1 Introduction

User centricity is generally acknowledged as a key concept in service design (Stickdorn and Schneider 2011). However, the term “user” has been overemployed and is also misleading because it implies all service users being very much alike each other. As this is not the case, it makes more sense to discuss stakeholders of services rather than just users. Indeed, each service comes with at least two kinds of stakeholders: service consumers and service providers. Stakeholders take on different roles. Broadly speaking, three kinds of roles can be distinguished among service consumers: citizen, public servant, and representative of an organization, such as company or non-governmental organization (NGO). Citizens in turn can be divided into groups depending on the roles they take in the society (Mead 1934).

Multiple stakeholders need to be taken into account when designing and developing public sector services and administrative processes supporting them. The nature of these stakeholders contributes to an increasingly complex system of services and processes. We need verifiable, justifiable and repeatable concepts and methods for dealing with that complexity. This article claims that by efficiently modeling stakeholders’ roles, goals, interactions, and knowledge, better public services can be designed and the efficiency and productivity in the public sector can be increased. We use as our methodology for service design Agent-Oriented Modeling (AOM), as proposed by Taveter and Sterling (2009). AOM has been successfully used for designing artifacts that consider the interests of various relevant stakeholders (Miller et al. 2014). In this article, AOM has been adopted to designing and developing public sector services and processes, notably including the users of the services to be designed performing their respective roles. Moreover, AOM also lends itself to fast prototyping of services to be designed and simulation of potential service scenarios in their social context. This article is confined to addressing the usage of AOM models of one particular kind – goal models – for facilitating discussions between different stakeholders.

On the other hand, the increasing complexity and growing capacity of technology and its use in the public sector has now created a situation where the computational power that we possess should be harnessed to truly serve the citizens and pre-empt their needs. However, all stakeholders do not have the same needs and levels of satisfaction (Hamilton et al. 2011). The goals, needs, and levels of satisfaction associated with different services may depend on the role citizens are taking, be it in a public sector, private sector or NGO, or looked at as an individual citizen. In this article we argue that designing proactive services of e-governance should be seen as the next stage in service design for e-governance. In our opinion, proactive public electronic services should be designed in a way that supports the automation and intelligent processing of already available information to reflect the purpose of meeting the needs of different stakeholders yet maintaining a people-first policy. For designing proactive services, AOM is again instrumental because as agent-oriented, it intrinsically supports the notion of proactivity, which is one of the main characteristics of agents or active entities, in addition to reactivity and social nature (Taveter and Sterling 2009).

Finally, it is proposed in this article that governments should introduce and implement the concept of service design thinking in the public sector in order to create public electronic services that would truly and purposefully meet the needs of citizens, businesses and NGOs. We claim that adopting service design thinking in the public sector can be facilitated by AOM. By modeling stakeholders’ goals, and through different service design methods, such as ethnography, creating personas, stakeholder maps and expectation maps among others (Stickdorn and Schneider 2011), it would be possible meet the needs of the stakeholders in order to provide high quality customer experience.

It has become increasingly important to understand how to design and provide ICT-driven public services efficiently. We plan to obtain that understanding by combining sociological qualitative research, interdisciplinary case study methods, and software engineering approaches in the research on service design thinking in the public sector. The overall research method used by us is Action Design Research (Maung et al. 2011). This article seeks to address the aims of service design and proactivity and service automation through the example of the family benefits system in Estonia.

The current article is structured as follows: literature review is followed by a short overview of the research methodology and the description of a case study on the current status of family benefits service in Estonia. The article illustrates the case of service design thinking and its relevance in the context of e-governance. The methodology of AOM is proving to be a valuable tool for approaching the design of proactive public services.

2 Literature Review

The role of the state is constantly changing and the functions of a government accumulate as increasingly more services are expected to be delivered in a more efficient manner (Sirendi 2012). Consequently, electronic government, a concept that initially emerged in the public administration of industrial countries (Schuppan 2009), developed as a reaction to these expectations. E-governance has become increasingly prevailing in delivering services and public value efficiently and in a timely manner.

The last couple of decades have brought along significant changes in how public sector organizations are run (Bode 2012), what technologies are implemented, and what management styles are relevant. Information is the catalyst and ICT impacts the way the world connects and knowledge is networked (Frasheri 2003). Countries have never been connected to one another to this magnitude before (Sirendi 2012).

