously so. As the service system delivered and was accoladed (Charter Marks, UK Council of the Year) the wider community of West Lothian (WL) took pride in its special service system. The CHCP holds regular public accountability meetings (including local doctors), which local people see as an important part of democratic accountability.

Analysis

Heritage and context

Heritage structures in Helsinki feature healthcare within the same organisation as social care, unlike WL where the two were organisational and functionally fragmented. Yet, the cases show the same challenges arose in each case suggesting that re-framing the issue not about organisational coordination but rather creating service systems ‘pulled’ by users overcoming barriers and boundaries. Nor is it possible simply to extrapolate top-down from institutional arrangements to predict levels of engagement in co-design: the WL case shows that even within less localised institutions, localism can be successful.

Whilst both projects aim to support independent living for older people, WL began life as smart housing replacing residential care, whereas Helsinki is an integrated ‘hub’ support to prolong independent living. Over its 16-year life the WL project migrated into an integrated service system, Helsinki’s final destination is yet to be set after four years. The co-design projects are quite different. In WL’s case, in a mixed welfare regime informal carers play a role in care and some vociferously opposed the new design, fearing a greater burden. In Helsinki’s Nordic welfare context, there are fewer onuses on informal carers in the service system design; the challenge there (Table 2) was getting diverse professionals to overcome functional separation – helped by all accepting the user-led system design brief. Both cases illustrate learning by doing. Unlike Bovaird and Loeffler (2012) we envisage co-production as resulting in practically beneficial services rather than abstract policy-making; experienced action learning from service delivery is the learning environment from which new service designs emerge.

As the cases show, in both instances wider accountability processes (Council, public discourse) could have rejected the new co-designed service, but instead approved it. We hypothesise that Finland has exceptionally strong wider accountability processes in public service design; in Scotland it is WL that is exceptional.

Learning processes

Both cases evidence the active learning (‘Complexity, empathy and action learning in local public services co-design’ section). For example, the individual WL social worker had an idea; it was distributed and quickly implemented as helping address the shared problem of independent living. The Finnish case describes the use of graphics illustrating service journeys as a way of distributing learning and arriving at a collective emergence. ‘Complexity, empathy and action learning in local public services co-design’ section also mentions Engeström’s (2007) expansive learning cycles, meaning continuously deeper investigations of a problem. The Helsinki case illustrates peeling away layers of meaning by professionals of what a user-centred service system looks like and eventually radically altering thinking. After 16 years, WL agents still tweak their online joint assessments to incorporate newly revealed good practice and current legislation mandating joint health and social care planning.

Both service systems involve an array of stakeholders apart from Council providers and users; these include technology providers, third sector groups and local companies. Setting the distributed learning footprint to include these external stakeholders has been important: WL became R&D partner with a technology provider, they also benefit from feedback on service satisfaction from voluntary organisations, with specific examples in the area of people with learning difficulties and in Finland with voluntary organisations representing older people.

The learning in these cases is shown to be social, action learning, reflecting ever more deeply into issues integrating social and health services for the elderly and to be providing a steady flow of learning that drives continuous innovation. As Figure 1 suggests, participants in the cases comment upon the storming and norming phases and the free play with service design ideas.Whereas for Rashman and Radnor (2005) and Rashman (2008) learning occurs in structured environments (best value, best practice transfer), here learning occurs in co-producing service delivery and then reflecting on how service design might improve and is in this way closer to the learning processes in Radnor et al. (2014) where learning feeds off emotional touch-points and waiting time: point-of-contact improvements ‘pull’ back office changes.

The new co-designed service system

Both projects had a heritage of strong empathy between providers and users. From the services-as-a-system perspective, users ‘pull’ delivery and influence service design: co-production supports co-design. Degrees of interactive empathy (enhanced, participative or passive) shape the quality of interactions, i.e. the degree to which ‘co’ features in design and production of services. It may be that at time particular users and providers are assigned to special projects (time limited, experimental, risk-laded); however, the norm is that for incremental changes and more radical change cumulating from more minor changes that user–producer interactions will occur during the delivery and continuous improvement of services.

