Measuring interoperability maturity in government networks

Transcription

1 2012 Measuring interoperability maturity in government networks # IKU K.L.L.G. Huijsman Universiteit Utrecht

2 Measuring interoperability maturity in government networks Master thesis # IKU Utrecht University Author: Koen Huijsman Master of Business Informatics Utrecht University k.l.l.g.huijsman@students.uu.nl Supervisors: External supervisor: dr. Marijn Plomp dr. Ronald Batenburg drs. Melior Hoekstra Department of Information Department of Information Advisory and Computing Sciences and Computing Sciences PwC Netherlands Utrecht University Utrecht University m.g.a.plomp@uu.nl r.s.batenburg@uu.nl melior.hoekstra@nl.pwc.com Supporting organization: PwC Netherlands 2

3 Abstract The purpose of this research was to create a model that describes the development of interorganizational collaboration in government networks that apply e-government. Contrary to several models that describe e-government from a government-to-citizen perspective, and primarily focus on the front office of e-government services, this research focuses on the collaboration that takes place in the back office to enable successful e-government services. A maturity model to describe and assess the level of interorganizational collaboration in government networks that apply e-government was developed. Through a structured literature review, 19 existing maturity models related to interorganizational collaboration and e-government were identified. These models were subsequently analyzed on the dimensions and stages that were used to define maturity. Furthermore, the characteristics of each stage and the preconditions for increasing maturity were studied. Based on this literature review and the analysis, a new maturity model was proposed, in which existing concepts are integrated and extended with a network perspective. This model describes the levels of interorganizational collaboration in government networks on three dimensions: system, information and process. Five levels of increasing interconnectedness describe how the interorganizational collaboration in government networks unfolds across these three dimensions. The model was empirically applied in case studies of three government networks. Medium- to large-sized networks of municipalities and their cooperating partners that apply e-government services in their permit application procedure were studied. The model appears to be applicable to assess the development of interorganizational collaboration within government networks that apply e-government in their service provisioning. Further research could focus on using the model to analyze additional growth strategies to create successful roadmaps through the model. 3

4 Acknowledgements I would like to thank my supervisors Marijn Plomp and Ronald Batenburg for their constructive comments during my thesis project. As my first supervisor, I would like to thank Marijn for his enthusiasm and positivism in his overall guiding of this project. By providing his comments he helped me to see any obstacles along the way as challenges to overcome. His straight forward way of providing feedback on my work is something I particularly liked. Although we were not always given the opportunity to meet face-to-face, modern communication tools such as Skype helped us out when needed. I would further like to thank Ronald as my second supervisor for his additional feedback on both the research paper and my thesis, providing me with additional food for thought on the crucial parts of my work. I would also like to thank PwC, the supporting organization where I worked on my thesis during a large part of my thesis project. Melior Hoekstra, my external supervisor provided me with feedback on several occasions as well as several other colleagues, providing me with additional insights of the events and developments that take place within the public domain. Finally I would like to thank my friends, family and particularly my girlfriend for supporting me during this last year by listening to my ideas and activities, providing me with refreshing insights and providing some distraction when needed. 4

5 Table of Content Abstract... 3 Acknowledgements... 4 List of figures... 8 List of tables Introduction Problem statement Relevance Scientific relevance Social relevance Research approach Literature review Artifact design Case study Scope Theoretical background Systematic literature review Planning Selection Extraction Execution Information technology in public administration E-government Conclusions Models of growth Characteristics of stage models Stage models in e-government Perspectives Levels Boundaries and transitions Conclusions Interorganizational collaboration Organizational growth in government chains Conclusions Interoperability Interoperability frameworks in e-government Interoperability models of growth in e-government Conclusions Artifact design Dimensions Selecting the dimensions

6 Labeling the dimensions Stages...40 Number of stages Labeling the stages Shifting between the stages New stage model for interoperability in government networks Independent stage Ad hoc stage Coordinated stage Domain stage Unified stage Context & use of the model Case study Objectives Case study design Case scenario Unit of analysis Site selection Data collection methods Data analysis Case 1: Municipality of Haarlem Case 2: Municipality of Ede Case 3: Municipality of Leiden Cross-case conclusions...60 Information systems...60 Information Processes Relations between dimensions Maturity levels cases Discussion Conclusions References Appendices Overview of models of growth found in the systematic literature review Protocol preliminary interview case study (Dutch) Introductie Casus afbakenen Protocol in-depth interview case study (Dutch) Introductie Vragen Contacts & interviewees

7 5. Simplified process model All-in-one-permit Complete list of search queries used in systematic literature review

8 List of figures Figure 1: A systematic guide to literature review development (Okoli & Schabram, 2010) Figure 2: DSRM Process Model (Peffers et al., 2008) Figure 3: Unit of analysis case study Figure 4: Research domains Figure 5: Articles per domain Figure 6: Publications per year Figure 7: Positioning model for e-government maturity models Figure 8: Stage model presented as change in time (Klievink & Janssen, 2009) Figure 9: Interorganizational network with reciprocal interdependence Figure 10: Interoperability Framework EIF v2.0 (European Commision, 2010) Figure 11: Design Science Research Method by Peffers et al. (2008) Figure 12: Interoperability maturity model for government networks Figure 13: Case study design (based on Yin, 2003) Figure 14: Setup one Figure 15: Setup two Figure 16: Maturity levels per dimension for each case Figure 17: Simplified process model of the all-in-one-permit application procedure List of tables Table 1: Main search queries per research domain Table 2: E-government maturity models analyzed (condensed table) Table 3: Four levels of interoperability (Kubicek & Cimander, 2009) Table 4: G2G models with their stages Table 5: Maturity model for interorganizational collaboration in government networks Table 6: Comparison maturity levels on three dimensions Table 7: Search queries

9 9

10 1 Introduction The way governments operate has changed dramatically with the introduction of information and communication technologies. Governments have gone online to provide public services to their citizens (Layne & Lee, 2001). With these technologies, governments aim to provide services that are more citizen-centric and that integrate their operations between different administrative levels and agencies (Layne & Lee, 2001). It helps creating improved efficiency and effectiveness in government processes and enables innovating them. Lee (2010) therefore states that this way of service providing will play an important role in public administration and governance in the future. These developments led to the introduction of the term e-government. The term first appeared in literature around 2000 and several different definitions were used to define its meaning. Schneider and Means (2000) defined e-government as the relationships between governments, their customers and suppliers (businesses, other governments, and citizens) by the use of electronic means. This definition covers all parties involved and refrains from stating explicit technologies that change over time and will therefore be adopted in this thesis. Although there has already been a considerable amount of research on IT in governments in the 1970s, 1980s and 1990s, the introduction of e-government services, mainly in the last decennium, has created new research grounds to study these new virtual services of the government (Andersen & Henriksen, 2006). A topic that particularly gained quite some interest from the research community is maturity models for e-government. Shifting towards a fully functional e-government does not happen overnight, instead several studies suggest it happens in stages (Lee, 2010). A well-known and one of the first maturity models for e- government growth is the Layne and Lee model (Layne & Lee, 2001). It consists of four stages namely (1) cataloguing, (2) transaction, (3) vertical integration and (4) horizontal integration. The first stage, cataloguing, simply exists of the government having several web pages with non-transactional information. In the second stage, transaction, the static web pages evolve into pages where electronic transactions take place. Rather than just automating and digitizing existing processes, the third stage is characterized by integrating systems within similar functional walls but across different levels of the government. This form of integration is called vertical integration. The final stage is involves horizontal integration, where government agencies integrate their systems across different functional walls and thereby become fully integrated (Layne & Lee, 2001). Other models begin with similar stages, such as the model of Heller and Bélanger (2001) and Ronaghan (2001), but go even further by stating that in the end e-participation by citizens will also be added to the e-government concept. While all these models describe e-government maturity, each uses different concepts and perspectives to illustrate this (Lee, 2010). Models in current literature are partly descriptive, partly predictive and partly normative. The number of descriptive models is limited since most models contain stages that are hard to reach in practice. In their early stages the models tend to be more descriptive, whereas in their later stages, they become more predictive and normative (Coursey & Norris, 2008). The majority of the models consist of four or more stages to describe the development. They are assumed to be cumulative, meaning that the next stage can only be achieved once the previous ones are also fulfilled (Lee, 2010). 1.1 Problem statement Current studies on maturity models in e-government only draw the global e-government picture, where the focus is mainly on the development of e-government in the front-office. However, many of these developments also take place in the back-office where different levels and functions within the government can be distinguished. Several government organizations cooperate within a certain government level (horizontal) and between different government levels (vertical) to provide services to the public. They form a chain of government agencies working together. Within these chains, government organizations do not only collaborate with each other, but also with other agencies, businesses and civilians when needed. 10

