Critical Features of a Digital Service Innovation Team at the Swedish Migration Agency

Introduction
Governments around the world face increasingly complex problems in an operating environment that is rapidly changing as digital transformation of society continues. Many governments and public sector agencies are looking to innovation to find new solutions and answers to societal challenges and citizen expectations. But innovation in the public sector is fraught with challenges, and increasingly public sector agencies are looking at organizational models alternative to the bureaucratic hierarchy in order to encourage innovation. Innovation teams that are partly autonomous from the parent organization are proposed as a possible solution to some of the innovation challenges that public sector faces.

This thesis looks at the relevance and viability of an innovation team concept in the context of digital communication service innovation at the Swedish Migration Agency (SMA). The SMA faces significant challenges in its communication with applicants for residency and citizenship, in particular due to the dramatically increased amount of applicants and stakeholders it needs to serve since the so called refugee crisis in the autumn of 2015. The political leadership of the Swedish government is expecting the SMA to fully use the potential of digitalization in the midst of these challenges.

The main research question of this thesis is:

How might an innovation team enable the Swedish Migration Agency to build communication services that are digital by default and meet the needs of the users?

When I conducted this research project I was social media manager at the Swedish Migration Agency, where I had been in charge of establishing new communication services in social media channels. From this experience I was encouraged by senior management to explore how the SMA generally could adapt its communication services to the digital era. An underlying assumption was that the SMA was struggling to use the potentials of digital, but any formalized knowledge of what was causing it and how it could be tackled was lacking.

To tackle the research question I place the concept of innovation teams within the theoretical framework of the ambidextrous organization. Using action design research together with colleagues, I build and evaluate critical organizational features of a recommendation for an innovation team at the SMA.

The Swedish Migration Agency’s challenge
The SMA is an autonomous government executive agency. Its primary task is to process and decide on applications for visas, residency, asylum and citizenship in Sweden. A significant secondary task is to provide housing and basic support to asylum seekers during the application process if they are not able to provide such themselves.

Partly due to the steep increase in the amount of asylum seekers that arrived in Sweden during the autumn of 2015, the SMA is facing significant challenges in processing applications within reasonable – even legal - processing times. This has resulted in increased costs both directly in the form of increased personnel costs, due to significant new recruitments to handle the increased amounts of applications, and increased housing costs due to more and longer stays in asylum centres, as well as indirect societal costs such as lack of skilled workforce in Sweden due to difficulty for foreign employees to gain work permits within reasonable time, and human suffering due to difficulty of gaining family reunion within the nine month timeframe stipulated by Swedish law.

Both the ministry in charge and the SMA itself are proposing efficient use of IT and digitialization as one means to tackle these challenges (Migrationsverket 2016a; Ekonomistyrningsverket 2015). However, while the SMA can show some significant cost savings due to ongoing digitialization (Migrationsverket 2015b), the Swedish National Audit Office (Riksrevisionen 2016) points to the fact that public sector organizations generally are not living up to the expectations of e-government, that government agencies can do more to reach national targets on digitialization, and that there is a tendency for Sweden to display a slower pace of digital development in the public sector than leading European countries.

In this context the government has decided to launch the digital by default (Digitalt först) programme for digital transformation of public services in Sweden. The stated objective for government executive agencies like the SMA is that digital services should be the primary choice of public sector contact with individuals and organizations (Regeringskansliet 2016). Furthermore, the digital by default programme suggests that solutions should be ‘innovative and built around people’s “life events”’, essentially a user-centered approach to innovation (Näringsdepartementet 2015). One such life event is moving to Sweden, where the SMA plays an important role.

Consequently, one challenge for the SMA is how it can innovate in communication services that are digital by default and meet the needs of the users.

The limitation to communication services is deliberate, based on the fact that infrastructure services digitization is where the SMA has had more success recently, whereas there is some anecdotal evidence that it is struggling more in client user interface digitization (Migrationsverket 2015b), with one key project (called Mina sidor – My Pages) being delayed with several years.

Organizational design and action design research
The move towards digital transformation in public service implies moving away from siloed government agencies and purely incremental performance improvement (Danneels and Viaene 2015). If an organization is to remain effective as it changes, managers must continuously evaluate the way their organizations are designed: for example, the way work is divided among people and departments, and the way it utilizes its human, financial, and physical resources. Organizational design involves difficult choices about how to control—that is, coordinate organizational tasks and motivate the people who perform them—to maximize an organization’s ability to create value. (Jones 2013)

But only little research has focused on how public sector organizations can foster innovation given their rather efficiency-oriented and hierarchical organizational design often considered not suitable for innovation (Jones 2013; Cannaerts, Segers, and Henderickx, 2016).

Research has shown that in corporate settings so called ambidextrous organizational design is positively associated with innovation. When it comes to launching innovative services, ambidextrous organizations are significantly more successful than other types of organizational design (O’Reilly and Tushman 2004). Building on O’Reilly’s and Tushman’s (n.d.) definition the concept of the ambidextrous organization design in a public sector context refers to:

The ability of government to both explore and deliver efficient and trusted services – to deliver with mature technologies and services where efficiency, control and incremental improvement are prized and to also explore new technologies and services where flexibility, autonomy, and experimentation are needed.

Ambidextrous organization designs create distinct explorative units that have their own unique processes, structures and cultures that are specifically intended to support innovation. These units, often comprised of one or more innovation teams, reside within the larger parent organization but have been set up to support the unique approaches, activities and behaviours required when engaging in innovation. (O’Reilly and Tushman n.d.). Public sector innovation teams can be argued to broadly fall within the explorative part of concept of organizational ambidexterity.

According to John Kotter (2014) these explorative teams run in parallel and should complement the existing hierarchy – allowing the so called exploitative hierarchical system to continue produce reliable outputs with maximum efficiency.

To gain more insight into how public organizations – and the Swedish Migration Agency in particular in the context of the challenges it is facing - can establish both efficiency and innovation, the following research question is proposed:

How might an innovation team enable the Swedish Migration Agency to build communication services that are digital by default and meet the needs of the users?

Using action design research (ADR) I aim to identify the critical features of the organizational design components of a proposed innovation team at the SMA, corresponding to an explorative part of a potential ambidextrous organization design at the SMA. These critical features build on those identified by Nadler and Tushman (1997), see table 1 below. Table 1: The four organizational components by Nadler and Tushman 1997

As the critical features are built and evaluated, they are marked with colour codes in the text. The colour codes are explained in the footnote visible on all pages of the thesis. This coding is done with the intention of making it clear where the critical features emerge from, without making the text too theoretical.

I start from an existing theoretical perspective (ambidextrous organization design), go to my primary and secondary data, back to theory and try to build new conceptual and practical propositions. The primary data sample consists of two workshops arranged for the purpose of this thesis at the SMA during October 2016, as well as official documents from the SMA, and two study visits to innovation teams in other Swedish public organizations.

ADR research consists of four stages (Sein et al 2011): 1) problem formulation, 2) building, intervention and evaluation, 3) reflection and learning, and 4) formalization of learning. In this thesis these stages have been put into action as illustrated in figure 1, and further explained below.

Problem formulation
This stage consists of practice inspired research driven by the SMA’s need to better understand how it can innovate in digital communication services. This phase was informed by academic and business literature on innovation in the public sector and other complex environments, including study visits to the Swedish Labour Service’s Digital Innovation Centre, the City of Stockholm’s and regional stakeholders’ OpenLab, previous case work on public sector innovation teams, as well as studying strategic documentation of the SMA.

The outputs of this stage are accounted for in chapter 2.

Building and evaluation
Inspired by the work done by Mead (2002) I wanted to offer the chance of participating in the research project to as wide a range of people as possible at the SMA without being overwhelmed by too many participants. A letter of invitation to an information session about the research project and the possibility to participate in it was drafted and e-mailed to hundreds of employees ranging from assistants to directors at the departments of SMA directly involved in the development of digital communication services at the SMA, as well as to the HR department. The open invitation was signed by the director of communication at the SMA in order to gain some recognition and legitimacy for the research. My involvement as facilitator was explicitly stated along with a brief description of the project. Participation in the information session was based on voluntary, informed self-selection.

Image 2: Fishbone diagram (Migrationsverket 2013), English translation by author.

About 20 persons showed up for the half-hour information session where some initial insights from the problem formulation stage were presented with the explicit intention of establishing a sense of urgency, as suggested by Kotter (2012). The participants ranged from experienced middle managers to experts at the beginning of their career. At least one person from each invited department showed up, with the exception of HR. In addition to the rationale of the research, I spoke about how people could contribute to building knowledge and evaluation if they decided to opt into participating in two workshops.

Five persons signed up for the two workshops lasting half a day each. Including myself, the group was mixed, with two middle managers, three experts, one entry level professional, four women and two men, representing strategy, development, production, and quality assurance of digital communication services at the SMA.

Image 3: Trends template for the future trends tool (Hyper Island n.d.)

In order to facilitate the workshops, I used the fishbone diagram tool (image 1) and the future trends tool (image 2.). During the first workshop we identified innovation barriers at the SMA with the help of the fishbone tool. We drilled down into those barriers that the workshop participants collectively prioritized as the most important ones, to find the underlying reasons. The outputs are documented in table 2.

During the second workshop we identified trends in innovation capacity building that could be of interest for the SMA in order to overcome the barriers from the first workshop. After classifying the identified trends into four different categories of public service innovation capacity building, we explored in depth one trend in each category that the workshop participants collectively decided held the most promise for the SMA to overcome barriers. In this manner focus was on the trends offering opportunities, rather than those posing threats.

The outputs of this stage are accounted for in chapter 3.

Reflection and learning
The third stage of the ADR process, reflection and learning, took place in parallel with the previous stages. I engaged in conscious reflection on the problem framing, innovation teams, public sector innovation literature and the emerging concept of an innovation team at the SMA to ensure that the research process was informed by both theory and the SMA context.

I also explicitly asked the workshop participants to reflect in writing after each workshop. The question I asked them to reflect on anonymously was:

‘Which are the most important insights for you and for the SMA from the workshop and why?’