A number of authors (Layne and Lee 2001; Reddick 2004; Andersen and Henriksen 2006; Lee 2010) address and describe the perceived phases of government. These models, describing the development of (e-)governance, allow us to make some subjective generalizations, by dividing the development into four general categories: cataloguing – governments focusing on establishing their online presence; transactions, where a government’s focus is on connecting “the internal government system to online interfaces and allowing citizens to transact with government electronically” (Layne and Lee 2001), vertical integration, which refers to the connectedness on a local, state, and federal level; and horizontal integration, where integration takes place across and between different functions and services (Layne and Lee 2001). Although these models may explain e-governance development in many countries and may cover the basic features of e-governance growth, these do not explain the potential phases conclusively.

Constant and strong pressure on public expenditure sets a demand to keep on finding ways to increase productivity, while at the same time addressing the needs of the citizens (Karwan and Markland 2006). There is an increasing demand on countries to make use of the allocated resources in a progressively efficient manner (Lindgren and Jansson 2013). The importance of service design in the public sector has arisen. Many service managers must design and re-design services in order to keep their offerings competitive, fresh, and desirable for customers. After each re-design, staff must re-learn to use a modified service system to deliver a high-quality customer experience (Heim and Ketzenberg 2011). Effective service design requires careful attention on different factors such as costs, service levels, efficiency, sales, profits (Narasimhan et al. 2005), and human aspects.

The recent couple of decades have brought insight into the changing concept of service design (Shostack 1982, Shostack 1984; Scheuing and Johnson 1989; Gummesson 1990; Hollins 1993; Kimbell 2011). Now a closer look into service design in the public sector will be taken (Karwan and Markland 2006). Service design uses techniques and research methods of different fields: ethnography, interaction design, and information science (Stickdorn and Schneider 2011), to name a few. The field seeks to understand and design methodologies for both the front and back office of an organization in order to create better, more user-friendly, more usable, and appropriate services (Smith and Fischbacher 2002). Effective service design involves developing a service concept that appeals to end-users while reflecting on operational limitations (Dixon and Verma 2013). The notion of a “service concept” as initially described by (Sasser et al. 1978) in (Karwan and Markland 2006) could be described as a “bundle of goods and services sold to the customer and the relative importance of each component to the consumer” (p. 14).

Unlike a product, service components are often not physical entities, but rather a combination of processes, skills, and resources, that must be integrated properly in order to result in the planned and designed services (Goldstein et al. 2002). When (re-)designing services, managers, designers, community members, and other stakeholders must make decisions about each component of the service (Goldstein et al. 2002). This way numerous decisions are made even for the simplest services as the involved processes are continuously on-going (Goldstein et al. 2002). The variety of stakeholders and their changing missions leave public agencies unable to conclusively achieve efficiency in their operations (Karwan and Markland 2006).

However, there is a growing focus reflected by an extensive amount of literature on the needs of users and growing interest in user-centric services even though little attention is paid to understanding users’ preferences (Venkatesh et al. 2012). The design of user-centric e-governance services will continue to be a challenging task, as citizens’ demands and needs change (Venkatesh et al. 2012).

As one of the most recent developments, Linders et al. (2015) showcase three illustrative case studies that support that proactive e-governance will be the future for the public sector.

The methodology chosen by us, AOM, has been successfully used in requirements modeling to engage diverse stakeholders (Miller et al. 2011, 2014). AOM enables to capture functional requirements in the form of functional goals for the service to be designed, which are arranged into a hierarchy. In the hierarchy of functional goals, non-functional requirements are represented as quality goals attached to the corresponding functional goals. Moreover, when attempting to understand human issues in service design, a range of consumer emotions need to be considered from the extremes of “customer delight” to “customer outrage” (Cook et al. 2002). Agent-oriented modeling can also be used to address these issues in the form of emotional goals attached to the relevant functional goals (Miller et al. n.d.). It is argued here that by using agent-oriented modeling in the service design context, it would be possible to address more precisely human aspect and gain a better understanding of the existing issues in public electronic services.

3 Research Methodology

It has become increasingly important to understand how to design and provide ICT-driven public services efficiently. We plan to obtain that understanding by combining sociological qualitative research, interdisciplinary case study methods, and software engineering approaches in the research on service design thinking in the public sector. The overall research method used by us is Action Design Research (Maung et al. 2011). We have chosen Action Design Research (ADR) because ADR is a research method for generating prescriptive design knowledge through building and evaluating interrelated ICT artifacts in an organizational setting. In our project this has a specific meaning of working out a repeatable method for designing proactive e-governance services in collaboration with their stakeholders. The ADR method consists of four stages. At Stage 1 the problem is identified and described in collaboration between researchers and stakeholders. At Stage 2 an artifact, which in our case is a prototypical implementation of the proactive family benefits service, is iteratively built and evaluated by stakeholders. At Stage 3 the artifact is rebuilt based on the results to apply to a broader class of problems. At Stage 4 the outcome is further generalized to design principles of proactive services of e-governance. This section describes Stage 1 of the case study in the research project undertaken by us. In our case study Stage 1 consisted of the following steps:

  1. 1.