If we assume that the four depths of user involvement in the new service system shown in Table 1 (passive, voice, participant and champion) apply to co-production and co-design then at the deepest level (champion in both co-production and co-design) users and providers are not only co-producing and co-designing, they jointly represent the service to the democratic community, achieving a wider accountability. Weaker co-production (participant) suggests a stable service model with incremental innovation, and weaker still (voice) that the services remain producer dominated with lower innovation velocity. At the weakest (passive) level of co-production, communications are one way and co-design non-existent. The participant and champion levels of co-production constitute an active learning system openly taking user/provider learning and implementing it in the technical-organisational sphere aligned with the socio-cultural context and (at champion level) giving democratic legitimacy to the service system: deutero-learning occurs in an open innovation system. Less open innovation and co-design occur at passive and voice depths of co-production simply because the learning environment is weaker and the voice of the provider dominates. At a passive level of co-production change is likely to be the result of top-down and programmed change such as lean initiatives with users given voice only when asked by providers: learning is single loop.

We cannot completely agree with Hartley’s (2005) point that public service research focuses too much on practical services and too little on governances: our cases show that innovations and new governances mutually catalyse. Also, unlike Moore and Hartley (2008) who focus on big governances changes, our cases and our experience suggest that altered roles, relationships and responsibilities occur between users and providers, between providers and very importantly within provider’s organisations – for example when point-of-contact decisions shift power from the back to the front office, redrawing hierarchic structure and redistributing power.

Implementation

In the early period at WL the Director of the project commented that integration is a scary word: only after time did smart housing migrate into integrated service design and implementation. For Helsinki the integration agenda was the raison d’être of the project. Sixteen years after starting, the process of service integration led by a design centred upon users is still active; changing ways of working and mindsets takes time. In the light of a recent study of self-directed services in Glasgow to improvise and immediately begin rolling-out the new system (Hunter et al., 2012), the rapid implementation in Helsinki is understandable. However, agreeing roll-out and its implementation are different: change involving altered mindsets and renegotiated social processes takes time: rapid transformation is for airport books, not reality.

Discussion and conclusions

We have suggested a four-part taxonomy of co-production and co-design. We recognise the taxonomy contradicts that relate to loosely coupled networks, rather than tightly coupled service systems.

Since we argue that co-production is the school for co-design, action learning is a useful way of conceptualising co-design. Noting Greenwald and Leavitt’s (1984) view that knowledge is the determinant variable in user involvement, we instead prefer ‘learning’; the cognitive contribution of active agents, and we taxonomise involvement into four degrees: passive, voice, participant and champion arguing that types of communications, the quality of interactivity, scope of user learning footprint and impact on co-design vary.

The cases reveal the interactions producing changed service delivery models and driving the integration of fragmented services. An important management implication is that if emergent service systems benefit from action learning, they must allow time for bottom-up remodelling, ideas testing, process negotiation and service walk-throughs. Experientially based learning sits uncomfortably with top-down plans (Pedler, 1983) and change coaches, since action learning presumes individual cognitive engagement, which is best not constrained; additionally, as the cases illustrate the fact that user and producers bring heritage pre-understanding as scaffolding (Wood et al., 1976) to the project is essential, not least for determining paths not to follow.

Previous research leads us to expect an absence or lower level of democratic accountability to public service co-design in Scotland than Finland: this is not what we find. We find that in Helsinki, the co-design was evaluated in local democratic process (local paper, Council meeting). Similar processes occurred in WL, in the initial period using specially convened local community meetings attended by informal carers users; later using press reports and periodic accountability meetings such as a the GP one. Do our findings dispute the case for Nordic exceptionalism in design (Ehn, 1993): we think not. Rather it is WL that is exceptional amongst areas of Scotland for reasons previously published (Kinder, 2010a, 2010b) including shared destiny and deep working class democratic traditions. This is not to dispute the literature low levels of local democracy in Scotland, merely to suggest that there are exceptions as Kinder (2012) shows.

Having defined service system innovation as a design with integrity, we support Wagner’s (2004) view that successful design emergence has social acceptability, which in a non-market (public service) context is the result of close (non-linear) learning between users and formal providers (Sanders et al., 2008). In these learning processes we anticipate activity by professionals educating users (and vice versa) in service parameters, and that the emergent design is subject to some wider democratic accountability. We have found that Ehn (1993) is right to claim a Nordic exceptionalism in wider democratic accountability of new service designs, noting that localism can prosper in areas such as WL despite a less auspicious institutional setting.