11 Whereas there is literature available on interorganizational collaboration through ICT in other sectors, such as the retail sector (Plomp & Batenburg, 2010), less is written on this topic in the public sector. Especially on achieving interoperability in the back-office of e-government networks. Since improved interoperability among government organizations is of great importance for overall e-government success, it is relevant to study this development (Gottschalk & Solli-Sæther, 2008). Gottschalk and Solli-Sæther (2008) made some progress in this direction by creating a conceptual model of maturity levels for interoperability in digital government. They identified five stages of interoperability, on (1) computer, (2) process, (3) knowledge, (4) value and (5) goal. There is however a need to identify characteristics and their values at each stage, more specific scales need to be defined between different levels. (Gottschalk & Solli-Sæther, 2008). This links to the lack of processoriented e-government studies as opposed to output and outcome-oriented ones (Yildiz, 2007). Different stages are often described, but not the conditions under which government agencies can reach them. The dimensions used in current models are underspecified (Coursey & Norris, 2008). Interorganizational collaboration furthermore takes place in networks of government organizations, but current models do not sufficiently apply a network view to describe the development. Based on the foregoing, this master thesis will focus on answering the following research question: How does interorganizational collaboration develop in government networks that apply e-government? With the following sub questions: (a) What is e-government and why is it relevant? (b) Which current models exist to describe the maturity of interorganizational collaboration in government networks that apply e-government? (c) Which dimensions can be used to describe the maturity level of interorganizational collaboration in government networks that apply e-government? (d) Which stages, their characteristics and preconditions can be distinguished to describe a maturity model of interorganizational collaboration in government networks that apply e-government? (e) How can this maturity model be used in describing and guiding the development in interorganizational collaboration in government networks that apply e-government? By answering these questions, the development of organizations towards a higher level of interorganizational collaboration through ICT will be mapped. Specifically focusing on describing the preconditions that have to be met to reach a higher level in collaboration and how these translate into transition processes for organizations that want to improve interorganizational collaboration in their government network. Creating and/or adopting a maturity model will aid in both describing and prescribing these processes. 1.2 Relevance Scientific relevance By focusing more on the processes that lead to a certain stage, instead of on the outcome of a stage, this thesis will give a more process-oriented approach, that was missing in previous literature (Yildiz, 2007). Studying the subject from this angle is relevant, since it creates a better understanding of why certain organizations or chains reach a higher level than others. It will hopefully reveal properties and preconditions that enable or prevent further development of interorganizational collaboration through ICT. Furthermore the exact dimensions in which the development is described are often underspecified in previous literature, this research will clarify these dimensions. This research can furthermore be linked to measuring interorganizational collaboration in ICT in other sectors, such as the retail sector (Plomp & Batenburg, 2010). Similarities and differences between different sectors can be analyzed, possibly leading to new insights in both domains. 11

12 Heeks & Bailur (2006) describe several points on which current e-government research can be improved. They noticed a strong theme of over-optimism, or what they call hype in many studies with regard to the capabilities of e-government initiatives. Keeping this in mind, this thesis will focus on maintaining a more balanced view on the impact of e-government. There is also a dominance of research methods that require no face-to-face engagement with the realities of e-government. With help of the case study that will be performed in this study, this thesis will make a start in filling that gap in current e-government literature. Social relevance Many e-government projects fail due to poor coordination and agencies that act too independently (Strejcek & Theil, 2002). By creating a better understanding of how networks of government agencies mature in their collaboration, this thesis will help agencies that are going through this transformation. Providing a way to indicate their e-government level, governments can assess their current level and define a strategy to reach the next level. A roadmap can be created for e-government projects, aiding organizations in their transformation. By describing the characteristics of different e-government stages, agencies can decide if their ambitions to reach higher levels of interorganizational collaboration, match the conditions of these stages. Based on the starting point of an organization and its characteristics, different scenarios can be created to aid in the planning and execution of transformations towards a higher level of collaboration. This can prevent initiating e-government projects that are too ambitious to execute at a certain level. Furthermore, lessons can be learned from similar chains of government agencies that have already gone through this transition by following their example and preventing known pitfalls. 1.3 Research approach Literature review The first part of the research method will consist of a literature review, which has several purposes for research in general. Among the purposes are providing a theoretical background for future research, learning the width of research on a topic of interest and answering practical questions by understanding what existing research has to say on the matter (Okoli & Schabram, 2010). Furthermore, Hart (1998) provides additional purposes that justify literature reviews in academic theses. He explains that in a literature review prior understanding on the subject can be demonstrated, a testament to the rigorous research dedication can be posed, scholarly conventions can be trained and applied and future research can be justified. Okoli and Schabram (2010) describe how a systematic literature review for information systems (IS) research can be conducted. They present eight major steps that they consider to be required for a systematic literature review and essential for the review to be scientifically rigorous. The steps are as follows: (1) Purpose of the literature review, identifying the purpose and goals of the review (2) Protocol and training, agreement about the procedure to follow (3) Searching for the literature (4) Practical screen, practical reasons for considering or not considering certain studies (5) Quality appraisal, setting criteria for quality (6) Data extraction, systematically extracting the applicable information (7) Synthesis of studies, analyzing and combining the extracted facts (8) Writing the review These steps can be categorized in four main phases in the review, being the planning, selection, extraction and execution phase, as depicted in Figure 1. 12

13 Figure 1: A systematic guide to literature review development (Okoli & Schabram, 2010) Since the subject of this thesis can be placed in the IS research domain, the method of conducting a systematic literature review presented by Okoli and Schabram (2010) is suitable to apply in this thesis. Artifact design Design science (DS) involves a rigorous process to create an artifact. An artifact being any designed object with an embedded solution to an understood research problem (Peffers, Tuunanen, Rothenberger, & Chatterjee, 2008). Such artifacts can include constructs, models, methods and instantiations. Peffers et al. (2008) describe the cycle of creating a new artifact in which they present principles, practices and procedures to perform this kind of research in the IS domain. They base their cycle on earlier design science research principles such as those mentioned by Hevner (2004). They describe six steps in design research which are problem identification and motivation, definition of the objectives for a solution, design and development, demonstration, evaluation and communication. Figure 2 shows the design science research methodology (DSRM) processes, as described by Peffers et al. (2008). The DSRM process model provides a rigorous method to create an artifact. The method provided by Peffers et al. (2008) can both help in creating the model and in structuring the whole research process and using it thereby has multiple purposes. The systematic literature review can be integrated in the first two phases in the design science research methodology. 13

14 Figure 2: DSRM Process Model (Peffers et al., 2008) From the literature review, existing maturity models are expected to be found and will be analyzed on similarities and differences. Based on the identified characteristics and problems with current models, objectives for a solution will be defined. A problem centered initiation of the research cycle is therefore applicable in this thesis (see Figure 2). The objectives will be used to design a new or extended artifact, in which the desired functionality and architecture of the model will be developed, resulting in a new model. This model will then be demonstrated and evaluated in several case studies that are described in the next section. The thesis and scientific paper form the main communication tools to share the designed artifact. Case study In a case study a phenomenon is examined in its natural setting, employing multiple methods of data collection to gather information from one or a few entities (i.e. people, groups, or organizations) (Benbasat, Goldstein, & Mead, 1987). The case study will be used to both validate and improve the created model. Unit of Analysis Before conducting a case study a unit of analysis should be determined, for example individuals, groups or entire organizations (Benbasat et al., 1987). This thesis will focus on interorganizational collaboration in government chains. An additional unit of analysis therefore applies in this case, being the chain or network of government organizations. The unit of analysis will therefore be organizations on the chain level and the chain as a whole. However, to be able to analyze a network of organizations, individual organizations have to be analyzed as well. Therefore, individual organizations are also seen as a unit of analysis in the case study. Figure 1 shows possible units of analysis and which of these units will be analyzed in the case study (marked blue). 14

15 Figure 3: Unit of analysis case study Criteria for selecting unit of analysis The main criteria for choosing an organization to perform a case study on, is whether or not it is a part of a mid to large sized government chain (depending on the average amount of organizations in government chains), to be certain that several organizations in the chain can be studied. Ideally, one small, one medium and one large organization within a specific chain will be part of the case study to fully cover the different sized organizations. The selection of organizations also depends on the role of the organization within the chain, ideally organizations that form an important node in the network will be selected. Furthermore, to be able to make a comparison between different chains, a second chain with one small, one medium and one large organization will also be part of the case study. This would preferably be a chain on the same government level, such as a similar government chain within a different province. Data collection methods Multiple data collection methods can be used in a case study, this thesis will focus on available documentation within organizations related to interorganizational collaboration through ICT. Furthermore, to complement documentation, expert interviews will be conducted with people in the organizations and people outside the organizations that are involved in implementing e-government projects. These interviews will be standardized open-ended interviews following a predefined procedure. 1.4 Scope E-government can be divided into three broad categories, Government-to-Government (G2G), Government-to-Civilian (G2C) and Government-to-Business (G2B) (Yildiz, 2007). The end goal of chains of government organizations is to provide services to civilians (G2C) and to businesses (G2B). But these efforts often take place at the beginning and end of the government chain, for example by initiating an information request by a civilian. In order to provide these services, several transactions between government organizations (G2G) are required. The integration of G2G transactions form the main obstacle for an effective e-government and the focus in this thesis is therefore mainly on G2G collaboration (Scholl & Klischewski, 2007). Bekkers (2005) furthermore states that the collaboration between back offices is often obstructed by interoperability problems and these problems are therefore an important factor to take into account when studying G2G transactions. The research will focus on chains of government agencies within the Netherlands. Although the results of this research may also apply to chains in other countries, this will not be further researched in this thesis. The main focus is on the public sector, organizations and other parties outside of the public sector will only be involved in the research if they are an essential part of a certain government chain. 15

16 This thesis will focus on the development of these chains from organizational and technological perspectives. Other dimensions such as social, political, economic and environmental influences will not be the main focus in this thesis. They will primarily be used to describe the context of the maturity model where needed. 16