    Interviewing stakeholders in Estonia’s family benefit area.

  2. 2.

    Analyzing interviews by means of qualitative research (structured interviews).

  3. 3.

    Modeling the hierarchy of goals with the associated roles of stakeholders and quality goals and emotional goals by AOM to describe the “ideal” service of family benefits.

  4. 4.

    Using the resulting goal model for facilitating further discussions with stakeholders and obtaining feedback from them.

For accomplishing Step 1, we have carried out three structured interviews with four stakeholders in Estonia’s family benefit area to investigate the perceptions of the interviewees about the potential areas of development for the family benefits service. The interviews were designed to understand the access points for increased effectiveness in providing the family benefits service to eligible persons and improving the service. The interviews were conducted with the employees of the Estonian Social Insurance Board. They were questioned regarding the involved stakeholders, potential improvements in efficiency, and the current situation with family benefits. As the debate on proactivity in the public sector is currently under way in Estonia, the interviewees were also questioned about the possibility of providing family benefits service proactively, as initiated by the Estonian Social Insurance Board. Future stages and iteration of the case study research will also include other stakeholders, such as citizens, in the discussion.

4 Family Benefits Service in Estonia

As was described in the previous section, the Estonian National Social Insurance Board was chosen as an example for this article on the current research by us. Estonia has developed into a novel example of a state able to implement practices in e-governance and other information technology solutions was done without in a short period of time. This having any substantial information infrastructure support (Sirendi 2012). Now, Estonia ranks among the first twenty countries in the world in e-governance development (United Nations 2010, 2012, 2014).

In Estonia, currently more than 170 databases are offering services via X-Road, the architectural backbone of e-Estonia – a data exchange layer that was launched in 2002 – which allows the nation’s different e-services’ databases, both public and private, to function in an interoperable fashion. Through that, estimably over 2000 services are available for use over X-Road (E-Estonia 2016).

However, only a fraction of these services are truly proactive, most notably the service of e-taxation. There is little to nothing written about proactive services in the public sector. Taking into consideration the ever-evolving nature of information technology that is embedded in the domain, it is becoming increasingly important to address proactivity in the public sector, and especially in the context of public services (Püüa 2008; Taveter 2014; Tallo 2015). This parallels with an increasing demand on countries to make use of the allocated resources in an efficient manner (Lindgren and Jansson 2013).

This article seeks to address the aim of proactivity through the example of the family benefits system in Estonia. The case study was chosen because it is a widely used service in Estonia. Family benefits are provided by the Social Insurance Board (Social Insurance Board 2016a) and are available to permanent residents and foreigners who have a temporary residence permit or the right for residence (Social Insurance Board 2016b).

The list of family benefits includes the following social benefits: childbirth allowance; adoption allowance; child allowance; childcare allowance; single parent’s child allowance, conscript’s child allowance or child allowance of a person in alternative service, foster care allowance, allowance for a start in independent life for person with no parental care, and allowance for big families with seven or more children (Social Insurance Board 2016b). An additional, needs-based family benefit is granted and paid by the local municipality (State Gazette 2016).

The application process is available through different channels both online and off-line. The standard process of applying for family benefits includes submitting the applicant’s passport or ID card and residence permit, if applicable. Depending on the type of allowance, one or more documents of the following types may be necessary: a certificate of employment, a Certification by the Defense Forces or the Defense Resources Agency, a document regarding the declaration of a parent to be a fugitive, a judgment on the establishment of guardianship or foster care contract, or a certificate by social welfare institutions or by a school for children with special needs (Social Insurance Board 2016b).

The interviews were designed to address the following list of topics: the stakeholders involved in the family benefits service, the efficiency of the service (both online and offline); potential points where an increase in efficiency is possible; the preference in the use of a channel by service users; and the potential weaknesses and constraints of the family benefits service.