Much of the literature on public services innovation focuses on knowledge management. Our approach alternatively emphasises active individual cognition (‘Complexity, empathy and action learning in local public services co-design’ section), its distribution within the design team and external stakeholder referencing (Figure 2). Our contribution here is not novel, but importantly rejects the idea of design processes as a black box or inside the minds of managers (Walsh, 1992) and instead insists upon active learning and engagement by individuals within the innovation context. Co-design then entails action learning.

In practical terms our framework may encourage and support local public services agencies shifting to a service systems co-design, including processes such as using expansive cycles of learning. We have mentioned various areas of further research; in particular, we draw attention to exploring more widely the interrelationships between co-production and co-design.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Alford

(2002) Defining the client in the public sector: A social-exchange perspective. Public Administration Review 62: 337–346.

Alford

(2009) Engaging public sector clients. From service delivery to co-production, New York, NY: Macmillan.

Angen

J. M.

(2000) Evaluating interpretive inquiry: Reviewing the validity debate and opening the dialogue. Qualitative Health Research 10: 378–395.

Argyris, C. (1977). Double loop learning in organisations. Harvard Business Review September–October: 115–125.

Bansler

(1989) Systems development in Scandinavia: Three theoretical schools. Scandinavian Journal of Information Systems 1: 3–20.

Bhaskar

(1986) Scientific realism and human emancipation, London, UK: Verso.

Bjerknes

Ehn

Kyng

(1987) Computers and democracy – A Scandinavian challenge, Aldershot, UK: Avebury.

Björgvinsson

Ehn

Hillgren

P.-A.

(2012) Agonistic participatory design: Working with marginalised social movements. CoDesign: International Journal of CoCreation in Design and the Arts 8: 127–144.

Bødker

Iversen

O. S.

(2002) Staging a professional participatory design practice. Proceedings of NordiCHI, Aarhus, Denmark. New York, NY: ACM Press, pp. 11–18.

10.

Bort

McAlpine

Morgan

(2012) The silent crisis: Failure and revival in local democracy in Scotland, Glasgow, UK: The Jimmy Reid Foundation.

11.

Bovaird

Loeffler

(2012) From engagement to co-production: The contribution of users and communities to outcomes and public value. VOLUNTAS: International Journal of Voluntary and Nonprofit Organizations 23: 1119–1138.

12.

Brereton

M. F.

(1996) Collaboration in design teams in analysing design activity, New York, NY: Wiley.

13.

Brown

J. S.

Duguid

(2002) Local knowledge innovation in the networked age. Management Learning 33: 427–437.

14.

Bryman, A., & Bell, E. (2007). Business research strategies. Business research methods. USA: Oxford University Press.

15.

Carroll

Rosson

M. B.

(2007) Participatory design in community informatics. Design Studies 28: 243–261.

16.

Carroll, J. M. (Ed.). (1995). Scenario-based design: Envisioning work and technology in system development. New York, NY: Wiley.

17.

Chapman

(2011) Emotionally durable design – Objects, experiences and empathy, London, UK: Earthscan.

18.

Chesbrough, H. W. (2006). Open innovation. Boston: Harvard Business School Press.

19.

Cottam

Leadbeater

(2004) RED paper 01: Health: Co-creating services, London, UK: Design Council.

20.

Csikszentmihalyi

Sawyer

R. K.

(1995) Creative insight: The social dimension of a solitary moment. In: Sternberg

R. J.

Davidson

J. E.

(eds) The nature of insight, Cambridge: MIT Press, pp. 329–364.

21.

Dindler, C. (2010). The construction of fictional space in participatory design practice. CoDesign: International Journal of CoCreation in Design and the Arts, 6, 167–182.

22.

Ehn

(1993) Scandinavian design: On participation and skill. In: Schuler

Namioka

(eds) Participatory design: Principles and practices, Hillsdale, NJ: Lawrence Erlbaum Associates Inc. Publishers, pp. 44–77.

23.

Ehn, P. (1988). Work-oriented design of computer artefacts. Sweden: Falköping, Arbetslivscentrum.

24.