17 2 Theoretical background 2.1 Systematic literature review Planning The primary goal of performing a systematic literature review is to gain literature that serves as the input for the model of growth that describes the development of interorganizational collaboration in government networks that apply e-government. The secondary goal is to provide a theoretical background for the thesis and learn the width and depth of the research domain. The systematic literature review is conducted following the steps provided by Okoli & Schabram (2010). This thesis combines concepts and ideas from different research areas and the systematic literature review is therefore also focused on subjects in various research domains. These research domains result from the research questions and can roughly be divided into three areas. These are models of growth, interorganizational collaboration and interoperability. This is illustrated in Figure 4 below. These research areas will furthermore predominantly be described from an e-government perspective, creating an additional overarching e-government domain. Figure 4: Research domains The information that is gathered from the literature review will be extracted, analyzed and combined. The result of the analysis of the literature is described in the following sections and the result of the combination of literature is the constructed model of growth that will be described in chapter 3. Protocol The protocol of the systematic literature review is based on the topics of the predefined research domains. Various search queries were derived from the main topic of these domains. Variations on these names, based on synonyms and related topics were added as well. The search queries in Table 1 show which main search queries were used for the different research domains. These queries were furthermore combined to provide results that focus on a specific area. For example, e-government was combined with maturity model to search literature that focuses on describing the different stages in the development of e-government. Additional queries were added based on findings from preliminary search results. 17

18 Table 1: Main search queries per research domain Maturity model(s) E-government Interorganizational collaboration Stage model(s) Electronic Interorganiz(s)ational government cooperation Growth Public administration Interorganiz(s)ational model(s) network(s) Level(s) E-business Crossorganiz(s)ational collaboration Stage(s) Interorganiz(s)ational network(s) Development Phase(s) Interoperability Technical interoperability Organizational interoperability Semantic interoperability Conceptual interoperability Interoperability framework(s) Electronic Data Interchange (EDI) These queries were subsequently entered into a set of scientific search engines. This set contained both general scientific search engines as well as engines that focus on a specific research domain. The general search engines used are the Omega search engine of Utrecht University and Google Scholar. Both these search engines cover many full-text papers from thousands of different digital journals and research areas. Specific IS related search engines used were ISI web of knowledge, IEEE Xplore and the ACM Digital Library, which focus predominantly on information systems research. Most of these search engines are aimed at finding full-text papers, but Google Scholar for example, also allows to search for and view published books. Selection The criteria for selecting a paper were based on the primary and secondary goal of the literature review. The primary goal of the literature review is to acquire input for the model of growth. Papers that were considered for selection should therefore focus on describing models of growth. Models of growth within the e-government domain are preferred, but papers describing models of growth in general or in different research areas were also considered for selection. Next to literature that describes specific models of growth, literature that describes how to construct (a part of) these models also fall under the criteria. The secondary goal of the literature review is to provide a theoretical background for the thesis and learn the width and depth of the research domain. This part of the literature focuses predominantly on the search areas of interorganizational collaboration and interoperability, mainly within the e- government domain, but literature outside the e-government domain is also considered for selection. Relevant articles were saved into folders that follow the structure earlier defined by the distinction between the different research domains. For every research domain there was a folder and separate folders were created for articles that fell into overlapping research domains, which can be seen in Figure 4. Reference managing software was used afterwards to order the literature and gather various statistics on publication year, journal name and authors. Selection of literature was further based on some practical considerations. For each query the first hundred results were analyzed for relevance, this threshold was set since experience shows that the chances of finding additional relevant literature decreases after this amount and keeps the selection stage within acceptable limits. All variations of the queries were applied on different search engines until they stopped delivering new hits in terms of relevant literature. Extraction The practical screen of the literature was followed by a quality appraisal. Literature that did not appear in a published book, refereed journal or related conference proceeding was excluded from the selection. The chance of encountering such articles already decreased with the selection of the search engines, which focus on scientific literature. However, sometimes articles containing information without a clear scientific background were excluded, such as popular literature and lecture slides. 18

19 To complement the systematic literature review, another approach was taken based on the overarching e-government perspective. Heeks and Bailur (2006) identify three leading e-government research outlets, which are Government Information Quarterly, Information Polity and The Proceedings of the European Conference on e-government (ECEG). The volumes of these journals were analyzed on literature that fall into one of the earlier defined research domains and that could further contribute to drawing the theoretical background or answering the research questions. Additional papers that resulted from this search and that adhere to the standard of relevance and quality set in the planning phase of the systematic literature review were added to the existing set of literature. All remaining relevant literature was further studied and applicable information was extracted, analyzed and combined into an overview of the literature. After reading the title and abstract of every article, the body of the article was scanned for relevant sections. The article was then further read based on relevance and the important conclusions and assumptions made in the text were noted. Finally, references to related articles were followed and added to the library if not found earlier. The result of this selection and extraction procedure is displayed in Figure 5. A total of 140 articles were found suitable to further analyze and process. Thirty articles covered general e-government topics, and 110 articles covered specific topics in the e-government domain, the distribution of these articles over the different research domains is depicted in Figure Interorganizational collaboration Models of growth 4 35 Interoperability Figure 5: Articles per domain Figure 6 furthermore shows the distribution of these articles over the years in which they appeared. The majority of the articles found were published in the 2000s. This observation can be explained, since the concept of e-government and the related research domain arose around the year 2000 and was predominantly present in the following years. There were only two papers that addressed e- government before 2000, dating from 1997 and 1998 and these can be seen as visionary literature on the e-government concept. The remaining literature that was published before 2000 describes maturity models and interoperability outside the e-government context. 19

20 18 Articles per year Articles per year Figure 6: Publications per year Execution The following section describes the results of the systematic literature review. The theoretical background is drawn based on the general e-government literature. A more in depth analysis of the various research domains is given in the subsequent sections, where interorganizational collaboration, maturity models and interoperability in e-government are examined. Conclusions based on the theoretical background form the foundation for the construction of the conceptual maturity model in the next chapter. In the DSRM process model by Peffers et al. (2007) this phase corresponds with the problem identification and finding the objectives for a solution. 20

21 2.2 Information technology in public administration During the last decennium, the use of ICT in governments and the public sector has undeniably increased (Gil-Garcia & Martinez-Moyano, 2007). Although ICT is now starting to become widespread within governments and many IT projects are initiated, this has not always been the case. The importance of ICT has often been neglected or at least underestimated in public administration theory. Even though public administration has long been a document driven domain which suits itself for the use of information technologies, few references to ICT in public administration literature were made during the emergence of ICT. This neglect was not only noticeable in the theory of public administration, but also in practice. There seemed to be an unspoken consensus that public administration would not be affected by the increasing spread of ICT in the domain. There was a notion that public administration would certainly not be changed, or worse, transformed by ICT. The reason for this sentiment was that ICTs were mainly regarded as instruments and not as influential factors in government processes. (Meijer, 2007). With the emergence of computers in public administration in the 1970s, opportunities for applying technology within the government first emerged. At that time, the main goal of using technology in governments was to enhance the managerial effectiveness of public administrators while increasing government productivity (Yildiz, 2007). This mainly consisted of automating existing processes, such as financial transactions and calculations. The introduction of personal computers in the 1980s and 1990s started a new period of IT use in the government, providing each public administrator with a personal information technology system. Computer systems operated independently and mainly within organizations (Kraemer & Dedrick, 1997). Although some interaction was possible within a single organization, there was no coupling of systems between different government organizations. The systems were closed and were therefore often called stovepipes (Yildiz, 2007). With the introduction of ICT in public administration, the topic also drew more attention of research. The transformational power of ICT within public administration was often discussed, sparking a discussion between supporters and adversaries of this theory. At that time, a majority of both scholars and practitioners did not truly believe that introducing ICTs in public administration would drastically change the way it was run. This still showed a substantial reluctance towards using ICT within the public administration domain. ICT was still mainly regarded as an instrument instead of an influential factor in governmental change. Or as Meijer (2007) puts it: The important role of information in government is recognized, but the role of technology is not. This results from viewing ICT as an independent variable and explains why it gained little attention in the past (Meijer, 2007). Since the discussion on the transformational power of ICT in public administration researchers have gained several insights. The first insight is that technologies that influence informational resources also trigger institutional change. This assumption is made because information forms a core component of government agencies. Technology can both strengthen and weaken existing orders since it influences the distribution of information resources within agencies. Changes in information patterns may also influence governmental regimes. Meijer (2007) therefore reasons that changes in core functions of public administration will inevitably affect its institutions. Bekkers (2005) further states that the characteristics of ICTs steer transformations in the public administration in the direction of more control over society. Meijer (2007) concludes that the major transformations of public administration are not only feasible because of ICTs, they are already happening because of them. Although several reasons to apply ICTs within government agencies were indicated, there was still a need of a catalyst to get this process starting. A turning point came with the emergence of the internet. In the private sector, increasingly more citizens and business started using the internet to acquire information and provide services to each other. Citizens and business expected that the government should be present on the internet as well. This put additional pressure on governments to follow this trend and encouraged them to start taking part in this new form of information exchange. While the use of ICT was mainly internal and managerial before the rise of the internet, after its emergence the role of ICTs within public administration was reconsidered. New goals for the use of IT in governments were increasing the convenience of government service provision, facilitating administrative reform and furthering democratic participation. Furthermore, customer-orientation was often mentioned as an important goal to pursue, as well as enabling new methods of production and increasing the flow and 21