The response rate was 8 of 10 expected answers.

Workshop participants have also been given the opportunity to give feedback on a draft of this thesis, with a response rate of one out of five. Furthermore, the SMA’s communication management board were offered the same opportunity to feedback on a draft as well as through a presentation of preliminary conclusions that I gave them in early December. This created some discussion that is accounted for in the  chapter Reflection.

The outputs of this stage stage are accounted for in chapter 4.

Formalization of learning
In the formalization of learning stage, situated learning is further developed into general solution concepts for a field of problems (Danneels & Viaene 2015). The generalized outcomes of this research are critical features of an organization design for an innovation team at the SMA and necessary organizational shifts that the SMA needs in order to implement an innovation team. This research is primarily focussed on building a solution for the SMA’s challenges, and should not as such be generalized to a broader class of problems. However, it does contribute to growing body of research on innovation teams in the public sector, and in that context brings to attention critical features of organizational design that can help connect a public sector innovation team to the core hierarchy.

The outputs of this stage are accounted for in chapter 5.

Critical and ethical discussion on theory and methodology
The research focus on the ambidextrous organizational design concept is chosen based on the theoretical potentials of the concept and the existing practical examples from similar contexts as the SMA. With that being said, the concept also has theoretical limitations. As Pan Fagerlin and Björklund (2010) write: ‘One distinctive limitation in the ambidexterity model is that knowledge creation requires the integration of new and existing knowledge that cannot be separated.’ Creating innovation teams may become a barrier to knowledge creation, and thus risk not advancing innovation. This thesis aims to the extent possible address these weaknesses and be open about them in the recommendations to the SMA.

In ADR the researcher takes on responsibilities for encouraging change (Bradbury 2013). In the case of this research I have encouraged change at the SMA in order to enable digital by default communication services that meet the needs of the users. While I value this as positive social change, it may not be unequivocally positive if critically scrutinized:

Defining the needs of the users is exceptionally difficult in the context of immigration, where the users literally come from almost every country of the world, and their levels of digital maturity are highly diverse. How can it be proven that building communication from a set of more philosophical principles – think Aristoteles’s ethos, pathos, logos - wouldn’t run lesser risk of negative change for all users than by attempting to build them according to user needs in the context of the SMA? Dissimilar from business, morally government cannot ignore any user. With about at least 200 cultural contexts as well as educational backgrounds ranging from illiteracy to Nobel laureates, according to what principles does the SMA define user needs without ignoring someone? This is something that the SMA needs to think of in the context of its work with user needs, but falls outside the scope of this thesis.

Encouraging the broader use of digital technologies is not a neutral act. Digital by default public services raise questions such as: Are public servants capable of re-thinking services and collaboration forms to allow for automation and user-control as the ‘default’? And are all users ready for handling this new paradigm and the responsibility that it implies? (OECD 2016) My answers to these questions are ’yes’, but in that answer is embedded my view of the world.

Furthermore, ADR as a method runs the risk of personal over-involvement of the researcher that may bias research results (Bradbury 2013). During the workshops in particular, I have consciously and predominantly acted as a facilitator, asking open questions and made an effort to avoid making value judgements. To the extent that I have been an active contributor in the workshops, I have strived to express clearly if I am referring to theories, observed facts, or my own opinions and experiences.

ADR is cyclic in nature and should contain several iterations in the building, intervention and evaluation stage (Danneells and Viaene 2013). This part-time research project has had a fairly limited time span (September – December 2016) and much of the more fundamental changes, such as observing actual change in the organizational design of the SMA as a result of the research is not within its scope.

= Problem formulation =

Definitions
Innovation in the public sector can be defined as (OECD n.d.):

‘The implementation by a public sector organisation of new or significantly improved operations or products.’

This definition of innovation can be contrasted to incremental or continuous development of existing solutions in the public sector (Steiber 2016). Furthermore, an innovation must be implemented, meaning that it cannot just be a good idea but rather must have been put in place operationally. Considering the SMA is not a research and development heavy organization the term innovation should refer mostly to the types of innovation that is new to the organization, the innovation does not have to be new to the world. (Bason 2010; OECD 2014a)

Thus in the context of the SMA building digital by default communication services that meet the user needs, innovation means the implementation by the SMA of digital communication service ideas that are new for the SMA, or significantly improved.

The terms innovation teams, innovation units, innovation offices, i-labs, i-teams or similar are often used in the public sector in the same sense as the more common term innovation labs. A nascent community of professionals and academics practicing and studying the work of public sector innovation teams or labs now exists as evidenced by professional conferences (e.g. SKL and VINNOVA 2016; European Commission 2016; OECD 2014), online discussions (see e.g. #psilabs on Twitter) and an increasing amount of professional (e.g. NESTA 2015b; Burstein and Black 2014) and academic articles (e.g. Tonurist et al 2015; Brown et al 2016).

Despite this growing body of knowledge the innovation team is not a well defined phenomenon. Some traits do however seem to be common to many public sector innovation teams, notably (Technopolis 2012; Modig 2016):

§  Significant autonomy from the parent organization(s);

§  Bringing together different disciplines and approaches from design, science, technology and business;

§  An exploratory and experimental approach

§  The active involvement of users at all stages of development (co-creation);

§  Multiple partners from private and public sectors;

§  Involvement of knowledge and expertise from the parent organization(s);

§  A dedicated space (real or virtual) for experimentation and developing new ideas;

§  Generating ideas for the renewal of government operations.

Academic and applied research on innovation teams have attempted at summarizing the definition of the innovation tea, form of organization:

''’Innovation [teams]  are a new institutional form that tackle policy and public sector challenges in a radically different way to the traditional approaches formerly undertaken. Their proliferation highlights that if governments are to open their doors to innovation, they need to create dedicated resource, skills, space - whether real or virtual - and executive support to enable innovation to happen. Recognising this, public sector managers are looking to partner with or insource teams that have the requisite skills and experience that will help to identify and implement the most promising ideas.’'' (Reynolds 2015)

''‘Innovation [teams] are an attempt to structure (radical) change processes within public organizations. [Teams] are supposed to approach problems in non-hierarchical ways and operate in a more horizontal manner across stakeholders and including professionals from a variety of backgrounds.’'' (Tonurist et al 2015).

Further building on the above definitions as well as the theory of ambidextrous organizations, and for the purposes of this paper, the definition used for a public sector innovation team is:

A network-like structure parallel to the traditional hierarchy designed to overcome barriers to innovation and build innovation capacity in the public sector.

Does digital by default require innovation?
Digital by default in public services can be interpreted either as a characteristic of the output an outcome of digital service innovation.

The UK’s Government Digital Service (n.d.) defines digital by default as an outcome:

‘Digital services that are so straightforward and convenient that all those who can use them will choose to do so whilst those who can’t are not excluded.’

The Swedish Government uses the term ’digital first’ (in Swedish ’digitalt först’) in a similar fashion.

But digital by default can also mean that digital becomes a starting point for creating service concepts, instead of an afterthought or supporting function (Danneels and Viaene 2015). When a digital service is not simply a digitization of an analogue process then it can be said to be a digital by default output. Offering a truly digital customer experience is something profoundly different than digitizing an existing product or service.

Building a digital service that is the preferred choice of users or public service providers is certainly possible without having to innovate. It can simply be the result of incremental development – it does not have to be something profoundly different than digitizing an existing product or service.

But given government’s stated desire to remodel itself around the needs of citizens,

it can arguably no longer afford to ignore innovation (Brown et al 2014). Governments of the world’s most developed countries have recognized that there is a link between successful digital government and innovation (OECD 2014b). The Swedish government is looking to advance ’innovative digital public services’ (Näringsdepartementet 2015). Successful digital communication projects are often seen to stem out of critical features that are typical to innovation (Eccles 2015):

§  Encouraging trial and error;

§  A culture of constant innovation;

§  Based on user needs and evidence;

§  Questioning everything and looking for efficiencies and improvements, rather than the status quo;

§  Constant iterative improvement.

Government leaders generally recognize that digital technologies are disrupting the public sector and that they are struggling to keep up (Deloitte 2015).

In a rapidly changing and digitalizing world, incremental change development is not enough, not even for governments – what is needed is innovation (Steiber 2016).

Why are innovation teams used in the public sector?
As governments become increasingly interested in developing innovation strategies, the idea of setting up innovation teams has become increasingly popular.

Outcomes
Why is it important for governments to innovate in order to keep up with the mounting complexity and rapid change caused and enabled in to a large extent by the digital transformation of society? What does ‘keeping up’ mean for the public sector? The answers to these questions can roughly be divided in two groups:

Firstly, those who argue that experimentation, calculated risk-taking, and investment in developing new approaches can help government do its job efficiently (Burstein and Black 2014).

Secondly, those who argue that ‘public organisations have until now focused overly on using digitisation to generate efficiency gains, rather than radically improving services for citizens and businesses’ (Bason 2010)

The evidence from the practice of innovation teams gives support to both logics. Impacts ‘ranging from much more powerful digital user experiences for citizens, to more empathetic and outcome-oriented health services for patients, to precise ‘nudges’ that shape large scale behaviour while reducing costs to tax payers’ can be observed (Bason 2016).

Furthermore, practitioners in the public sector see innovation as ‘a means to address growing budgetary pressures, through more efficient administration or service delivery, and new societal demands, through different and more effective service design’ (Technopolis 2012).

In this context of dual outcome expectations on efficiency and improving services for citizens, it is important that public sector innovation teams clearly articulate on what outcomes they are measured. The focus on efficiency gains becomes especially delicate because the exact hierarchies that innovation teams are granted autonomy from are indeed designed to create maximum efficiency – and are considered to stifle innovation (Kotter 2014; Kattel and Karo 2016).

Organizational theory offers a broader perspective on the question of the goal of organizational change (Jones 2013):

‘The goal of organizational change is to find new or improved ways of using resources and capabilities to increase an organization’s ability to create value, and hence its performance.’