Eisenhardt

K. M.

(1989) Building theories from case study research. The Academy of Management Review 14: 532–550.

25.

Elster, J. (1998). Deliberative democracy (Vol. 1). Cambridge, UK: Cambridge University Press.

26.

Engeström

(1987) Learning by expanding – An activity-theoretical approach to development research, Helsinki: Orienta-Konsultit.

27.

Engeström

Miettinen

Punamäki

R. L.

(1999) Perspectives on activity theory, Cambridge, UK: CUP.

28.

Engeström

Kerouac

(2007) From workplace learning to inter-organisational learning and back: The contribution of activity theory. Journal of Workplace Learning 19: 336–342.

29.

Glock

(2009) Aspects of language use in design conversation. CoDesign: International Journal of CoCreation in Design and the Arts 5: 5–19.

30.

Greenwald

A. G.

Leavitt

(1984) Audience involvement in advertising: Four levels. The Journal of Consumer Research 11(June): 581–592.

31.

Griffiths

Williamson

(2011) Evolving new methodologies in practitioner research: The case of visual methods. In: Khine

M. S.

Saleh

I. M.

(eds) Practitioner research (Chapter 3), Hauppauge, NY: Nova.

32.

Gronroos

(2001) The perceived service quality concept – A mistake? Managing Service Quality 11: 150–152.

33.

Gronroos

(2007) In search of a new logic for marketing, London, UK: John Wiley.

34.

Habermas

(1971) The post-national constellation: Political essays, Cambridge: MIT Press.

35.

Hajer, M. A., & Wagenaar, H. (Eds.). (2003). Deliberative policy analysis: Understanding governance in the network society. Cambridge, UK: CUP.

36.

Hartley

(2005) Innovation in governance and public services: Past and present. Public Money and Management 25: 27–34.

37.

Hubka

Eder

W. E.

(1995) Design science, Heidelberg, Germany: Springer.

38.

Hunter, S., Manthorpe, J., Ridley, J., Cornes, M., & Rosengard, A. (2012). When self-directed support meets adult support and protection: Findings from the evaluation of the SDS test sites in Scotland. The Journal of Adult Protection, 14, 206–215.

39.

Illeris, K. (2004). A model for learning in working life. Journal of workplace learning, 16, 431–441.

40.

Iser, W. (1993) The fictive and the imaginary: Charting literary anthropology. Baltimore, MD: John Hopkins Press.

41.

Iversen

O. S.

Halskov

Leong

T. W.

(2012) Values-led participatory design. CoDesign: International Journal of CoCreation in Design and the Arts 8: 87–103.

42.

Jordan, B., & Henderson, A. (1995). Interaction analysis: Foundations and practice. The Journal of the Learning Sciences, 4, 39–103.

43.

Katzan

(2011) Essentials of service design. Journal of Service Science 4: 43–60.

44.

Kinder, T. (2002). The introduction of an infrastructure for tele-working and implementation of joined-up-government in local public administration: a sociotechnical analysis of a West Lothian case study. Research Policy, 31, 329–355.

45.

Kinder

(2010a) e-Government service innovation in the Scottish criminal justice information system. Financial Accountability & Management 26: 21–41.

46.

Kinder

(2010b) Evolving accountabilities and outcome agreements – Experience and prospects form Scottish public services. In: Ball

Osborne

(eds) Social accounting and accountability (Chapter 6), London, UK: Routledge.

47.

Kinder

(2012) Learning, innovating and performance of locally delivered public services. Public Management Review 14: 403–428.

48.

Kinder, T., Laitinen, I., Aulton, K., & Stenvall, J. (2014). Making sense of service integration: From inter-organisational collaboration to service systems. University of Edinburgh Working Paper.

49.

Koskinen

Battarbee

Mattelmäki

(2003) Empathetic design user experience in product design, Helsinki, Finland: IT Press.

50.

LaDantec

C. A.

(2010) Situating design as social creation and cultural cognition. CoDesign: International Journal of CoCreation in Design and the Arts 6: 207–224.

51.

Lapsley

(2009) New public management: The cruellest invention of the human spirit? ABACUS 45: 1–21.

52.

Lemke

J. L.

(2000) Across the scales of time: Artefacts, activities and meanings in ecosocial systems. Mind Culture and Activity 7: 273–290.