22 accuracy of information (Landsbergen & Wolken, 2001). This did not only affect the use of ICTs within government agencies, but also its role in public administration research. E-government The new goals that were formulated by governments to provide more effective utilization of ICT resulted in a shift in the research themes of ICT in public administration. Focusing on a more customer-oriented approach of service providing, the use of ICT became an important factor in the transformation of governments. Propelled by the internet, interorganizational collaboration became an important issue for government organizations. The integration of different organizations to realize better service provisioning attracted many researchers. This new field of research differed from traditional public administration research. There was a strong overlap with research on information systems, its theories and views were therefore widely adopted. At the same time it leaned on the heritage of public administration research, especially on the organizational side (Heeks & Bailur, 2006). In the overlap between public administration and information systems research, a new research domain emerged. The term e-government appeared to fill the gap between these two domains. Because of the former reluctance towards ICTs in public administration research, this new research area never quite merged with traditional public administration research. The term can be used to describe the research domain as well as the practice. To define the exact meaning of e-government as a practice a definition will be provided that will be used throughout this thesis. Definition e-government The phenomenon of electronic services to support government processes can be described with the term e-government. The term first appeared in literature around 2000 and several different definitions were used to define its meaning. Some focus more on explicit technologies, while others like Schneider and Means (2000) maintain a generic definition of the term: The relationships between governments, their customers and suppliers (businesses, other governments, and citizens) by the use of electronic means. This definition covers all parties involved and refrains from stating explicit technologies that change over time and is therefore adopted in this thesis. Main topics in e-government field Several research topics can be distinguished in the e-government domain. In the first decade of academic e-government research, the focus was on the observation and evaluation of the output of e- government initiatives. The initial presence of the government on the internet was a starting point for research. Analyzing content of (web) services and describing best practices for benchmarking these services formed a large part of research. This phase was followed by describing the development of e- government, from a maturity perspective, resulting in several models that described the growth in service levels provided by e-government (Yildiz, 2007). Initially describing these processes on a national level and later also focusing on regional and local level (Gil-Garcia & Martinez-Moyano, 2007). More recently, studies started focusing on the processes in the background that lead to good e- government at the front office, such as interorganizational collaboration and interoperability between different government organizations. Heeks and Bailur (2006) analyzed a large amount of e-government literature which they identified in three large e-government journals. They sorted the literature, based on different perspectives of impacts associated with e-government, which range from optimism to pessimism. Next to these perspectives on the impact of e-government, the causes of these impacts were divided into technical causes and social causes. Their analysis shows that most of the researches have a neutral to optimistic view on the impacts of e-government, with a tendency towards optimism. The causes of these impacts are most often attributed to a combination of technological and social determinants, with a slight tendency towards technical causes. The larger part of the literature that Heeks and Bailur (2006) analyzed was based on describing a model. Many of these researches describe a model of growth that is linked to the development of different aspects of e-government. A large amount of literature is also category-based. Which can be 22

23 defined as work that presents a set of categories or a list of factors that have an impact on e- government, for example consisting of critical success factors for e-government. Conclusions Although the use of ICT and its transformational power was not immediately accepted in public administration, its importance nowadays can hardly be neglected. This shows from the efforts that are put into implementing various ICT solutions by government agencies and the description and analysis of these developments by researchers. The focus is mainly on the field of e-government, where governments use electronic services to provide services to its citizens and other government agencies. An overview of the literature shows that there is much research performed on describing the developments that lead to efficient electronic services. The early days of e-government research mainly focused on describing the level of service provided by the electronic government, while the last couple of years the focus is shifting towards the back office of e-government, where the interorganizational collaboration using ICT between government agencies must be optimized to successfully deliver services to the customers. Since the description of e-government developments in models of growth is an important subject in both e-government research and this thesis, the following section will further analyze this area of e-government research. 2.3 Models of growth As concluded by Heeks and Bailur (2006), many researches in the e-government are based on describing a model, often a model of growth. Models of growth can be used to describe the development of various phenomena in different research areas. They form a common approach of describing growth in several domains, from the growth of organizations and the sales levels of products to the growth of living organisms (King & Teo, 1997). These models of growth are often called maturity models or stage models. From the models of growth encountered in literature, the term stage model was found to be the most common description of these models and will therefore be used further on in this thesis. A model that has influenced the popularity of stage models comes from the hand of Nolan (1973), who created a stage model of organizational growth. By creating a model that was intuitively attractive, he sparked a series of other stage models. These models can among others be used to direct transformational processes and form a roadmap to guide these transformations. Some critique on stage models such as the one by Nolan (1973) was given by Benbasat, Dexter, Drury and Goldstein (1984), who came to the conclusion that empirical support for the stage hypothesis was unconvincing. Several studies on the model indicated that the various maturity criteria did not reliably move together, or even always in the same direction. This conflicted with Nolan (1973) who stated that the characteristics of each stage should be distinct and empirically testable. This critique does however not immediately reject the stage hypothesis. Many of the original problems in the model by Nolan (1973) can be attributed to the small number of organizations that were used in the study and the early point of development the IS field was in at that time. Since the IS field developed and changed rapidly, so did the stages that were initially described. This also indicates that maturity is not a static concept, but can change over time. Another explanation is that the operationalization of the model was insufficiently described by Nolan (1973) which led to differences in interpretation. The critique by Benbasat et al. (1984) can be taken into account when creating a new model of growth. The characteristics of each stage should furthermore be selected carefully, especially with regard to the level of detail. They have to be distinct and empirically testable, but at the same time not with a level of detail that is too high. This to prevent creating a model that is only applicable to a small amount of cases and thereby threatens external validity. Describing the creation and operationalization of the model is therefore also required to ensure the reproducibility. Before a model of growth can be successfully applied to a wide variety of cases, it has to go through a cycle of consecutive development and validation within a sufficient large pool of cases, to change or even reject certain characteristics. Finally it is important to assess the level of change in the environment that is described by the model. A rapidly changing environment can affect the characteristics of the stages and thereby the durability of the model. One could even question the feasibility of creating a 23

24 model for such rapidly changing environments. When taking these points of critique into account and applying them to a suitable environment, models of growth are well applicable. An example of a model of growth in IT which takes these points of critique into account can be found in Leem, Kim and Yun (2008). Characteristics of stage models Stage models stem from the evolutionary perspective, which unlike the contingency perspective emphasizes the presence of predictable patterns in growth. These predictable patterns lead past various stages or levels which are sequential in nature. Furthermore, these stages generally occur in a hierarchical and often irreversible progression (King & Teo, 1997). Once a certain stage has been reached, it is unlikely that one would revert to a lower level. Each stage can be viewed as a set of problems that has to be solved to enable a transition to a next stage (Kazanjian & Drazin, 1989). In this context problems can also be viewed as tasks or environments that have to be completed or created. The specific problems to be solved may differ between stage models, but all models suggest that once the problems of one stage have been solved, new problems arise that prohibit a transition to the next stage (Greiner, 1972). These problems create unique demands that need to be met, by developing certain capabilities within the organization for example. A fit has to be created between the current situation and a future stage. The gap between these two stages is created by the problems that have to be overcome. Stage models in current literature are partly descriptive, partly predictive and partly normative. The number of purely descriptive models is limited since most models contain stages that are not easily reached in practice. In their early stages the models tend to be more descriptive, whereas in their later stages, they become more predictive and normative (Coursey & Norris, 2008). The majority of the models consist of four or more stages. Many of them are cumulative, meaning that the next stage can only be achieved once the previous ones are also fulfilled (Lee, 2010). The majority of these models however, do not explicitly state that all previous stages have to be acquired in order to shift to a higher level. Plomp and Batenburg (2010) describe the accumulation assumption when referring to this attribute in models of growth. The majority of the stage models have emerged from daily practice and experience. There often lacks a theoretical background on which these models are based and they mainly rely on empirical evidence to support them. Furthermore, many models focus on describing the characteristics of the stages without paying attention to the transition between the stages. Preconditions and capabilities needed to move from one stage to another are often neglected, making them less interesting to use in practice. Stage models in e-government Within the e-government domain several stage models have been suggested. These models take on different perspectives and describe e-government development on different levels. A well-known and one of the first maturity models for e-government growth is the Layne and Lee model (Layne & Lee, 2001). It consists of four stages namely (1) cataloguing, (2) transaction, (3) vertical integration and (4) horizontal integration. The first stage, cataloguing, simply exists of governments having several web pages with non-transactional information. In the second stage, transaction, the static web pages evolve into pages where electronic transactions take place. Rather than just automating and digitizing existing processes, the third stage is characterized by integrating systems within similar functional walls but across different levels of the government. This form of integration is called vertical integration. The final stage is involves horizontal integration, where government agencies integrate their systems across different functional walls and thereby become fully integrated (Layne & Lee, 2001). An overview of all e-government related stage models identified along with their characteristics is shown in Table 2. Other models start with similar stages, such as the model of Heller and Bélanger (2001) and Ronaghan (2001), but go even further by stating that in the end e-participation by citizens will also be added to the e-government concept. While all these models describe e-government maturity, each uses different concepts and perspectives to illustrate this (Lee, 2010). The two models described here, 24