The value of public services and public service innovation cannot be measured simply by economic benefits (Wegener 2012). For example, Osborne et al (2016) refer to four ideal types of value in public service delivery:

§  meeting of an individual social need

§  meeting of community needs

§  individual well-being

§  social capital in an individual and/or community

The SMA has implemented and strives to work according to Lean principles, meaning that it has well-developed analysis of value creation in its processes. Policies at the SMA indicate that Lean principles are in place i.a. in order to advance innovation and development, and that the SMA should create value in the form of added value for the customer, quality and financial value. In the SMA’s Lean handbook emphasis is on added value for the customer, resulting in legally certain, simple, flexible and quick immigration process. (Migrationsverket 2013; Hellgren and Börje 2013)

Function
Typically, public sector organizations face significant barriers to innovation that hinder the organization from developing necessary innovation capacity and from exploiting opportunities. The organizational factors and opportunities that innovation teams are said to increase are:

Internal capabilities: The surge in innovation teams in the public sector can be tied to open and user-centred approaches, such as design thinking that focuses on user experience, frequent experimentation and multiple analytical perspectives. The competencies and mindsets needed for successfully using these approaches are not the same as those required for stable, daily operations and service at the front line. They aren’t even the same as needed for traditional, linear project design and ‘stage gate’ implementation. This means, for example, having a staff that is more willing to take some risks and is more willing to adapt new ways of organizing their work than the ‘regular’ staff. (Timeus 2015; Tonurist et al 2015; Bason 2010). Therefore, it is argued, these methods and skills are best used and fostered in a network-like structure (Kotter 2014) such as an innovation team.

Access to external knowledge: When public sector organizations start taking a more user- centred approach to innovation, they invariably discover that many other organizations play critical roles in people’s lives. User-centricity forces organizations to take a much broader, collaborative, and inclusive view of who needs to be part of the process of co-creating initiatives that will actually work in the real world. But public innovation that takes a user-centred and value-oriented approach is ultimately disruptive to the traditional government bureaucracy. It is severely challenging to the command-and-control logic of hierarchical organizations and to the linear logic of the policy-making process (Bason 2013), and better placed in an innovation team.

Organizational culture: Many claim that innovation teams seem to be particularly capable of creating their internal ‘culture of innovation’ within the lager environment of their organization. Innovation activities are vulnerable in public organisations, especially at department level where day-to-day managing of crises may take precedence over more strategic project work. Innovation activities may become a ‘nice to do’ rather than a need a ‘need to do. The creative processes are at risk of being caught in a stop-go process, teams fragmenting as key members are picked off for other pressing tasks. Innovation teams could stay consistent for the entire innovation process and operate in an environment where the innovation process is a professional discipline and not a rare, singular event, and where people can meet, interact, experiment, ideate and prototype new solutions. (Bason 2010)

Fully exploiting the opportunities of technology in government: Technology plays a central role in the formation of many public sector innovation teams. Many of the tasks innovation teams carry out are directly or indirectly related to developing digital solutions for the citizens as well as public sector. The adoption and diffusion of digital and the possibilities it has created, from participatory feedback mechanisms to using web analytics and big data, are central for many teams. Innovation teams are used as a conduit to start using technologies that are not yet fully grasped in the public sector. (Tonurist et al 2015)

Risks
Establishing an innovation team is not a guarantee of improving an organization’s innovation capacity. Literature and experience indicate there are many risks that have to be mitigated and pitfalls that have to be avoided to achieve greater likelihood of success. Innovation teams have been observed to suffer i.a. from (Tonurist et al 2015; Bason 2010; NESTA 2009)

§  Lack of connection to and understanding of core organisation;

§  Lack of backing from top leadership;

§  Perception of a privileged group of employees;

§  Difficulty documenting results;

§  Driven by marketing rather than change motives.

A key challenge is communication with the core hierarchy and showing relevancy of the innovation team.

To mitigate these risks, the innovation team should be based on genuine change motives, help employees in the core hierarchy become innovators, help identify problems and showing genuine curiosity and interest in the central activities of the organisation, as well as create strong and valuable connections with the mid-level managers that are so crucial for sponsoring real organisational change (Ibid.).

= Building and evaluation = This chapter continues to build on the insights from the problem formulation by showing how organizational design can help the SMA increase its innovation capacity in digital communication services. The four subchapters delve into the four main functions of public sector innovation teams; organizational culture, internal capabilities, access to external knowledge and exploiting the opportunities of technology. Building notably on insights from workshops at the SMA, the critical features of an innovation team are evaluated in relation to overcoming innovation barriers at the SMA (table 2) and seizing corresponding opportunities. Table 2: Innovation barriers identified during workshop at the SMA, October 2016. Table inspired by Australian Government 2010. (1) ‘The lean period’ signifies the years in the early 2010s when Lean government principles were first introduced and applied with apparent vigour at the SMA.

Organizational culture
According to workshop participants

‘The SMA focusses on short-term objectives and [exploitative] operational tasks.’

It’s time for the SMA to change its mindset and understand what digital service innovation is and what it requires from the organization. The employees and leadership need frames of mind and the proper courage and mentality to get things done!

Consciousness about innovation
During the workshop, the terminology of innovation in the public sector was introduced by me according to the definition from the problem formulation stage. At the time of the production of this thesis the word innovation was not actively used at the SMA, and indeed when searching for the word innovation at the SMA’s intranet I got zero hits. By discussing the overlap as well as the differences in the definitions of development and innovation, this thesis can contribute to give employees and managers at SMA a notion of in what way capacity building for innovation is different than capacity building for continuous development.

This was evidenced in one of the reflections from the workshop participants:

‘[…] perhaps it is even more difficult to define the word innovation than to implement innovation.’

The concept of continuous development is much more well established at the SMA than innovation. As introduced in the chapter on definitions, innovation is indeed partly different from continuous development work, and the SMA needs to be explicit about how. The Australian Government’s (2011) ‘Innovation 101’ offers a clear explanation, that the SMA can use to gain clarity on the issue, see box 1.

In order for management and employees at the SMA to gain clarity on what is expected, the SMA is well advised to introduce the word innovation into its planning. Despite the clear expectations from government, the mere introduction of the word is likely to face resistance internally. To quote one participant’s thoughts: ‘Is innovation just a more fashionable word for [continuous] development?’

Indeed, currently the SMA formally includes innovation activities under the umbrella of continuous development, using the term ‘completely new ideas for development’ (Migrationsverket 2016a) in a manner akin to innovation. In the prioritization of large scale development projects, these ‘completely new ideas’ are given the lowest priority, while maintenance projects are given the highest priority.

The workshop participants concluded that there is low awareness among leadership about what kind of organizational culture a creative or innovative organization requires, and how it is different from the current organizational culture. Bringing the issue of innovation and organizational culture on the agenda of the leadership would be a necessary step to influence the internal politics of the SMA.

This being said, it emerged during workshop discussions that while awareness is low, there are individual exceptions:

''‘Right now it is an exceptionally difficult time to put forth new things. But it is important that innovation work is embedded with the “right” persons so that you can get some pull from them.’'' (Quote from workshop participant)

Apparently there is a lack of champions of digital service innovation (table 1) but as expected from applied literature (see e.g. Tiesinga and Berkhout 2014), change agents – people who can be described as intraprenuerial auditors who are working to transform the way their institutions work from the inside – are to be found at different levels of hierarchy also at the SMA.

Courage of leadership and employees
Even when succeeding in bringing innovation onto the leadership agenda, group participants were doubtful whether the issue of budgeting for innovation could ever be overcome at the SMA. The fact that – apparently according to the financial director of the SMA -  all SMA development investments have to provide immediate financial return in the form of cost reductions was considered a big barrier to innovation. This is a reflection of a risk-averse organizational culture not considered favourable for innovation. In particular, because innovation involves exploring, testing, and sometimes failing in addition to sometimes succeeding. But success and failure are by definition unpredictable in innovation work. (Bason 2010)

Interestingly the workshop participants (with the exception of myself) did initially not voice risk aversion as an issue when it was understood as pertaining to the behaviour of individual employees. Perhaps one reason for this is that one of the official organizational values of the SMA is courage (Migrationsverket 2015a). One could argue that with courage should come the ability to take risks. The SMA needs to let the value of courage be extended to risk-taking in the context of innovation of digital communication services and let this value be reflected in budgeting practices that allow for failure. A further reflection on how the values of the SMA compare to those of select innovation teams in table 3. Table 3: Formal organizational values of the SMA (exploitative) compared to similar values at innovation teams (explorative).

A key insight for the SMA is to understand and embrace that in today’s digital communication projects the cost of risk is much lower than the SMA seems to assume, due to the fact that cheap prototyping and failing fast is cheaper than trying to avoid errors through bureaucracy – even in government (Christy 2016).

The current incapacity of top SMA leadership to speak about innovation, and thus make it a legitimate activity at the SMA was identified in the workshop as a barrier to innovation:

‘[Many managers and leaders have] an open mind to suggestions for solutions, but this is not well communicated to the employees.’

That leadership currently does not promote or prioritize development, innovation and creativity is further evidenced from messages from the director general on the SMA intranet stating that in the late autumn of 2016 ‘all focus lies on producing [immediate decisions in residency cases]’. To quote a workshop participant: ‘The leadership must talk about innovation because it is not in opposition to production, even though that seems to me to be the general perception [at the SMA] sometimes.’

It would seem from the point of view of advancing digital by default services that meet user needs that the SMA’s leadership is currently struggling for balance between exploitation and exploration. ''’There are big differences in what management wants and what employees feel and what conditions they operate in. It seems unfortunately that innovation is lower on the agenda and that the SMA doesn’t feel innovation produces value.’'' (Quote from workshop participant)

According to Bason (2010) visionary leadership is at the core of innovation. This entails an ability to formulate a compelling future and show in practice how it might be created. A top executive of a government agency such as the SMA is advised to ask the question: ‘Have I clearly extended a licence to innovate to the organization that I am heading, and do my actions show in practice that I mean it?’