53.

Llewelyn

(2003) What counts as theory in qualitative management and accounting research? Accounting, Auditing and Accountability Journal 16: 662–708.

54.

Manzini, E., & Jégou, F. (2008). Collaborative services: Social innovation and design for sustainability. Poli Design, 111.

55.

Mattelmäki

(2008) Probing for co-exploring. CoDesign: International Journal of CoCreation in Design and the Arts 4: 65–78.

56.

McIntosh, et al. (1999). Moving forward: Local government and the Scottish Parliament. Report of the Commission on Local Government and the Scottish Parliament.

57.

Memon, A., & Kinder, T. (2014). Management in an increasingly collaborative and integrated public sector. Edinburgh: University of Edinburgh, Working Paper.

58.

Meroni, A. (2008). Strategic design: Where are we now? Reflection around the foundations of a recent discipline. Strategic Design Research Journal, 1, 31–28.

59.

Moore

Hartley

(2008) Innovations in governance. Public Management Review 10: 3–20.

60.

Mouffe

(2000) The democratic paradox, London, UK: Verso.

61.

Noffke

(1997) Professional, personal and political dimensions of action research. Review of Research in Education 22: 305–343.

62.

Normann, R. (2002). Services management: Strategy and management in service business. 3rd edn. Chichester, UK: John Wiley & Sons.

63.

Ogle, R. (2007). Smart World: Breakthrough creativity and the new science of ideas. Boston: Harvard Business School Press.

64.

Osborne, C. M. (2012). Editorial: On-going reforms in adult education in Europe. Journal of Adult and Continuing Education, 18, 1–3.

65.

Osborne

S. P.

Radnor

Vidal

Kinder

(2014) A sustainable business model for public service organisations? Public Management Review 15: 165–172.

66.

Osborne

S. P.

Strokosch

(2013) It takes two to tango? Understanding the co-production of public services by integrating the services management and public administration perspectives. British Journal of Management 24: S31–S47.

67.

Ostrom

(1973) The intellectual crisis in American public administration, Tuscaloosa, AL: Alabama University Press.

68.

Oxman

(2002) Structure and status in design teams: Implications for design management. Academic Review of the Design Management Journal 1: 55–53.

69.

Palumbo

D. J.

(1992) Bucking the tide. In: Dunn

Kelly

(eds) Advances in policy studies since 1950, New Brunswick, NJ: Transaction Publishers, pp. 59–80.

70.

Parker

(2007) Unlocking innovation. Why citizens hold the key to public service reform, London, UK: Demos.

71.

Pascal

Thomas

Romme

A. G. L.

(2013) Developing a human-centred and science-based approach to design: The Knowledge Management Platform Project. British Journal of Management 24: 264–280.

72.

Paulini

Murty

Maher

M. L.

(2013) Design processes in collective innovation communities: A study of communication. CoDesign: International Journal of CoCreation in Design and the Arts 9: 90–112.

73.

Pedler, M. (Ed.). (1983). Action learning in practice. Aldershot, UK: Gower.

74.

Petersen

(2004) Embracing values in designing domestic technologies. Proceedings of European conference on cognitive ergonomics, New York, NY: ACM Press, pp. 161–171.

75.

Petroski

(1996) Invention by design: How engineers get from thought to thing, Cambridge, MA: Harvard University Press.

76.

Pieper

(1997) Technology and the social triangle of home care: ethical issues and the application of technologies to dementia care. In: Bjorneby

van Berlo

(eds) Ethical issues in use of technology for dementia care, Knegsel, the Netherlands: Akontes, pp. 1–31.

77.

Rabinow

Sullivan

W. M.

(1985) Interpretive social science, (2nd ed. Berkley: University of California Press.

78.

Radnor

Osborne

S. P.

Kinder

Baranova

(2014) Operationalizing co-production in public services delivery: The contribution of service blueprinting. Public Management Review 16: 402–423.

79.

Raelin

J. A.

(2000) Work-based learning: The new frontier of management development, Reading, MA: Addison-Wesley.

80.

Ramirez

(1999) Value co-production: Intellectual origins and implications for practice and research. Strategic Management Journal 20: 49–65.

81.

Rashman

Radnor

(2005) Learning to improve: Approaches to improving local government services. Public Money and Management 25: 19–26.