25 mainly focus on the front-office of e-government, the actual service level or level of presence towards citizens and businesses, while several other perspectives can be identified in literature. Table 2: E-government maturity models analyzed (condensed table) Author(s) Year Main focus Process oriented Nr. stages Grijpink 1999 G2G Yes 5 Baum and Di Maio, Gartner 2000 G2C/G2B No 4 Layne and Lee 2001 G2C/G2B Yes (Challenges for each stage) 4 Hiller and Bélanger 2001 G2C/G2B No 5 Silcock, Deloitte 2001 G2C/G2B No 6 Ronaghan, UN 2001 G2C/G2B No 5 Wescott 2001 G2C/G2B No 6 Netchaeva 2002 G2C/G2B No 5 Chandler and Emanuels 2002 G2C/G2B No 4 Peristeras, Tsekos and Tarabanis 2002 G2G No 4 West 2004 G2C/G2B No 4 Siau and Long 2005 G2C/G2B No 5 Wauters, EU Commission 2006 G2C/G2B No 4 Andersen and Henriksen 2006 G2C/G2B No 4 Papantoniou et al G2C/G2B Yes (change management) 4 Gottschalk and Solli-Sæther 2008/2011 G2G Yes (developing capabilities) 4 Sarantis, Charalabidis and Psarras 2008 G2G Yes (developing capabilities) 5 Klievink and Janssen 2009 G2G Yes (dynamic capabilities) 5 Janssen 2010 G2G Yes (capabilities) 4 quadrants Perspectives Many stage models show e-government from the customer perspective, the government providing services to its citizens and businesses (G2C/G2B). Within the customer perspective, several models focus on the level of service provided by the government, the different levels of communication/presence of the government and the level of participation. 25

26 Other stage models focus more on the integration of the back office. They describe the development of interorganizational collaboration between different government agencies. These models take on a different perspective that focuses on government organizations providing services to each other (G2G). Interoperability between different government organizations plays an important role in these models. They look at the development of for example organizational, semantic and technical interoperability between organizations. There are however only few models that take on a G2G perspective. Examples of these types of stage models are the models by Gottschalk and Solli-Sæther (2008) and Klievink and Janssen (2009). Lee (2010) also identified and analyzed 12 e-government maturity models. He made a distinction between a citizen/service theme and an operation/technology theme that were apparent in the models. He also noticed that different perspectives such as technology, organization, management and politics are divided over the different models and there is no model comprehensive enough to contain all of them. The G2C/G2B models mostly use a citizen/service theme which are sometimes combined with a technology theme to describe the technologies that are used to provide a certain service. The G2G models predominantly have an operation/technology theme and discuss the different technologies that create interoperable organizations. Next to the G2C/G2B versus G2G perspective, there is also a difference between outcome or process oriented stage models. Many of the current stage models describe what a certain stage looks like. Characteristics are described for each stage without looking at specific processes between stages. Some models go further by not only describing the characteristics of the stages, but also the processes that lead to transitions between stages. One of the few examples of a process oriented model is that of Klievink and Janssen (2009), who use dynamic capabilities to describe transitions between stages. Levels Another distinction between the various stage models in the e-government domain can be made on the level or scale on which they describe the developments. Especially the earlier models of e- government describe the development of e-government on a global scale, by describing how the development of e-government on national level will unfold for example. An advantage of these models is that they give a clear overview of the possibilities of e-government, but they are less practical in the implementation of e-government, due to lack of detail. An example of these kinds of models is the Layne and Lee model (2001). Other models also describe the growth from a more local perspective, such as on the level of a municipality, making the model more practical in use. Figure 7 shows a positioning model for the current maturity models in e-government research. The model is divided into two dimensions. One dimension is based on the orientation in service providing, if a model mainly describes the front office of e-government, G2C and G2B services, or if it describes the back office processes in e-government. The other dimension describes if the model only describes the characteristics of the stages or if it also describes the processes and preconditions that lead to another stage. Especially on the G2C/G2B side, the majority of these models only discuss the characteristics of its stages instead of also describing how to go from one stage to another. On the G2G side two observations can be made. First of all the number of e-government models that focus purely on G2G development is scarce and secondly the models that do focus on G2G are largely process oriented. However since there are only four models focusing on G2G, no further conclusions can be made based on the second observation. 26

27 Figure 7: Positioning model for e-government maturity models Boundaries and transitions There are several ways to define the boundaries between different stages. Although the majority of the stage models do not explicitly describe how these boundaries can be distinguished, it forms an important aspect of a stage model, in order to describe the transition between stages. Solli-Sæther and Gottschalk (2010) describe the use of benchmark variables to indicate the stage in which an organization currently resides. The measurement of these variables are often carried out by using a Guttman or Likert scale. For this approach certain benchmark variables have to be determined and defining these variables and making them quantifiable can be challenging. There have to be clear criteria for selecting a certain variable, which should be explainable from its context. Klievink and Janssen (2009) introduce a different concept, using discontinuity to mark the boundaries of a stage. Discontinuity in their view is seen as the presence or absence of change. They set boundaries by distinguishing periods of incremental change and substantial change. The stages can be viewed as periods in time in which there are only incremental changes noticeable in the unit of analysis. This can be viewed as a period of relative stability. Once there is a substantial change (a discontinuity), this can be seen as a period of transform, in which the unit of analysis moves towards a higher stage. When the change decreases again and returns towards incremental change, the next stage is reached and the period of transform ended (Figure 8). 27

28 Figure 8: Stage model presented as change in time (Klievink & Janssen, 2009) These periods of change are propelled by a concept of dynamic capabilities. Klievink and Janssen (2009) describe these dynamic capabilities as prerequisites for a shift towards a higher stage. For each stage, certain capabilities must be developed to overcome the problems that each stage has. Developing these capabilities creates a substantial change from the current situation and thereby thrives the shift towards a higher stage. Different stages can require a different set of capabilities to be developed. This makes the capabilities dynamic, where for one stage the focus may be on developing organizational capabilities, whereas for the next stage technological capabilities may be more important for growth. Conclusions Within the field of e-government a total of 19 stage models were identified. The majority of these models were created in the early 2000s and focused on broadly describing e-government development. The majority of these models and especially the early ones describe these developments from a customer perspective (G2C/G2B). These describe the different service levels that the government provides. Only four models were identified that focus on collaboration between government organizations (G2G) and thereby described the back office of e-government. These were largely created in recent years and can be seen as a reaction on the large amount of G2C/G2B models. The G2C/G2B models are largely outcome oriented, meaning that they describe the outcome or characteristics of a certain stage in e-government without describing the processes and preconditions that lead to that stage. Process oriented models are predominantly found among the G2G models. The process oriented models describe the boundaries between stages and how transitions take place. Capabilities are mentioned to describe prerequisites that are required to make a transition between different stages. The recent rise in G2G stage models shows that there is a need to describe the development of interorganizational collaboration using ICT between government agencies. The currently available G2G stage models give an idea of the issues that arise when dealing with interorganizational collaboration and its development. To further research this domain within e-government and stage models, the next chapter will focus on interorganizational collaboration. Relevant factors of interorganizational collaboration, that are partially discussed in the found G2G models, will be further uncovered to lay a foundation for the design of the artifact in the third part of this thesis. 28

29 2.4 Interorganizational collaboration An important aspect in realizing e-government at the front-office, is creating an effective back office. Within the e-government domain, several organizations in the public sector collaborate with each other to provide a joint service. Both on the organizational side and on the technological side, this collaboration has to grow to realize effective service provisioning. To describe this growth and the changes it entails, the organizational context has to be described first. Networks form an important concept to describe the organizational dimension in public administration. The delivery and management of public services increasingly rely on complex networks of interdependent organizations. Network governance is seen as one of the alternatives to the traditional procedural bureaucracy (Pardo & Jiang, 2007). Networks of organizations in public administration are mainly researched based on organizational studies. These view networks as a non-hierarchical, non-market form of organization in the public sector and see them as an alternative to bureaucracy. Bureaucracy can be described as a large, formal organization with hierarchical structure and division of labor. They are considered as highly efficient for performing standardized activities. However, there are also several downsides to this structure, such as routinization, resistance to change, overreliance on regulation, over conformity and trained incapacity (Pardo & Jiang, 2007). Networks on the other side are based on relationships, mutual dependency and norm reciprocity. Organizations in a network can be depicted in a graph as nodes, where lines represent their interrelations (such as in Figure 9). They furthermore have open-ended relational features that greatly improve their ability to transmit and learn new knowledge. These capabilities make them suitable in situations where efficient information exchange with reliable information is necessary. Networks can furthermore adapt to unexpected environmental changes, making them more flexible than bureaucracies and better able to control resource dependencies (Pardo & Jiang, 2007). Bekkers (2005) describes that viewing a group of collaborating government organizations as a network is increasingly valued as a means of steering complex societies in terms of co-operative production and management. Several networks emerged in which separate, relatively autonomous organizations execute adjacent or even slightly overlapping administrative tasks. Janssen (2010) further indicates the developments that have contributed to the rise of networks in governments. He mentions the following three: - increased contracting out of services to semi-autonomous agencies and private parties. - increased collaboration to improve efficiency and profit from each other s knowledge, resources and capability by avoiding duplications of efforts. - integrated service delivery requiring public organizations to collaborate to provide services provided by multiple organizations as it was provided by one organizations. The collaboration as described in the e-government domain also resembles collaboration in a chain as described by Grijpink (2009). To clearly define a chain of organizations in the public sector, it has to be explained within its context. The concept of a chain is commonly used to describe a variety of concepts. In this context a chain is defined as large scale collaboration between a large number of independent organizations and professionals. Societal chains produce a collective, immaterial product for social security, health or safety and government organizations are often active in these chains. These chains exist in various domains and in various configurations. There are usually huge numbers of independent organizations and professionals collaborating in ever changing combinations, depending on the cause of collaboration. Increasing specialization make organizations more dependent on each other and raise the importance of chain collaboration. There is an increase in hectic due to societal demand along with an increase in expectations which demands more coordination in the chain (Grijpink, 2009). 29