Reflecting this, the workshop participants identified clear mandates for innovation as an interesting trend for the SMA to explore, and one that clearly presents an opportunity to balance the current situation ranging from explicit non-prioritization to confusion. If clear innovation mandates would be given, then the group envisioned the SMA as an organization where i.a. risk is tolerated to a greater extent than today.

But if employees are to ‘feel that innovation is part of my duties’, as expressed as a vision for the future during the workshops, then it is not enough that there is a clear mandate and supporting formal organizational arrangements. Another key is that colleagues and leadership show appreciation for the job done: Social influences are a powerful agent to spark innovative behaviour. Classical monetary incentives are not the primary driver to incentivize the intended behaviour – probably even less so in work for the public good (Bason 2010). But there is an inherent need for social validation from peer groups (Deloitte 2016). Celebrating successes is one commonly used way of achieving social validation. But in the most successful innovative organizations, failures are celebrated too. ‘Successful organisations reward success and acknowledge or celebrate failures, for example, by creating opportunities to openly discuss and learn from mistakes.’ (Martins and Terblanche 2003)

Team based mentality
Working in teams is a Lean government practice that thrives at the SMA, but according to my own observations only when a team is formed from within a unit. Teams in the Lean sense, that also would work in digital service development and include members from different silos are rare at the SMA.

Most work that crosses department lines is placed in different forums such as ’process networks’ where civil servants that usually work in their own silos meet to iron out digital communication project plans and progress. Such forums are actually not called for by any of the official project management approaches and techniques of the SMA. In all essence the SMA has bastardized the project management models, and while well-meant enhancements (Kotter 2014), academic analysis of government agencies show that such forums can easily turn into arenas for ‘issues of conflict, trust and a risk averse blame culture that prevents key stakeholders from building and fostering a collaborative and co-operative collective that can help project progress.’ (Berger 2007)

During the workshops, a frustration with these forums was raised: ‘The process networks perform to varying degrees of satisfaction’. Perhaps a tendency to blame culture and mistrust could be hinted in comments during the workshops such as:

‘Well, it is your unit not working well’

Compared to these forums the Lean teams hold considerable potential to build trust among team members across silos. This is explained by the fact that Lean places great emphasis on a team-based mentality, encouraging autonomy and giving individuals the environment and support they need to get the job done. Leadership is shared and practices such as iteration planning, daily stand-ups, and retrospectives help foster trust. (McHugh et al 2012)

Internal capabilities
Are processes that force a silo mentality and prevent change hampering the SMA’s chances of enabling digital by default services that meet user needs? In this chapter the current work processes of the SMA are evaluated and management practices that are better adapted to the innovation process emerge, ultimately creating value for the user. The emergent practices are a new way of working at the SMA that require a new way of combining skills.

Innovation process and management
At several instances workshop participants made the implicit assumption that the innovation process of digital communication services starts with an idea. Although there was some critical reflection around this, and that in fact user needs should be the point of departure, further reflections often return to the ’problem’ of ’placing the idea’ rather than generating ideas. Quoting a workshop participant:

’Entry level employees must be informed that new ideas are welcomed and where you can present these ideas. But it needs to work in practice otherwise the employees will become tired of engaging.’

The underlying notion is that ownership of the initial idea is transferred to someone else than the person(s) who initially came up with the idea. Furthermore, how ideas are generated is not covered by the current continuous improvement process at the SMA, but these are implicitly expected to spring from individuals as part of their every day work.

Among the workshop participants there was recognition that ‘the continuous improvement process doesn’t work’.

The SMA should revise its thinking around the idea being the first step in its innovation work. Basically all established public sector innovation process models start with user needs, with further steps before idea development comes into play (see eg. Image 4 Swedish Association of Municipalities and Regions 2015; Stockholm County Council 2015; OpenLab n.d.).

Formally there are two different project management methods at the SMA that are supposed to support innovation projects; stage gate XLPM and Lean government. The XLPM model is consistently used for larger scale development projects at the SMA, but according to most contemporary academic literature linear stage gate approaches are not considered conductive to innovation (see e.g. Brown et al 2014). My personal observation is that for mid and small scale digital communication projects often ‘no particular’ method is followed instead of the formally recommended Lean government.

The SMA’s IT department internally uses agile for its development projects – often siloed ‘orders’ that are a delivery stage of a broader digital communication project.

Further confusion about project management approaches is added by the fact that the SMA’s quality management department encourages the use of user centred design, service design and systems thinking. At several instances workshop participants raised the question about how all these approaches differ, overlap and interlink. This is explained in box 2.

Iterative agile processes are a success factor identified by most current literature on public sector digital service innovation and Lean government is an agile method (Brown et al 2014; Danneels and Viaene 2015; Government Digital Service n.d.). The SMA could choose to develop its Lean government method so that it is fit for use in the digital service innovation process.

Making Lean government fit for innovation in digital services indicates applying Lean principles to improve user experience so that it meets the needs of the users. The Lean UX approach by Gothelf (2013) suggests a four-step iterative process illustrated in figure 1:

Figure 1: The Lean UX process according to Gothelf (2013)

Contrasting to current approaches at the SMA, Lean UX presents the opportunity to have an approach that:

§  is not linear, just like the innovation process;

§  includes in depth research of user needs;

§  includes the declaration of assumptions about user needs early in the process;

§  creates testable prototype services (or minimum viable products MVP) potentially together with the user as ascribed by service design;

§  runs experiments where the MVP is tested with the user;

§  seeks feedback from the user on the MVP in order to improve the service in short sprints;

§  is explorative, just like the innovation process, and allows jumps back and forth between the different stages depending on what is discovered.

The main difference compared to Lean government is that the first step is not planning for what is to be done or for an output, but rather declaring assumptions about what is worth exploring and why (Gothhelf 2013).

As identified by workshop participants, this doesn’t completely solve the issue of who decides which assumptions are taken forward into the next step of the Lean UX process.

But by requiring the employees to be more clear about assumptions and expected outcomes for the user, they are forced into more reflection on their ideas.

In fact, Bason (2010) identifies outcome focus as key innovation driver in the public sector: Focusing on outcome based results is a starting point as well as a way of measuring results. The SMA has well developed outcome objectives in the form added value for the user, see chapter 2.2.1, and starting with them in combination with digital by default can shape an outcome focussed innovation approach.

One workshop participant said: ’Even if the employees at the SMA can come up with creative solutions and improvements, often these ideas do not reach leadership and therefore the improvements may not happen and the level of innovation is low.’

This thought doesn’t elaborate on in which format and at what stage of development the ideas or improvements reach the leadership. But it is perhaps not surprising that senior leadership doesn’t have an interest for new ideas if their value is not somehow evidenced when presented to them. After all:

‘Everyone who's ever taken a shower has an idea. It's the person who gets out of the shower, dries off and does something about it who makes a difference.’ – Nolan Bushnell

Before an idea is taken to senior leadership it should have ran through the Lean UX process, probably had a few iterations, while documenting insights that support evidence of potential in reaching outcome goals. Really it should be only one step short of an innovation.

People and cross-functional teams
The workshop participants identified educational, recruitment and awareness raising efforts related to the implementation of new methods of working (such as Lean UX) as a potential pitfall, due to the fact that it can be a massive effort to introduce new processes, methods and tools in the whole organization of currently about 8.000 employees.

But as emerged in later discussions, it is the employees working with innovation of digital services at the SMA who need the knowledge about how creative and innovation processes and related methods and tools work. Not all 8.000 employees need this knowledge.

Even in an innovation team not all members need the full knowledge to start with. Quoting a workshop participant'': ‘I think the most important steps for the SMA is to disseminate knowledge [to employees about how to work with innovation and development] and in that context to work in cross-functional teams. It gives us the opportunity to educate employees theoretically but also try things out in practice.’''

During the workshop it emerged that

‘[the SMA’s] organization makes it difficult to be innovative as innovation requires cross-pollination of competencies.’

The participants identified cross-functional teams where users can be directly involved, along with employees with different competencies from different parts of the SMA as an interesting opportunity for the SMA to explore.

A step along the way supporting such teams would be to dedicate employee working hours to innovation work. On the details on how to gather such a team, a suggestion from the workshop participants was to dedicate a cross-functional team to the exploitative ‘core’ functions of the SMA, and let them work together with the civil servants there. Kotter (2014) and Bason (2010) point to the fact that borrowing volunteers from the core exploitative hierarchy into the innovation team is a key feature of successful innovation teams.

In addition to the skills of project management (Lean UX), and the volunteer experts from the core – that in the case of the SMA are experts on the immigration process – Bason suggests that social research experts, eg. ethnologists that can help understand user needs, should form the core of a cross-functional innovation team.

INSERT IMAGE

Some skills in e.g. user experience, lean, agile, and service design exist at the agency, but they are mostly in siloed units and are experiencing apparent difficulties in cooperating flexibly.

Furthermore, there is a well founded fear that any outputs from the innovation team – even when they meet the needs of the users - won’t find a stable use or a home. That worry is founded on the current problems of the idea management process:

''‘The communication channels don’t work that well. There are initiatives and ideas that don’t seem to reach further than the closest [colleagues].’''

The team must have an understanding of the innovation process, including how to scale an innovation at the SMA (Bason 2010). More on best practices for scaling in chapter 4.

Finally, during the workshop the notion that the physical space can support creative and innovative behaviour was discussed. It was suggested that the physical space can support the quick gathering of cross-functional teams if spaces are easily available for these teams to meet in. Furthermore, due to the importance of visualizations in user centred design tools, the physical space can potentially be used for better visualizations, as evidenced during a visit to the The Swedish Labour service’s digital innovation centre (image 5).

Image 5: A physical space at the Swedish labour service dedicated to visualization of a user centred design process. (photo credit: Arbetsförmedlingen)

Access to external knowledge
Another aspect of innovation teams is to work directly with users – with the SMA’s applicants and stakeholders. The SMA wants to work in a user centred manner, but is struggling with establishing the necessary practices. This chapter builds and evaluates organizational features that can help the SMA stay close to people outside the organization.