The interviewees identified a diverse range of stakeholders, such as those covering support functions, procedural activities, e-service hosting on the state portal, and different registries on the architectural backbone for Estonia – X-Road. A number of stakeholders are involved whether in the application procedure or covering the support functions: the Social Insurance Board, the Ministry of Social Affair, the Estonian Information Systems’ Authority, different registries, service developers, and different officials at the Social Insurance Board, who are working on the procedural aspects of the service.

The interviews revealed that increasing the efficiency of the service, both online and offline, would involve moving from an e-service (which currently enables submitting an application by electronic means) to an automated e-service (allowing to gather information from different registries and databases). Also, the offline service would benefit from switching from a service-specific application to a unified application, where the user of the service would be able to apply for a number of services at once to save time on doubling the provided by him/her application data.

Although the e-service is available on the state portal, it turned out from the interviews that only an average of 40 % of the service users are currently opting for that. The numbers vary from city to city, reaching up to 50 % in bigger cities (Tallinn and Tartu), whereas only around 10 % of families are currently using the e-service in the North-Eastern part of the country. The interviewees pointed out that a stronger focus on promoting the service and delivering clarifications in different languages would be beneficial.

The interviews showed that the current situation, where the necessary data is available in the registries, allows the application process to be relatively seamless and quick, taking from 15 to 20 days to process an application. However, should the applicant have been previously working or living abroad, the process would be longer, as each situation would usually be looked at in a case-by-case way.

Currently, the use of the family benefits service requires an individual to show initiative and apply for the benefit. According to the interviewees, proactivity in the provision of family benefits service may be hindered by the fact that different stakeholders may be eligible for the same benefit (e.g. both parents may apply for the childcare allowance). However, a proactive proposal could anyway be made to both.

Fig. 1.
figure 1

The first draft of the AOM goal model, affiliated with Stage 3 of the ADR method

Figure 1 represents an initial goal model describing the “ideal” proactive service of family benefits that was designed for further discussions with stakeholders. The figure represents the main goal of the family benefits service – ensure the well-being of the people living in a country – with its two sub-goals, representing the financial and social aspects of ensuring the well-being. Attached to the goals are the quality goals and emotional goals, applying to the corresponding goals. Figure 1 also shows the two stakeholder roles Citizen and Social Insurance Board.

5 Conclusions and Future Research

It has become increasingly important to understand how to design and provide ICT-driven public services efficiently. The authors have planned to obtain that understanding by combining sociological qualitative research, interdisciplinary case study methods, and software engineering approaches in the research on service design thinking in the public sector. The overall research method used by us is Action Design Research.

Technological advancements have opened up new possibilities for developing public services, e.g. personalized government services (Pieterson et al. 2007), which bases its core thesis on designing services around citizens’ needs, rather than around the needs of the service providers. However, the author believes that the technological environment and current computational capacities would make it possible to take personalization of public services to a next, “proactive”, level. Currently, the most relevant discussion regarding proactive public services and proactive government has been initiated regarding the Taiwan’s fourth e-governance policy (Linders et al. 2015), which claims that instead of the current “pull” approach of traditional e-governance, where citizens must seek services provided by government, a “push” approach should be implemented where government proactively and seamlessly delivers timely, customized, and relevant services to its citizens. In our opinion the key to achieving proactive and public services lies in tailoring the ICT infrastructure of a state for the needs of citizens, businesses and NGOs rather than for supporting ministries and governmental organizations per se. The design of targeted proactive and personalized services requires a method for modeling sociotechnical systems that are associated with multiple agents. The AOM approach is suitable for this purpose, as it allows designing services that are associated with many different stakeholders. Moreover, AOM lends itself to analyzing problem domains of services and their stakeholders through different lenses, such as the goals, stakeholder roles, problem domain knowledge to be represented in registries and databases, interactions, and behaviors. Also, AOM is geared towards the notion of proactivity, which is one of the three fundamental characteristics of an agent. Additionally, as showcased in Miller et al. (n.d.), AOM includes the emotional dimension of modeling stakeholders’ attitudes, which makes it especially useful in the service design context, where user-centered approach predicates greatly on emotional dimension.

The chosen approach – analyzing service design thinking and proactivity in the context of public service provision and through the lenses of the AOM methodology – allows the authors to synthesize the already existing information into a conceptual model that would provide mainly public sector organizations, but also other stakeholders such as private companies and NGOs, with an approach that would help them concentrate on the needs of the citizens and users of the service, playing different roles. The new approach – proactive service design – would facilitate automation of e-governance and enable maximum usage of information that is already stored in different registries and information systems run by government. Future research will seek to define how AOM and service design thinking could be the solution for building greater proactivity into the public sector.