30 Grijpink (2009) describes chain collaboration as a reaction on a dominant problem within the chain. A dominant chain problem is a problem that affects the chain as a whole, endangers the functioning of the chain and may result in discrediting the chain. This dominant chain problem cannot be solved by just one or a few partners within the chain. It can only be solved when the chain co-operates as a whole. The dominant chain problem has a dynamic nature, over time it can shift towards a different dominant chain problem. Interorganizational networks in the public sector are characterized by reciprocal interdependence. Every organization in the network has ties with one or more other organizations and communication goes in both directions. Figure 9: Interorganizational network with reciprocal interdependence Grijpink (2009) also makes a distinction between different levels of decision making. Firstly, on the level of individual organizations, decisions are made that seem rational on the level of that individual organization but may not be on the level of the entire chain. It does not seem self-evident for organizations to reason from a chain perspective instead of from an individual perspective. Subsequently, it should be logical to distinguish a chain level. Chain collaboration theory is based on the presence of bounded rationality on the chain level. Unclear or contradicting objectives lead to unexplainable and often irrational decisions on the chain level. What may seem to be a rational decision on the level of a single or a few organizations, may turn out be completely irrational on the level of the entire chain. Because there is a lack of overall authority, this irrational decision making is difficult to counter. With these different levels, so called fallacies of the wrong level are commonly made. Experience and practices on the level of a single or few organizations are applied to the whole chain. These are bound to fail, since the rules of these levels do not apply. Therefore it is important to distinguish two levels of analysis, one for the ground level (individual organizations) and one for the chain level (Grijpink & Plomp, 2011). Although collaborating with each other and being dependent on resources controlled by other organizations, each organization maintains a certain autonomy, by having a set of specific resources in the form of information, knowledge or competences. There is no organization that can enforce its will on all organizations in the network, both Grijpink (2009) and Bekkers (2005) acknowledge this. There is a unique sphere of influence ownership and control over information, which makes negotiating over exchanging information more difficult (Bekkers, 2005). Organizations fear that they might lose their independence and rather want to increase dependencies that other organizations have on them. When deciding over shared resources, such as a shared information system, a struggle arises over the type and amount of information that is shared with other organizations. Bekkers (2005) calls this process information politicking. Janssen (2010) describes these networks of government organizations by using the term public service networks (PSNs). He states that the development of a PSN is a difficult task that requires time 30

31 to create the necessary authority structures, procedures and mechanisms to be able to collaborate. For a PSN to persist over a long period of time, uncertainty regarding the continuity should be avoided according to Janssen (2010). Stating that a shared infrastructure should be created that is useable for all organizations involved, to prevent organizations to remain as a silo. Janssen (2010) furthermore states that formal arrangements that ensure quality and performance seem to be necessary to stimulate the maturity of the network. In order to reach an environment in which these government organizations collaborate as a network, several change management initiatives are describes by Burn and Robins (2003). For the organizational dimension an important issue to address is the network relationship balancing. Organizations should recognize the need to manage both internal and external forces to effect change (Burn & Robins, 2003). They use the constructs of interorganizational linkages and cross functional cooperation to describe the level of interorganizational collaboration in a network. Since the negotiation over information exchange can be quite difficult, especially when there are multiple parties involved, according to Bekkers and Homburg (2002) a different approach should be taken to reach interorganizational collaboration in networks. Conventional approaches such as applying project management may be found to be less effective than approaches that focus on defining a collective goal and identify situations that are fruitful for the majority of the organizations. Issues in various arenas influence each other, such as system design, data ownership and legislation, in order to successfully collaborate, these issues should be linked to each other and taken into account when working towards collaborative ICT initiatives. To deal with these issues, Bekkers and Homburg (2002) suggest that business process management approaches may be more suitable to apply than project management, since they provide interesting strategies to deal with the challenges in this field. Hellman (2010) furthermore indicates several organizational barriers that have to be overcome for organizations in a network to successfully collaborate. These barriers were translated into opportunities, which include the broader involvement in modeling the business processes, introducing performance indicators for these interorganizational processes and the use of a knowledge base to learn from previous projects. Organizational growth in government chains Chain computerization theory uses a method to analyze to what extent a shared information system is needed and feasible within a chain of organizations. Next to describing the goal of the chain, the coordination mechanisms that are applied and the information that is shared throughout the chain, there is another instrument to measure the level of collaboration in the chain. The so called cooperation profile that uses five stages to describe this collaboration (Grijpink, 2009). The stages are sequential in nature and start with a phase in which there is only informal consultation. The next phase is characterized by more formal consultation, but only on peripheral issues without any decision making on chain level. It requires participation from the chain partners and some sort of agenda. The third phase is characterized by joint decision making on chain wide supportive issues. This is an important moment in the collaboration but at the same time hard to realize. The different parties with their own perspectives and goals have to agree on issues that affect them all. The more parties involved, the harder it will be to reach an agreement. After this stage, the collaboration reaches a level on which ad hoc joint action becomes feasible, in the shape of a chain project. Finally, the cooperation reaches a state in which organization are so far integrated that they can be viewed as a common network organization. Another model that describes the organizational growth in organizations is the interorganizational interoperability model by Clark and Jones (1999). This model describes five phases of organizational interoperability, ranging from independent to unified and describing the characteristics at every stage. It was created to describe the level of interorganizational collaboration between government organizations and is often referred to in e-government literature as a tool to describe a part of e- government maturity. 31

32 Conclusions The creation of an effective back office is crucial to provide effective e-government services. The focus on efficiency in service delivery has caused increased collaboration between different organizations, both governmental and private and demands a suiting organizational structure. The concept of a network or a chain of organizations is often used in e-government literature to describe the structure of the collaboration in the back office. Knowing how such a network develops, what barriers can be expected in the development and how they can be overcome are important for the overall development of interorganizational collaboration. An interesting observation is that although different organizations have to collaborate to jointly provide a service and they partly depend on each other, they all maintain a large degree of autonomy. There is not a single organization that can enforce its will on all other organization in the network. This creates a situation in which negotiation over ownership and control of information decides if and how shared information systems will be introduced. There are however also certain forms of interorganizational collaboration enforced by legislation and political motives, which differentiates the collaboration in the public sector from that in the private sector. These characteristics should be taken into account when describing the development of interorganizational collaboration in e-government. 2.5 Interoperability To create efficient public services Kubicek and Cimander (2009) mention two ways of achieving this, either by the integration of back-offices through centralization of data and services, or by standardizing and clearing, i.e. interoperability. There are several reasons why public administrations cannot fully centralize their tasks and data. These often have to do with legal issues in which central data storage is controversial or forbidden or due to the amount of stakeholders that have their own systems and data, making it difficult to reach agreements on shared services. In these cases interoperability forms a preferable alternative to achieve an efficient back-office, where individual organizations maintain a large part of their internal systems and data. By using the interoperability concept to create a seamless exchange of data between government agencies, the information systems in the back-offices of these agencies have to be linked up (Kubicek & Cimander, 2009). Because of its important role in the development of interorganizational collaboration in e-government and the many definitions that can be found in literature, the interoperability concept will first be clarified. Initially the interoperability concept was mainly described from a technical perspective and was concerned with the coupling of diverse and disparate IT systems. The Institute of Electrical and Electronic Engineers (IEEE) defined interoperability as "the ability of two or more systems or components to exchange information and to use the information that has been exchanged." (IEEE, 1990). In this context, interoperability mainly refers to ensuring that data can be exchanged between diverse systems that are located in different locations. This includes creating and agreeing on the signals exchanged, such as protocols for data transfer. Furthermore decisions on processing the received data have to be made, by creating and agreeing on standardized data exchange formats such as XML. However, interoperability in an e-government context goes beyond this technical dimension, by also including the re-organization of different administrative processes, aligning different organizational structures and agreeing on the meaning of what is exchanged. These aspects make the problem more complex and require a broader definition of the interoperability concept. A definition of interoperability with a broader scope comes from the European Interoperability Framework for pan-european e- Government Services (EIF) (European Commision, 2004). The second version of this framework (European Commision, 2010) defines interoperability as follows: " the ability of disparate and diverse organizations to interact towards mutually beneficial and agreed common goals, involving the sharing of information and knowledge between the organizations via the business processes they support, by means of the exchange of data between their respective information and communication technology (ICT) systems". 32