User needs and user involvement
The disruptive nature of working in a user-centred manner and how challenging it can be to work in this way in a hierarchical organization (Bason 2013) was clearly evidenced during both workshops. In addition to the lack of employee skills discussed in the previous chapter, the group also identified issues such as organizational introversion and a reluctance to let go of previous ways of working (Table 1).

‘We think we know [what the users’ needs are.]’

‘There is a risk that the SMA will relax and trust in old insights that are taken for the truth. Reality is not always what you think it will be.’

These quotes from the workshop indicate the SMA is facing a real challenge in accessing necessary external knowledge in the form of user needs.

But there is much work already done that the SMA can build on: For example, the SMA’s quality assurance department has recently launched a tool box and also some requirements on when and how to understand user needs when working with service design.

Service design has an explicit co-design approach, involving the user in the actual development, something that is absent from the SMA’s Lean government method. This reinforces the confusion on methods already mentioned in chapter X. One quote from the workshop was that user needs run the risk of becoming ‘a mantra that no-one believes in’ and another participant said ‘there are prejudices against working with user needs.’

One can conclude then that the SMA needs to be even more explicit on why, when and how to consider user needs in the innovation process – and Lean UX gives a clear answer to this, see chapter 3.2.1.

A key question is also who the vehicle of user needs is – is it the employee ‘through the eyes of the user’ of is it the user him or herself through direct user involvement? Or is it all about quantitative data? In this context it is worth reinforcing the SMA’s Lean manual’s message that user needs should be understood in depth at every step of the Lean UX process, they are not something that can be based on the assumptions or what any individual employee thinks the user needs are. Or as expressed by a workshop participant:

’With empirical material backing you up it is always easier to start new innovations. Meaning that it’s not only I or someone in the organization that thinks something.’

One workshop participant took an optimistic twist on the mental challenges of working with user needs: ‘We should identify the prejudices and misunderstandings about working in a user needs oriented way and show through examples that it is not at all so difficult.’

External partnerships
It is likely that innovation in digital services regularly requires involving other organizations in order to address all the issues facing the user when trying to achieve a particular outcome. But the major challenge that this cooperation poses was heavily emphasized during the workshops, with the difference in frameworks, lack of centralized governance structures, and the need to work in project-based formats.

Much of the thinking in the workshops, apparently influenced by the current standard way of working across government agencies in Sweden when developing new services, focussed on project organization. Every new cross-governmental ‘life event’ project is given a new project organization, where new balances have to be struck in host of areas, such as terminology, resources, and priorities, as exemplified by the workshop participants.

The workshop participants called for ‘a method to involve others’ in innovation work, including a management model for cross-governmental agency processes.

Just as in the digital work crossing internal agency silos, it is essential that cross-agency work doesn’t get bogged down in command and control forums like the ones described in chapter 3.1.3. A team that works together throughout the Lean UX process is key even when involving employees from other organizations.

Three particular challenges emerge:

How to avoid long negotiations before a project between government agencies?

Committing an employee to a core team will indeed involve up front negotiations on time frames and an understanding on what outcome is expected from the innovation work. But instead of negotiating a completely new project organization upfront, some government agency employees simply temporarily join existing innovation teams in other organizations without any strong formal agreements. An example is the Swedish Labour Service’s employees joining Spotify’s innovation week.

Relating to this, it is notable that there have been digital design and innovation challenges in Sweden focussing on refugees’ needs (UNHCR 2016). These have not had any proactive involvement of the SMA, and so their output has not directly contributed to the outcomes that the SMA is working to produce. An innovation team would be advised to mobilize the skills that exists among civil society and businesses to contribute to the goals of the SMA.

How to avoid that the team doesn’t expand and become too big?

Involving other organizations in innovation doesn’t have to be more complicated than establishing a core team for a co-creation project. A key success factor is to keep the core team small, and have a broader group of stakeholders that become involved in the process as it unfolds (Bason 2010).

How to avoid being hindered by data protection legislation?

Much of the work done at the SMA involves accessing potentially classified personal information, and there are notable difficulties in transferring this information even from one government agency to another. In this context the innovation team can become a space where working with external partners is standard modus operandi and with the courage to be open to collaboration while respecting data protection legislation. This probably includes finding ways of prototyping without the need to access classified information.

Fully exploiting the opportunities of technology in government
The SMA has digitialization related strategies in place, and workshop participants implicitly seemed to agree that there is further unexploited opportunity in this field. The technology solutions that were briefly mentioned during the workshops were open data and automation. Not much focus was given to discuss these per se, as we dived into organizational issues. But the limiting effect of legal regulations on exploiting these technologies were mentioned several times and with emphasis. Quoting a workshop participant reflecting on the contents of table 2:

‘It strikes me that some parts of the rules and processes will be of more significance than other trends, e.g. tougher regulations and less state funding.’

The requirements coming from central government were perceived as conflicting; more transparency is expected, but stricter privacy rules are introduced. Indeed, new EU rules on data protection will have some impact on government use of cloud services and to some extent social media (Gorge 2012).

On the other hand, security requirements are nothing new to government, and there are valuable experiences at hand to learn how to best deal with them. The UK’s Government Digital Service (n.d) suggests that security issues should be seen as part of improving the user experience of digital services, with the hypothesis that users won’t use your service unless you can guarantee it’s confidential. This has consequences for how the organization approaches security issues in its work. The whole team working in a digital communication project needs to think about security in a proportionate way. To cite James Stewart of the UK’s Government Digital Service (Turnbull 2014):

‘All too often, it’s been the case that people have approached security as something that either people who deal with compliance and writing documents deal with, or that the techies deal with. It’s a fundamental part of the service; it’s not this separate thing that one team thinks about […]’

Academically assessed examples show that information security issues can be incorporated into the models similar to Lean UX  by (Al-Hamdani 2014):

§  framing – in particular stating assumptions about – information security;

§  tackling information security as part of collaborative design;

§  creating MVPs that take information security into consideration in a balanced way;

§  and experimenting with them, getting feedback and doing further research from an information security perspective.

= Reflection = The workshops explored innovation barriers and trends in an open manner, without an explicit reflection on how they relate to the concept of the ambidextrous organization. So this begs the question: How to connect the innovation team to the exploitative core hierarchy of the SMA?

In each component of organizational design there are critical features that are known to help scaling innovations to the core hierarchy (What Works Scotland Evidence Bank 2015; NESTA 2015a; Brown et al 2014; Labour Digital Government Review 2015; Kotter 2014):

§  Focusing on outcome goals: The innovation team should align outcome goals with those of the SMA in the form of added value for the user.

§  The team draws on expertise from the core in the field of the immigration process, volunteer experts from the hierarchy and other stakeholder organizations are part of the innovation team, as well as the team being borrowed to particular functions of the organization. Not only is the innovation more likely to succeed when it builds on expertise, but volunteers from the hierarchy can help build trust between hierarchy and innovation team. This requires clear, direct and frequent communication between all stakeholders, often in the form of informal contacts with the hierarchy.

§  Being as open as possible with the innovation, including showing through examples how the work is done, sharing insights that support evidence of potential in reaching outcome goals, and openly discussing and learning from mistakes, as well as celebrating success and failure together with the hierarchy.

§  Procurement focus not on monolithic contracts that create system lock-ins, but rather on procuring the right knowledge and services to be able to build digital services based on open standards.

Visionary leadership applies to the senior leadership of the whole ambidextrous organization, not to only to the innovation team where the leadership is shared among team members.

Can the top leadership be convinced that innovation and production, exploration and exploitation, can exist side by side at the SMA, that they mutually reinforce each other, and that there is a need to communicate this duality to employees? This type of rhetoric would clearly reflect the ideas underpinning the ambidextrous organization (O’Reilly and Tushman 2004).

When I presented the identified critical features of innovation team to the SMA’s communication management board reactions were mixed. While the idea of working iteratively, autonomously, and cross-functionally was welcomed as ‘obvious’, the necessity to involve employees from across departments became a barrier for this siloed management board, highlighting the need to reach out to management structures where senior leadership across departments are represented.

Validation of digital by default and meeting user needs
The workshops mostly explored achieving digital by default and meeting user needs through the proxy of capacity building for innovation, assuming – based on the problem formulation – that increased innovation capacity can achieve digital by default and user needs. But looking directly at the enablement of digital by default and meeting user needs, can the identified features be validated?

Approaching digital by default as an outcome goal enables testing along with validation of the service innovation’s achievement of SMA outcome goals in the form of added value for the user.

The outcome goals in the form of added value for the user are in themselves a strong driver for meeting user needs. A second feature that reinforces this is the use of project management models where user needs are included at every stage and where the user him or herself can be directly involved.

Thorough experience validates the enablement digital by default and meeting user needs with the help of these critical features - the outcome focus, and the integration of user needs throughout the Lean UX process (see eg. Open Lab n.d; SKL & VINNOVA 2016; Bason 2010; Brown et al 2014).

= Formalized learning =

Critical features of a digital service innovation team at the SMA
In the building and evaluation stage, as well as in the reflection, a number of critical features of what the organizational design supporting digital communication services at the SMA might look like have emerged. These critical features are presented in the table 4. Table 4: The four organizational components and their critical features for an innovation team at the SMA. Those features marked with * indicate a feature with a connection to the hierarchy

Summarizing these critical features in a stylized fashion, allows us to answer the main research question of this thesis:

How might an innovation team enable the Swedish Migration Agency to build communication services that are digital by default and meet the needs of the users?

An innovation team in this thesis is defined as a network-like structure parallel to the traditional hierarchy designed to overcome barriers to innovation and build innovation capacity in the public sector. In the context of the SMA an innovation team could enable user-centered innovation in digital communication services by:

1.          Using the SMA’s outcome goals in the form of added value for the user as an innovation driver, but also by having the outcome as the starting point, by formulating assumptions about how an outcome could be reached with the help of a new or radically improved digital communication service;

2.          Developing and combining skills flexibly in a cross-functional team, notably including skills in Lean UX project management, social research, the immigration process, relevant skills from other organizations, and the user;

3.          Engaging in work that is exploratory, requires risk-taking, and that to a large extent cannot be planned upfront in any detail;

4.          Having strong formal autonomy and an organizational space, including budget, time and mandate, within which to operate;

5.          Using a project management method that is outcome driven, iterative, involves the user, and supports the innovation process – the suggested model is Lean UX;

6.          Improving consistency in accessing external knowledge, by creating a team where understanding user needs in depth, involving users and external partners in the actual development of a new service is a standard way of procedure;

7.          Fostering an open team based mentality that reflects the SMA’s values and builds strong relations with senior leadership.