82.

Rashman, L. J. (2008). Organizational knowledge and capacity for service improvement in UK public organizations. Doctoral dissertation, University of Warwick.

83.

Ravald

Grönroos

(1996) The value concept and relationship marketing. European Journal of Marketing 30: 19–30.

84.

Richardson, A. (2010, November 15). Using customer journey maps to improve customer experience. HBR blogs. http://blogs.hbr.org.

85.

Sanders, E.B.-N. (2001). Collective creativity. Loop AIGA Journal of interaction design education. Retrieved from http://www.aiga.org/inspiration/.

86.

Sanders

E. B. N.

Stappers

P. J.

(2008) Co-creation and the new landscapes of design, 2008. CoDesign: International Journal of CoCreation in Design and the Arts 4: 5–18.

87.

Schön

D. A.

(1983) The reflective practitioner, New York, NY: Basic Books.

88.

Schuetze

H. G.

(2014) From adults to non-traditional students to lifelong learners in higher education: Changing contexts and perspectives. Journal of Adult and Continuing Education 20: 37–55.

89.

Sennett

(2008) The craftsman, London, UK: Penguin.

90.

Shaw

B. G.

(2010) A cognitive account of collective emergence in design. CoDesign: International Journal of CoCreation in Design and the Arts 6: 225–243.

91.

Suchman, L. (1993). Foreword. In D. Schuler & A. Namioka (Eds.), Participatory design principles and practices. Hillsdale, NJ.

92.

Svihla

(2010) Collaboration as a dimension of design innovation. CoDesign: International Journal of CoCreation in Design and the Arts 6: 245–262.

93.

Thomas

J. C.

(2012) Citizen, customer, partner: Engaging the public in public management, New York, NY: M.E. Sharpe.

94.

Thorpe

Gamman

(2011) Design with society: Why socially response design is good enough. CoDesign: International Journal of CoCreation in Design and the Arts 7: 217–230.

95.

Tuckman, B. W. (1965). Developmental sequence in small groups. Psychological Bulletin, 63, 384.

96.

van Rijn, H., Sleeswijk Visser, F., Stappers, P. J., & Özakar, A. D. (2011). Achieving empathy with users: The effects of different sources of information. CoDesign, 7, 65–77.

97.

Vargo

S. L.

Lusch

R. F.

(2007) Service-dominant logic: Continuing the evolution. Journal of Academic Marketing 36: 1–10.

98.

Von Hippel

(1988) The sources of innovation, Oxford, UK: OUP.

99.

Vygotsky

L. S.

(1978) Mind in society: The development of higher psychological processes, Cambridge, MA: Harvard University Press.

100.

Wagner

(2004) Open planning – A reflection on methods. In: Boland

Collopy

(eds) Managing as designing, Redwood City, CA: Stanford University Press, pp. 153–163.

101.

Walsh

Roy

Bruce

Potter

(1992) Winning by design, Oxford, UK: Blackwell.

102.

Wikström

(1996a) The customer as co-producer. European Journal of Marketing 30: 6–19.

103.

Wikström

(1996b) Value creation by company-consumer interaction. Journal of Marketing Management 12: 359–374.

104.

Winschiers-Theophilus

Bidwell

N. J.

Blake

(2012) Altering participation through interactions and reflections in design. CoDesign: International Journal of CoCreation in Design and the Arts 8: 163–182.

105.

Wood

Bruner

Ross

(1976) The role of tutoring in problem solving. Journal of Child Psychology and Child Psychiatry 17: 89–100.

106.

Yanow, D. (1999). Conducting interpretive policy analysis (Vol. 47). Sage Publications.

107.

Yanow

(2000) Conducting interpretive policy analysis, Thousand Oaks, CA: Sage.

108.

Yanow

(2003) Accessing local knowledge: Policy analysis and communities of meaning. In: Hajer

M. A.

Wagenaar

(eds) Deliberative policy analysis, Cambridge, UK: CUP, pp. 228–246.

109.

Yarnit, M. (2015). Whatever became of the learning city? Journal of Adult and Continuing Education, 21, 24–35.

110.

Yin

R. K.

(1984) Case study research: Design and methods, Newbury Park, CA: Sage.