33 This definition takes a much broader perspective on interoperability than the traditional technically oriented definitions and reflects the different aspects that have to be taken into account in an e- government setting. An important addition is the emphasis on mutually beneficial and agreed common goals, which is the main driver in e-government for individual organizations to collaborate. Next to sharing information and knowledge, this definition also mentions the business processes involved, which have to be taken into account to not only make a fit from the technical dimension, but to align the business processes that are supported by the IT systems as well. It was mentioned before that e-government as a research domain draws from various disciplines, this also applies to the interoperability issue. The body of scientific literature on interoperability is spread across different research areas such as information systems research, computer science, public administration, organization science and information management. These various domains tackle the interoperability problem from both technical and non-technical perspectives (Scholl & Klischewski, 2007). The combination of different research areas also adds to the complexity of the problem, different disciplines have to be combined to come to a suitable solution. Scholl and Klischewski (2007) state that the extent and complexity of the challenges surrounding interoperability initiatives in public administration are not yet fully understood in theory and practice. Many interoperability initiatives fail due to this complexity, which prevents creating more sustained levels of interoperability (Soares & Amaral, 2011). Since interoperability is an important success factor for overall interorganizational collaboration in e-government chains (Pardo, 2011), it is important to further study the development towards interoperable government organizations. Interoperability in e-government From an e-government perspective interoperability in government is driven by an increasing demand on better quality of service, not just by citizens and businesses, but also by other government organizations (Abramowicz, Bassara, Wisniewski, & Zebrowski, 2008). Various individual government agencies have to share information and link their (administrative) processes. However, existing information systems and infrastructure do not sufficiently support this interaction. This inevitably requires the existence of interoperability solutions between different information systems and organizations in public administration (Soares & Amaral, 2011) (Scholl & Klischewski, 2007). Interoperability is therefore considered to be a critical success factor to make progress in the online provision of public services (Pardo, 2011). The last few years there has been written a substantial amount of literature on the interoperability issue in e-government. The literature first mainly focused on the technical aspects. As concluded earlier, the interoperability problem goes beyond the technical level and appears to be more complex. Pardo (2011) recognizes the overreliance on the technical dimension of interoperability as a barrier to e-government interoperability. He views e-government interoperability as a set of policy, management and technology capabilities that are necessary for chains of government organizations to achieve successful information sharing. The challenge is to identify these different levels and create a fitting approach to tackle the problem. Later studies took on this challenge and started to broaden their focus more towards the other dimensions of interoperability. Abramowicz et al. (2008) adopted the semantic and organizational dimensions from the EIF and described the governance issues for realizing interoperability in e-government as well. Scholl (2005) takes an even broader perspective by stating that interoperability is achieved by solving issues on legal, policy, organizational, managerial, social process and technological levels. Literature on interoperability in e-government shows that there is no doubt that achieving interoperability between different organizations in public administration is a complex process that requires different disciplines to be combined and applied on various dimensions. Some initiatives have been taken to increase the chances of successful interoperability in e-government, such as Archmann (2007), who identified several critical success factors for achieving interoperability in e-government. However, tools to structure and analyze the relevant dimensions are a first step to solve the interoperability issue. 33

34 Interoperability frameworks in e-government Several frameworks have been proposed to analyze and solve the problem of creating interoperable systems and organizations, both general and e-government specific. They can be used to analyze the various dimensions and aspects that come into play when working towards an interoperability solution. From a more practical perspective interoperability frameworks can be seen as tools for guaranteeing interoperability. They function as a reference for the basic technical and organizational specifications that all organizations involved should adopt (Guijarro, 2007). They furthermore give guidance to practitioners as of what to consider and to do in order to enable the seamless interaction (Kubicek & Cimander, 2009). Kubicek and Cimander (2009) review three main interoperability frameworks in e-government. They start with the European Interoperability Framework for pan-european e-government services (EIF), a framework that is often mentioned in e-government literature on interoperability. This framework is part of the IDABC programme, managed by the European Commision. It was initially created to encourage the delivery of cross-border public sector services to citizens and enterprises in Europe and to improve the efficiency of the collaboration between European administrations. Many national governments adopted the principles of this framework in order to create interoperability frameworks on national level. The EIF describes three main dimensions of interoperability, a technical, semantic and organizational dimension. These three dimensions form the basis of the majority of the interoperability frameworks and will be further discussed. Technical dimension The technical dimension deals with traditional interoperability concerns of linking computer systems and services. This covers among others open interfaces, interconnection services, data integration and middleware, data presentation and exchange, accessibility and security services (European Commision, 2010). Agreements therefore have to be reached on protocols for data transfer and standardized data exchange formats (syntactic level). Technical standards for the infrastructure are fairly mature, protocols for network communication such as TCP/IP are used in many organizations (Kubicek & Cimander, 2009). Other technical standards are however less evident. The area of data representation for example is concerned with the selection of data format standards, such as character sets (European Commision, 2010). Furthermore there is the selection of platforms, databases and database models and the use of programming languages for shared (web)applications. In certain government domains there are software packages available that will avoid many of these technical interoperability barriers, these package often only support limited interaction between different parties in the network and do not cover domain specific needs. Agreements should therefore also be reached on standards that are aimed at a specific domain and cannot be solved by standard solutions. Since these standards have to be created, agreed upon and subsequently maintained, it shows that different levels of technical interoperability can be distinguished. Semantic dimension The semantic interoperability level addresses issues on the level of information. Once the data is exchanged, it has to be processed and interpreted by the receiving organization to be of use. The EIF defines semantic interoperability as enabling organizations to process information from external/secondary sources in a meaningful manner (European Commision, 2010). Agreements have to be reached on meaning and format of the information that is exchanged. To achieve this, common data elements and protocols for sharing and using information need to be established as well as common directories, data keys and ontologies. In the best case an information management lifecycle will be present in the participating organizations to monitor the quality of information exchange. The semantic level is challenging to implement, as necessary tools and methods are being developed, but have not been regularly applied before on large scale. The large scale on which these agreements have to be reached also poses a big challenge for overall e-government interoperability (European Commision, 2010). When discussing the semantic dimension, ontologies often play an important role. An ontology can be defined as a formal specification of a conceptualization and enables concepts from the real world to 34

35 be processed by machines. Ontologies can be further divided into domain ontologies, these represent abstract models of how people think about concepts related to a particular area (Orgun, Dras, & Nayak, 2008). A domain ontology in their view, consists of concepts or terms related to the domain in a hierarchical fashion. Concepts have attributes associated with them, defining their properties and relationships with other concepts. When two concepts have similar attributes but use different syntax to communicate instances of such concepts, the interoperability problems are situated on the technical dimension, since a standard syntax has to be defined to ensure seamless information exchange. However, when two concepts have been defined with different attributes and only partly overlap, the problem resides on the semantic dimension. In this case an agreement has to be reached between different parties to define the exact attributes of a concept, eventually leading to a new ontology. Organizational dimension Finally, the organizational interoperability level describes the linkage of different business processes that are supported by the different systems. In the EIF the organizational dimension is described as reaching detailed agreements on how processes will interact (synchronize/cooperate) in order to deliver public services where needed. It deals with aligning workflows between different organizations to ensure that the data and information does not only efficiently reach other organizations, but connects to the processes in these organization as well. In order to do this, the business processes in each organization should be described in an agreed upon manner. This means that business process management can be applied in various degrees. Measures that have to be taken to achieve organizational interoperability are agreeing on a standard that has to be used to describe the (cross-border) business processes, creating a business repository of business processes and best practices to motivate stakeholders to exchange information and reuse the best practices. Kubicek and Cimander (2009) describe the development in each of these dimensions and conclude that the technical & syntactic are fully developed in theory and are applied widely in practice. The semantic interoperability dimension is developed in theory but from an implementation view is still in an early stage. This concern is also put forward by the European Commission (2010), which points out that semantic interoperability is mainly difficult to reach on a large scale. Other dimensions A second version of the framework was created (EIF v2.0), expanding these three dimensions with a legal and political view. Legal interoperability deals with the legislation concerning the data sharing between the cooperating organizations. This legislation mainly applies to data that originates in one organization and is transferred and stored in a different organization. Since EIF also focuses on collaboration between organizations in different EU member states, in this case the legal dimension covers differences in legislation between different countries involved. Legislation is usually influenced by decisions that are made by the government. In this way, the legal dimension links to the political dimension. Interorganizational collaboration is influenced by decisions taken in politics as well. EIF describes the political dimension (or context) as cooperating partners having compatible visions, and focusing on the same things. Any changes to these visions should be coordinated and aligned appropriately (European Commision, 2010). An overview of the interoperability dimensions as presented in the EIF (European Commision, 2010) is depicted in Figure

36 Figure 10: Interoperability Framework EIF v2.0 (European Commision, 2010) The other two frameworks that were analyzed by Kubicek and Cimander (2009), describe the three main dimensions but combine these with additional interoperability aspects. The European Public Administration Network (EPAN) describes the customer support at the front office and governance aspects as well. The European standardization institute ETSI split the technical dimension, resulting in an additional dimension called syntactic interoperability. In the EIF this syntactical dimension, concerning data exchange format is incorporated in the technical dimension. While there are global standards and protocols for the levels of technical and syntactic interoperability and recognized concepts and methods exist for the semantic interoperability, very heterogeneous elements and aspects are assigned to the level of organizational interoperability, which are described only vaguely or which formulate requirements instead of options for action as at the other levels (Kubicek & Cimander, 2009). Table 3 shows the progress in each of the dimensions according to an assessment of Kubicek and Cimander (2009). Table 3: Four levels of interoperability (Kubicek & Cimander, 2009) Level of interoperability Aim Objects Solutions Technical/Syntactic interoperability Semantic interoperability Organizational interoperability Technically secure data /Processing of received data Processing and interpretation of received data Automatic linkage of processes among different systems Signals/Data Information Processes (workflow) Protocols of data transfer/ Standardized data exchange formats, e.g. XML Common directories, data keys, ontologies Architectural models, standardized process elements (e.g. SOA with WSDL, BPML) In Europe, interoperability frameworks have shown up as a key tool for interoperability in the deployment of e-government services, both at national and European level. They initially focused on 36