This contrasts with the dominating features of the SMA’s current organizational design that:

a)          engages in work that is mostly about predictable service production, continuous improvements, risk-minimizing, and relies heavily on planning;

b)         outputs that are presumed to produce desired outcomes are planned upfront, a way of working that is not conductive to advancing innovation.

c)          consists of a variety of individuals with different skills, but whose variety of skills rarely meet in teams that stay put throughout a project, and who lack the tools or agreement on how, when and why to include external partners and users in the work;

d)         is based on strong hierarchies, organizational silos, control mechanisms, and where budget, time, and mandate for innovation is usually very low;

e)          uses a variety of project management methods, including a stage-gate based approach that isn’t iterative, and some of which have been misunderstood, and their interrelations create confusion;

f)           fosters team based mentality within units, promotes the SMA’s values, and gives preference to formal management hierarchies.

With the possible exception of bullet point e) above that highlights the confusion around project management methods at the SMA, it is probably not desirable for the SMA to shift critical features of its hierarchy’s organizational design towards the rather different critical features identified for an innovation team.

This speaks in favour an ambidextrous organizational design for the SMA, where the hierarchy is left intact and can continue to produce reliable outputs with increasing efficiency. In parallel an innovation team is introduced with its own organizational design features.

Critical features that help connect the two systems in the case of the SMA’s hierarchy and a suggested digital innovation team can be summarized as:

8.          Volunteers from the core inhabit the innovation team, and the innovation team inhabits core units;

9.          Leadership of the core and the innovation team is ultimately the same visionary leadership integrated into the existing senior management hierarchy. It is a leadership that is capable of communicating the dual operating system to the entire SMA;

10.     The team being open in all stages of its work, celebrating successes and failures together with the core hierarchy, and in particular reporting openly and as clearly as possible about its progress towards outcome goals.

= Critical and ethical discussion = The proposed organizational design of an innovation team at the SMA is an educated best guess of how the design, implemented in certain ways in the particular cultural and organizational environment of the SMA, will affect the ability of people to perform the kinds of work required to meet the needs of users through digital by default communication services.

‘There are never any guarantees that people will act according to plan or that unforeseen external events won't necessitate a shift in strategy— in fact, they probably will.’

(Nadler & Tushman 1997)

With this insight in mind, the concept of an innovation team that has been developed in this thesis, should be treated for what it is: A concept that has been validated through action design research, but that has not been implemented and thus not evaluated in terms of actual results. The concept is based on a review of best practices, academic knowledge, as well as the situated knowledge of practitioners at the SMA. What it is not, is a guarantee of success, but rather a concept that is argued to have a higher likelihood of succeeding in enabling the building of digital by default communication services that meet user needs than the continued status quo at the SMA.

Despite these limitations, the suggested organizational design shifts of this research project should be seen in a larger societal context. An innovation team at the SMA would constitute a small step in the movement away from public administration through the principles of control and excessive governance typical to New Public Management and constitute a step towards a trust-based governance, characterized by less output objectives, rules and control and more trust in civil servants. While this type of governance is only emerging in the public sector, it is sanctioned i.a. by the Swedish Agency for Public Management (Statskontoret 2015).

The proposed innovation team at the SMA also implies more in-house development of digital services and less use of external contractors. Again, this reflects a movement away from the still dominating principles of New Public Management towards an emerging governance model sometimes labelled Digital Era Governance.

While Digital Era Governance is considered to make possible better outcomes for the user, it has consequences for other individuals and organizations that may not be equally positive for everyone involved. For example Mergel (2016) points to the experience of public innovation units in the USA that have been ’criticized by the private sector witnesses for their opaque operations, vague agreements, and that they introduce an agile delivery purchase process that not all vendors and contractors want to follow due to intellectual property right protection.’

While opaque operations and vague agreements are certainly not morally acceptable, the fact that some operators are not willing to adhere to open standards in public procurement for the benefit of the development of public goods is not a reflection of bad governance. On the contrary; openness is a moral value at the centre of good governance (see e.g. Open Government Partnership 2015).

Ethics
The proposed organizational design for an innovation team at the SMA is akin of a decentralized organization structure that includes an important delegation of authority and individual responsibility. Such decentralized organizations have been observed to sometimes deviate form organizational objectives and serve their individual interests. If employees in the innovation team misinterpret their autonomy this may lead to immoral acts by the employees. (Weiss 2014)

Other studies show the opposite: Strict hierarchies can make employees passive in terms of moral judgements in their work. Increased autonomy for the employees then correlates with more active and hence better moral judgements. (White 1999)

In fact, the tension of the exploitative and explorative functions is also an ethical tension. The civil servant’s professional standards imply a need for stability, but innovation requires change. Pragmatism is one possible answer to this dilemma and instructs professionals to ’cease searching for a single, immutable principle and jump in, think, then act’  (Andrews Emison 2010). In a pragmatic approach judgment and action co-evolve. Interestingly, the Swedish code of conduct for the civil servant (Regeringskansliet 2015) says that

’(s)he who is very powerful needs to be very kind.’

The value of kindness towards others is well in line with the value of empathy that both the SMA and the Lean UX approach embrace. An autonomous group like an innovation team at the SMA then has a better chance of pragmatic moral development while adhering to the value of empathy than the law-and-order mentality of the conventional hierarchy.

Alternatives to innovation teams
Should the SMA choose that it doesn’t want to introduce an ambidextrous organization, there are other avenues that it may want to explore in order to increase innovation capacity. These alternatives include (Burstein and Black 2014; Muller-Krogstrup 2011):

§  Improved employee driven innovation

§  An innovation and leadership training programme for selected staff

§  Membership in organization that promote knowledge sharing around government innovation

§  Changes in recruiting practices to attract different skill sets

§  Public-private partnerships

§  A host of other programs and projects

= Starting small: A suggestion for the SMA’s leadership = Setting up an innovation team at the SMA with the critical features identified in this thesis does not have to entail a lot of complicated formal decisions. Some organizational shifts are necessary though. I propose a few recommendations on how leadership can tackle these shifts in the short term when setting up a first innovation team at the SMA with the purpose of building communication services that are digital by default and meet user needs:

Shift 1: No more production only, but production, development and innovation

Current leadership, preferably the director general, or a combination of senior managers from those departments that are key for digital communication services innovation (communication, IT, development, quality assurance, HR), make a clear statement on why the SMA needs to work with innovation in digital services, what is understood by innovation, and how it relates to the core tasks of the SMA. Here is a suggestion of what such a statement could look like:

''Innovation is necessary in order for the SMA’s communication to meet user needs. In the field of digital communication, production and continuous development need to be complemented by a small but focussed effort to enable new implementable communication services that provide added value to the user. The need to not only build on existing digital communication services, but to also build entirely new ones, is highlighted by the opportunity to take advantage of the rapid digital transformation of society, and public sector digital innovation is expected by the government of Sweden.''

''Innovation work is different in nature than production, and to some extent also from continuous development, but does not stand in opposition to either. By exploring, testing and seeking feedback on new ideas, we can find new services that help the SMA reach its outcome goals in the form of added value for the user.''

''To attain an organization that is capable of digital service innovation we must make sure that we have the right capabilities, organizational culture, access to external knowledge and can exploit the opportunities of technology. Therefore, we are starting an innovation team for digital communication services that will include volunteer experts from core operations and where the team can be dedicated to core functions.''

Shift 2: Shifting from outputs to outcomes

The current leadership needs to select an outcome goal for the innovation team’s first project based on the added value for the user (the applicant for residency permit, asylum or citizenship). Formulating a challenge for the innovation team is a challenge in itself, but fortunately leadership can find support e.g. from the Swedish state innovation agency Vinnova’s guide on innovation challenges (NESTA 2014). This could take the form of a challenge statement such as:

How might we enable a faster residency application process aided by a new or radically improved digital by default communication service that meets the needs of asylum seekers?

This is a quite broad challenge, and the feedback from the SMA’s communication management board suggests that it should be narrowed, however without tripping into the pitfall of defining a desired output, as is now often the case with digital communication development activities.

Leadership should also give mandate, budget and time frames, including setting a deadline for delivery, e.g. one month for a minimum viable product.

Based on these outcome goals and initial frames the team builds hypotheses they would like to test and the team then prioritizes them based on risk, feasibility and potential success, as suggested by the Lean UX method (Gothelf  2013).

Shift 3: Moving from limited roles to collaborative capabilities

Find a change agent at the SMA with the potential of being a visionary leader and who is interested in innovation of digital communication services. Make this person the assembler of the innovation team. Perhaps let employees put themselves forth for this task or suggest colleagues that they find suitable for the task.

The SMA should explicitly state that it expects, allows and encourages innovation team members to use their full range of skills, talents and competencies. In an otherwise formal organization like the SMA, roles are often very strict, meaning conversations across silos and disciplines often include mistrust and finger pointing. By not only using their core competencies, but also secondary competencies that can make the team work better. Also this allows for natural conversations across disciplines, thus creating a climate of trust.

Shift 4: Creating small, cross-functional teams

Identify the skills needed for assembling the team, which of these skills are available at the SMA and which that are missing. Identify where the missing skills can be found among volunteers in civil society and business who would be interested in contributing. To work in a credible manner with service design, the team should throughout the process involve a user in the development of the new service.

The innovation team should be a small team as smaller teams work better than large teams, and experience from successful large organizations suggests breaking down big teams into smaller teams of about 5 to 10 persons (Gothelf  2013). While the outcome goals are likely to be very broad, it is up to the innovation team to break down the problem and work on a hypothesis that is feasible in the context of the team size.

Shift 4: Eliminating big design up front and navigating documentation standards

Set the innovation team about working in the Lean UX method and let them retain autonomy throughout the innovation process. Do not expect the team to provide plans on their work or outputs upfront.

The innovation team should work in an iterative method; this thesis has suggested Lean UX i.a. because of its adaption to the innovation process and the linkages to Lean government that the SMA already partly works in. Furthermore, Lean UX is well suited for work beyond software development, with strong service design underpinnings.

The Lean UX approach includes not having to wait for complete documentation from any previous phase, but quickly moving into prototyping and testing mock-ups. This is significantly different than the XLPM stage-gate approach in use elsewhere at the SMA, with its heavy requirements on documentation. It is also different from the way ideas management has been implemented at the SMA which includes shuffling ideas up the hierarchy before having any possibility to start testing them.

There will be a need for documentation no doubt, in particular in relation to regulatory compliance. Here, the Lean UX approach suggests to ‘lead with conversation, and trail with documentation.’ (Gothelf  2013) This means that the IT security balances that will need to be struck should be part of the problem solving in the early stages of the project life cycle, not something that comes in at the end when a few security experts evaluate what has been done. This means that legal and security will often need to be part of the innovation team expertise.

Shift 5: Creating open, collaborative workspaces

The innovation team should set up offices in a clearly visible space at the SMA and should engage in dissemination activities at the SMA, eg. during a weekly informal update ‘coffee and cake’ open for anyone interested at the SMA.

The SMA should co-locate the innovation team members and offer them a space that keeps them visible to each other and colleagues in the organization, but also allows them to visualize their own work.

This may prove to be challenging considering the SMA is devolved around more than 20 locations around Sweden. Most of the core competencies for digital service innovation are to be found in Norrköping where the headquarters is located, and in Stockholm. Mimicking what a digital innovation team did at Scandinavian airlines (exhibit 1), the SMA’s innovation team could set up highly visible temporary offices in either Norrköping or Stockholm with a clear invitation for employees to pop in.

Image 6 Innovation team room at SAS.

Shift 6: Procurement and working with third-party vendors

The SMA needs to look further into how it can combine outputs from its innovation team with the likely need of having to procure some parts of the solutions. It is outside the scope of this thesis to look at procurement practices in detail, but there is expertise and best practices that the SMA should take advantage of in this field: e.g. recommendations from the Swedish Competiton Authority (Konkurrensverket 2016) and the National Agency for Public Procurement (Upphandlingsmyndigheten 2016). This is a clear avenue for further research.

Shift 7: Reaching up and out

At the end of the set time period, leadership should hold the innovation team accountable through quantitative data and user testimonies for what is known about how its minimum viable product performs towards the outcome goal. If it presumably performs well enough, then it should go through a policy check, before scaling it.

The innovation team must be allowed to pursue its own agenda in order to discover features that will best service the user. But this means that the team also must constantly communicate its agenda to members of the hierarchical organization, so that they are aware of what is being explored and developed. The innovation team must reach out to different departments in the hierarchy proactively and cannot ignore them. They will be far less resistant to the changes the innovation team’s digital services are making. (Gothelf 2013)

Final reflections
This research project has relied on the theory of the ambidextrous organization. In his book ‘Accelerate’ John Kotter (2014) combines the theory with his change management approach ‘the eight accelerators of change’. Change management is vital, as the introduction of an innovation team constitutes the type of change that the SMA knows from previous experience can lead to e.g. (Migrationsverket 2015c):

§  Negative reactions from employees when the current organization is questioned.

§  Internal politics and prestige becoming a barrier to change.

The SMA has some tools for change management of its own that are based on Lean government (Migrationsverket 2013). Nonetheless, should change agents and/or leadership at the SMA decide to further pursue the implementation of the concept of a digital innovation team, one is well advised to have a verified change management strategy. Further research on openness and resistance to change, what factors underpin the obstacles and openings to change at the SMA, and how they can be managed to create the change that is necessary for the introduction of an innovation team at the SMA.

Image 7:  The change curve by Migrationsverket (2013)

I will place this text online in a wiki so that anyone who is interested can continue to build, evaluate and reflect on the critical features of an organization design for a digital service innovation team at the SMA.

= Bibliography = Al-Hamdani, W.A. (2014) Design Thinking Approach in Teaching infomration Security,.

Andrews Emison, G. (2010) ‘Ethics of Innovation for Public Service Professionals’, The Innovation Journal: The Public Sector Innovation Journal, Volume 15(3), 2010, article 2,.

Arbetsförmedlingen (2016) Från ego- till eko-system.

Australian Government (2011) Innovation 101. Available at: https://innovation.govspace.gov.au/innovation-101/ (Accessed: 22 October 2016).

Bason, C. (2010) Leading public sector innovation: Co-creating for a better society. Portland, OR: Univ of Chicago Pr.

Bason, C. (2013) ‘Design-led innovation in government’, Stanford Social Innovation Review, Spring 2013.

Bason, C. (2016) ‘The Rise of Innovation Labs: Three Stumbling Blocks Along the Yellow Brick Road’, The Journal of Design Strategies, 8(1).

Berger, H. (2007) ‘Agile development in a bureaucratic arena—A case study experience’, International Journal of Information Management, 27(6), pp. 386–396. doi: 10.1016/j.ijinfomgt.2007.08.009.

Bradbury, H. (2013) The sage handbook of action research: Participative inquiry and practice. Edited by Peter Reason and Hilary Bradbury-Huang. Thousand Oaks, CA: SAGE Publications.

Brown, A., Thompson, M. and Fishenden, J. (2014) Digitizing government: Understanding and implementing new digital business models. Basingstoke, United Kingdom: Palgrave Macmillan.

Brown, Culliton, Robb and Staszowski (eds.) (2016) The Journal of Design Strategies, 8(1).

Burstein, R. and Black, A. (2014) A Guide for Making Innovation Offices Work. IBM Center for Business and Government.

Cannaerts, N., Segers, J. and Henderickx, E. (2016) ‘Ambidextrous design and public organizations: A comparative case study’, International Journal of Public Sector Management, 29(7), pp. 708–724. doi: 10.1108/ijpsm-12-2015-0210.

Christy, A. (2016) ‘Government Goes Agile’, Stanford Social Innovation Review, Spring 2016.

Danneels, L. and Viaene, S. (2015) ‘Simple rules strategy to transform government: An ADR approach’, Government Information Quarterly, 32(4), pp. 516–525. doi: 10.1016/j.giq.2015.09.006.

Deloitte (2015) The journey to government’s digital transformation. .

Deloitte (2016) Cultivating innovation at work. Available at: https://dupress.deloitte.com/dup-us-en/deloitte-review/issue-18/cultivating-innovation-at-work.html (Accessed: 12 November 2016).

Eccles, D. (2015) How do you make an organisation ‘digital by default’? Available at: http://numiko.com/blog/how-do-you-make-organisation-digital-default (Accessed: 23 November 2016).

egovlab (no date) Vision and values. Available at: http://www.egovlab.eu/index.php/en/about-us/vision-and-values (Accessed: 19 November 2016).

Eknomistyrningsverket ( The Swedish National Financial Management Authority) (2015) Regleringsbrev 2016 Myndighet Migrationsverket - Ekonomistyrningsverket. Available at: http://www.esv.se/statsliggaren/regleringsbrev/?RBID=16910 (Accessed: 23 October 2016).

European Commission (2016) LAB CONNECTIONS policy labs in Europe, for Europe. Available at: https://blogs.ec.europa.eu/eupolicylab/lab-connections-policy-labs-in-europe-for-europe/ (Accessed: 28 September 2016).

Experio Lab (no date) Experio lab. Available at: http://experiolab.se/en/about-us/ (Accessed: 19 November 2016).

Gorge, M. (2012) The implications for storage of EU data protection regulation. Available at: http://www.computerweekly.com/feature/The-implications-for-storage-of-EU-data-protection-regulation (Accessed: 9 November 2016).

Gothelf, J. (2013) Lean UX: Applying lean principles to improve user experience. Edited by Josh Seiden. Sudbury, MA, United States: O’Reilly Media, Inc, USA.

Government Digital Service (no date) Service manual. Available at: https://www.gov.uk/service-manual/ (Accessed: 4 November 2016).

Hellgren, A.H. and Börje, M.D. (2013) Implementering av Lean på Migrationsverket (Implementation of Lean at the Swedish Migration Agency). Lunds universitet. .

Hyper Island (no date) Future trends. Available at: http://toolbox.hyperisland.com/future-trends (Accessed: 30 October 2016).

Jones, G.R. (2013) Organizational theory, design, and change. 7th edn. United Kingdom: Pearson Prentice Hall.

Kattel, R. and Karo, E. (2016) ''Start-up Governments? Blogpost''. Available at: http://lipse.org/blogpost/item/4 (Accessed: 13 September 2016).

Konkurrensverket (2016) IT-upphandlingar ger inlåsningar och begränsar konkurrensen. Available at: http://www.konkurrensverket.se/nyhetsbrevsartiklar/it-standarder-inlasning-och-konkurrens/ (Accessed: 10 December 2016).

Kotter, J.P. (2012) Leading change. Boston, MA: Harvard Business School Press.

Kotter, J.P. (2014) Accelerate: Building strategic agility for a faster-moving world. Boston, MA, United States: Harvard Business Review Press.

Labour Digital Government Review (2015) ''Changes. How procurement and needs have evolved''. Available at: http://digitalgovernmentreview.readandcomment.com/creating-better-outcomes-by-building-digital-partnerships/changes-how-procurement-and-needs-have-evolved/ (Accessed: 19 November 2016).

Martins, E.C. and Terblanche, F. (2003) ‘Building organisational culture that stimulates creativity and innovation’, European Journal of Innovation Management, 6(1), pp. 64–74. doi: 10.1108/14601060310456337.

McHugh, O., Conboy, K. and Lang, M. (2012) ‘Agile practices: The impact on trust in software project teams’, IEEE Software, 29(3), pp. 71–76. doi: 10.1109/ms.2011.118.

Mead, G. (2002) ‘Developing Ourselves as Police Leaders: How Can We Inquire Collaboratively in a Hierarchical Organization?’, SystemicPracticeandActionResearch,Vol.15,No.3,June2002,.

Mergel, I. (2016) Digital innovations in the public sector. Available at: https://inesmergel.wordpress.com/ (Accessed: 3 December 2016).

Migrationsverket (Swedish Migration Agency) (2013) Leanhandbok (Lean user manual). .

Migrationsverket (Swedish Migration Agency) (2015a) Migrationsverkets uppförandekod. Available at: http://www.migrationsverket.se/download/18.2d998ffc151ac38715911f96/1472031023400/Migrationsverkets+uppf%C3%B6randekod.pdf (Accessed: 10 November 2016).

Migrationsverket (Swedish Migration Agency) (2015b) Redovisning av verksamheten. Available at: http://www.migrationsverket.se/Om-Migrationsverket/Styrning-och-uppfoljning/Redovisning-av-verksamheten.html (Accessed: 23 October 2016).

Migrationsverket (Swedish Migration Agency) (2015c) Slutrapport Projektet helhetsinriktad utveckling och styrning (HuS). .

Migrationsverket (Swedish Migration Agency) (2016a) Projekt är utveckling (Projects mean development). .

MindLab (no date) About MindLab - MindLab. Available at: http://mind-lab.dk/en/om-mindlab/ (Accessed: 28 September 2016).

Modig, S. (2016) Innovations- och policylabb i Sverige och omvärlden - miljöer för innovation i offentlig verksamhet.

Müller-Krogstrup, J. (2011) How to solve 7 challenges in employee driven innovation. Available at: http://www.innovationmanagement.se/2011/10/20/how-to-solve-7-challenges-in-employee-driven-innovation/ (Accessed: 7 December 2016).

Nadler, D.A. and Tushman, M.L. (1997) Competing by design: The power of organizational architecture. New York: Oxford University Press.

NESTA (2009) Everyday Innovation. Available at: http://www.nesta.org.uk/publications/everyday-innovation (Accessed: 7 September 2016).

NESTA (2014) Innovationstävlingar - en guide. Available at: http://www.vinnova.se/upload/epistorepdf/innovationstavlingar.pdf (Accessed: 20 November 2016).

NESTA (2015a) Scaling and growing your innovation lab’s project. Available at: http://www.nesta.org.uk/blog/scaling-and-growing-your-innovation-labs-project (Accessed: 17 November 2016).

NESTA (2015b) World of labs. Available at: http://www.nesta.org.uk/blog/world-labs (Accessed: 14 September 2016).

Näringsdepartementet (Ministry of Enterprise and Innovation) (2015) Godkännande och undertecknande av avsiktsförklaring mellan staten och Sveriges Kommuner och Landsting för en digital förnyelse av det offentliga Sverige. Available at: http://www.regeringen.se/contentassets/7aca65dd753d4242b4916572a632d007/beslut-om-bemyndigande-att-underteckna-en-avsiktsforklaring-med-skl-om-digital-fornyelse.pdf (Accessed: 21 October 2016).

OECD (2014a) Innovating the public sector: From ideas to impact - OPSI. Available at: https://www.oecd.org/governance/observatory-public-sector-innovation/events/pastevents/innovatingthepublicsectorfromideastoimpact.htm (Accessed: 28 September 2016).

OECD (2014b) OECD recommendation on digital government strategies. Available at: http://www.oecd.org/gov/digital-government/recommendation-on-digital-government-strategies.htm (Accessed: 23 November 2016).

OECD (2016) Digital Government Strategies for Transforming Public Services in the Welfare Areas. .

OECD (no date) Public-sector innovation. Available at: https://www.oecd.org/sti/outlook/e-outlook/stipolicyprofiles/competencestoinnovate/public-sectorinnovation.htm (Accessed: 5 October 2016).

Open Government Partnership (2015) Open government partnership. Available at: http://www.opengovpartnership.org/ (Accessed: 11 December 2016).

OpenLab (no date) About. Available at: http://openlabsthlm.se/about/ (Accessed: 28 September 2016a).

O’Reilly, C.A. and Tushman, M. (no date) ‘Organizational Ambidexterity: Past, present and future’, SSRN Electronic Journal,. doi: 10.2139/ssrn.2285704.

O’Reilly, C.A. and Tushman, M.L. (2004) The ambidextrous organization. Available at: https://hbr.org/2004/04/the-ambidextrous-organization (Accessed: 25 November 2016).

Osborne, S.P., Radnor, Z. and Strokosch, K. (2016) ‘Co-Production and the Co-Creation of value in public services: A suitable case for treatment?’, Public Management Review, 18(5), pp. 639–653. doi: 10.1080/14719037.2015.1111927.

Pan Fagerlin, Wen and Björklund, Matts (2010) ‘In Pursuit of Ambidextrous Organizations: Using Knowledge Sharing and Organizational Learning to bridge structural ambidexterity and contextual ambidexterity’, Sweden: Lund University Publications. .

Regeringskansliet (Swedish Government) (2015) Att arbeta med den statliga värdegrunden en handledning. Available at: http://www.vardegrundsdelegationen.se/media/Att-arbeta-med-den-statliga-v%C3%A4rdegrunden-en-handledning.pdf (Accessed: 3 December 2016).

Regeringskansliet (Swedish Government) (2016) Regeringen ser över styrningen av digitala tjänster. Available at: http://www.regeringen.se/pressmeddelanden/2016/05/regeringen-ser-over-styrningen-av-digitala-tjanster/ (Accessed: 23 October 2016).

Renzi, A. and Simone, C. (2011) ‘Innovation, tangible and intangible resources: The ’space of slacks interaction’’, Strategic Change, 20(1-2), pp. 59–71. doi: 10.1002/jsc.885.

Reynolds, S. (2015) The landscape of public sector innovation labs. Available at: http://www.designforeurope.eu/news-opinion/landscape-public-sector-innovation-labs (Accessed: 7 September 2016).

Riksrevisionen (Swedish National Audit Office) (2016) ''Den offentliga förvaltningens digitalisering – En enklare, öppnare och effektivare förvaltning? (RiR 2016:14)''. Available at: http://www.riksrevisionen.se/PageFiles/24636/RiR_2016_14_DIGITALISERING_ANPASSAD.pdf (Accessed: 2016).

Sein M; Maung K.; Henfridsson, Ola; Purao, Sandeep; Rossi, Matti; and Lindgren, Rikard (2011) ‘Action Design Research’, MIS Quarterly, (35: 1) pp.37-56,.

Simonds, F. (2016) ''Human Centred Design vs Design Thinking vs Service Design vs UX …. What do they all mean?'' Available at: https://www.linkedin.com/pulse/human-centred-design-vs-thinking-service-ux-what-do-all-simonds (Accessed: 3 December 2016).

SKL and VINNOVA (2016) Innovationsdagen 2016.

Statskontoret (Swedish Agency for Public Management) (2015) Vägen till en lärande och effektiv förvaltning – utvecklingsområden och prioriteringar 2015–2016. Available at: http://www.statskontoret.se/globalassets/publikationer/2015/2015104.pdf (Accessed: 3 December 2016).

Steiber, A. (2016) Innovationsledning – att skapa organisatoriska förutsättningar för innovation..

Stockholm County Council (SLL Innovation) (2015) Vår innovationsprocess. Available at: http://sllinnovation.se/var-innovationsprocess (Accessed: 9 November 2016).

Swedish Association of Municipalities and Regions (SKL) (2015) Modell för innovationsarbete. Available at: https://skl.se/naringslivarbetedigitalisering/forskningochinnovation/innovation/modellforinnovationsarbete.4780.html (Accessed: 9 November 2016).

Technopolis (2012) Trends and Challenges in Public Sector Innovation in Europe. .

Tiesinga, H. and Berkhout, R. (eds.) (2014) Labcraft: How social labs cultivate change through innovation and collaboration. United States: Labcraft Publishing.

Timeus, K. (2015) Innovation Units in Executive Government. Available at: http://lipse.org/blogpost/item/2 (Accessed: 13 September 2016).

Tonurist, P., Kattel, R. and Lember, V. (2015) ‘Discovering Innovation Labs in the Public Sector’, Working papers in technology governance and Economic Dynamics, 61.

Turnbull, G. (2014) Government digital service. Available at: https://gds.blog.gov.uk/2014/02/10/striking-a-balance-between-security-and-usability/ (Accessed: 9 November 2016).

UNHCR (2016) UNHCR northern Europe, UNHCR innovation and Transformator design 12-hour workshop for refugees. Available at: http://www.unhcr-northerneurope.org/what-we-do/participation-empowerment/unhcr-service-design-workshop/ (Accessed: 12 December 2016).

Upphandlingsmyndigheten (2016) Innovationsupphandling. Available at: http://www.upphandlingsmyndigheten.se/omraden/innovation-och-upphandling/ (Accessed: 10 December 2016).

Wegener, C. (2012) ''Public sector innovation: Value creation or value loss? - elm - European lifelong learning magazine''. Available at: http://www.elmmagazine.eu/articles/public-sector-innovation-value-creation-or-value-loss (Accessed: 23 October 2016).

Weiss, J.W. and Bentley (2014) Business ethicsa stakeholder and issues management approach, sixth edition. 6th edn. San Francisco, CA: Berrett-Koehler Publishers.

What Works Scotland Evidence Bank (2015) 1 What Works Scotland Evidence Review: Scaling-Up Innovations. Available at: http://whatworksscotland.ac.uk/wp-content/uploads/2015/06/WWS-EB-evidence-review-Scaling-Up-Innovations-June-2015.pdf (Accessed: 15 November 2016).

White, R. (1999) ‘Organisational Design and Ethics: The Effects of Rigid Hierarchy on Moral Reasoning’, International Journal of Organisational theory and Behaviour 2 (3&4), 431-456, 1999,.