37 technical interoperability, such as syntax issues, but recently inclusion of specific issues such as semantics has started. (Guijarro, 2007). Within the European union there are several initiatives on national level to improve e-government interoperability. In the Netherlands there is an initiative called OSOSS, which stands for the Open Standards and Open Source Software in Government (Archmann, 2007). It contains a list of several standards that should be used when applying certain technology, such as XML, SOAP and TCP/IP. Another initiative in the Netherlands to improve the interoperability in e-government projects is the Dutch Governmental Reference Architecture called NORA ( Nederlandse Overheid Referentie Architectuur in Dutch), which contains common development agreements for the design of e- government information systems. Similar initiatives exist for regional (PETRA) and local (GEMMA) e- government initiatives. These frameworks however mainly pay attention to the technical dimension and leave other dimensions such as the semantic dimension largely out of scope (Abrahamse, Wisse, & Oude Luttighuis, 2009). Furthermore, these frameworks mainly provide generic interoperability solutions, whereas there is no guidance for domain specific interoperability issues. Interoperability models of growth in e-government Next to the interoperability frameworks, there are also several models that distinguish different levels of interoperability. Within the e-government research domain, there are only few maturity models that focus on interoperability. From the four G2G e-government maturity models identified earlier, two models focused on interoperability. One relatively often cited model by Gottschalk and Solli-Sæther (2008) and one by Sarantis, Charalabidis and Psarras (2008). Gottschalk & Solli-Sæther (2008) created their maturity model for interoperability in e-government based on several stages that were labelled according to a meaningful characteristic. The different levels of this model do not necessarily measure the level of connectedness between organizations, but rather seem to look at the level of collaborative achievements that can be made at a certain level. The first stage for example is focused on aligning work processes, the second stage focuses on the exchange of knowledge, the third stage deals with joint value creation and the final stage is aimed at aligning strategies of collaborating organizations. An organizational and semantic dimension were used to describe these different levels. Solli-Sæther (2011) has extended the initial model by adopting the three interoperability dimensions from the European Interoperability Framework (EIF) and thereby adding the technological dimension to the existing two. Although Gottschalk and Solli-Sæther (2008) speak of developing capabilities to make a transition towards a higher stage, these are not further indicated for each stage. Sarantis et al. (2008) created a maturity matrix based on the LISI model, a maturity model for technical interoperability (C4ISR Architectures Working Group, 1998). They adopted the five stages from the LISI model but created two additional dimensions next to the technical one. The semantic and organizational dimension were adopted from the EIF framework to create three dimensions for every level of interoperability, equal to the one of Solli-Sæther (2011). The difference with the Solli-Sæther (2011) model is that this model bases its stages on the LISI model. The LISI model distinguishes the isolated, connected, functional, domain and enterprise stage. The model furthermore uses so called interoperability attributes to measure the current level of maturity. Sarantis, Charalabidis and Psarras (2008) state that the model is applicable for both local, regional and national administrations. The interoperability attributes that are assigned to the different levels however make this model less useful for local and regional networks of government agencies. In the organizational dimension for example there are attributes for interoperability not only at local level, but also at national and international level. Such attributes seem irrelevant when looking at a network of government agencies on a local level. The organizational dimension furthermore lacks the focus on interorganizational business and work processes that one would expect to be described at the different levels. Although the dimensions and levels in the model of Sarantis er al. (2008) seem justifiable to use in describing interorganizational collaboration between government organizations, there are also some remarks to be made on the model. There does not seem to be a clear scope as to what kind of network of government organizations the model focuses on. This reflects in the attributes they use to describe each level of collaboration. The organizational dimension could be defined more clearly and is not suitable to describe different levels of collaboration on a local or regional level. 37

38 Conclusions Achieving interoperability is an important success factor for successful interorganizational collaboration in e-government. Where the focus was mainly on technical interoperability in the past, the focus has now been expanded to take semantic and organizational aspects of interoperability into account. The European Interoperability Framework (EIF) is becoming an important reference when interoperability issues are discussed and is found to be used in various e-government papers, including two G2G e-government stage models. From the analysis on stages models in e-government, four models were identified that focus on G2G collaboration maturity. From these four models, two also took interoperability aspects into account. Taking a closer look at these two models shows that they have both similarities and differences. Both models adopt the three dimensions from the EIF. Solli-Sæther (2011) however, uses these dimensions to describe the different levels of collaborative achievements from government organizations. Although this is an interesting outlook on the interorganizational collaboration issue, it is more interesting to look at the prerequisites of these achievements. Sarantis et al. (2008) apply this perspective to their model which uses the same dimensions but different levels to describe the developments. The downside to this model is that it does not define the organizational dimension clearly and there is no clear scope on the level. They describe local, regional and national interoperability instead of focusing on just one of these levels. The G2G models and interoperability frameworks will be used primarily as input for the artifact design in the next chapter. The other e-government models will further be used to combine several aspects and add to the existing models. 38

39 3 Artifact design As discussed in the research approach chapter, the design science research method by Peffers et al. (2008) will be used to create the stage model. The schema of the design science research method is depicted again in Figure 11, where the blue highlighted area shows which steps have been completed so far. In the introduction of this thesis, the problem being researched has been identified and a motivation by means of the scientific and social relevance has been given. Further exploration of the problem and defining possible solutions were performed during the literature review. The results of this review were formulated in the second part of this thesis. The third chapter runs parallel to the third step in the design science research method. It describes the design and development of the artifact, in this case the maturity model. Figure 11: Design Science Research Method by Peffers et al. (2008) The literature review has laid an important foundation for the development of the conceptual model. It has answered several sub research questions and showed the progress that has been made in various research areas of e-government research. Important conclusions from the literature review with regard to the conceptual model will be discussed in the following sections. First the selection of dimensions that will be used to describe the interorganizational collaboration in government networks will be explained and these dimensions will be described. Secondly the selection of stages and their characteristics will be described and the stages will be described. Finally the context of the model will be described in the final section. 3.1 Dimensions Selecting the dimensions The levels of interorganizational collaboration using ICT can be measured in different dimensions. From the literature review, several dimensions stood out. First of all the technical dimension is unmistakably an important dimension to consider when realizing collaboration through IT. From the interoperability frameworks that were studied, all of them focused on the technical dimension, which is discussed the most in e-government literature in general as well. In various literature and interoperability frameworks the organizational dimension is mentioned as well. Although there are differences in the exact definition of the concept. When focusing on the organizational dimension, the alignment of business processes between different organizations is most often mentioned. A dimension that is recognized more and more to be important in interorganizational collaboration through ICT is the semantic dimension. This dimension is part of a majority of the interoperability frameworks, such as the EIF and describes the information layer between different organizations. 39

40 Organization realize that it is no longer a matter of using technology to connect to each other, but agreement on the structure and meaning of what is exchanged have to be made as well. In the design of the dimensions of the new model, the decision was made to follow these three main dimensions. By deciding to select the technical, semantic and organizational dimension, the conceptual model adheres to the dimensions mentioned in the European Interoperability Framework (EIF). Because of their importance to interoperability (Kubicek & Cimander, 2009) these dimensions will be incorporated into the model first. Other dimensions were considered as well such as a political, governance or legal dimension. These dimensions are not selected to incorporate into the design of the maturity model itself, since they can be better used to the describe the context of the model. Labeling the dimensions The original three dimension in the EIF are quite generic and to make the model more specific the naming was adjusted to better suit the context. The definitions for each dimension will be largely adopted from this framework and are adjusted to the context of e-government networks where needed. Information System dimension: The technical dimension deals with traditional interoperability concerns of linking computer systems and services. This covers among others open interfaces, interconnection services, data integration and middleware, data presentation and exchange, accessibility and security services. (European Commision, 2010). Information dimension: Semantic interoperability enables organizations to process information from external/secondary sources in a meaningful manner. In practice, it will involve the establishment of common sector specific sets of data structures, data elements and protocols. The partners need to agree on meaning and format of the information to be exchanged. (European Commision, 2010). Process dimension: The partners need to reach detailed agreements on how their processes will interact (synchronize and cooperate) in order to deliver public services where needed. Organizational interoperability in practice means the seamless integration of business processes and the exchange of information that they manage between the organizations (European Commision, 2010). 3.2 Stages Next to defining the dimensions in which to measure the level of interorganizational collaboration, the stages were defined. The stages will represent levels in interorganizational collaboration that intensify with each additional stage. The division of the stages in the found maturity models was analyzed first. Since the new model will focus on G2G collaboration, these models will primarily be used (Table 4), the other models that were identified and focus more on G2C/G2B collaboration will however also be taken into account. Table 4: G2G models with their stages Nr. Peristeras, Tsekos & Tarabanis 1 Islands of Automation 2 Automated Process Chains Gottschalk & Solli- Sæther Sarantis, Charalabidis & Psarras Independent Klievink & Janssen Aligning Work Stovepiped applications Processes Sharing Knowledge Ad hoc Integrated organizations 40

41 3 Reengineering through Information Technology Joining Value Creation Collaborative Nationwide portal 4 Total Reinvention Aligning Strategies Integrated Interorganizational integration 5 Unified Demand-driven joined up government Number of stages From the models it shows that two of the four models maintain a number of five stages. The average amount of stages in all nineteen found models also lies around five stages. Although no literature was found that states that a division of a stage model in a certain amount of stages is required, the division into five stages can be chosen from a practical point of view. By defining five stages, two extremes can be defined, balanced by the third stage. The second and fourth stage, make a gradual shift between these extremes possible. Furthermore, for statistical purposes five stages can be practical when applying a five point Likert scale to determine certain capabilities. This model will therefore also use an amount of five stages. Labeling the stages For each stage, the models of growth found in the literature review were analyzed. Several requirements were set for the selection of the names for the stages. The name should represent a level of interoperability between organizations in an interorganizational network. The name should furthermore be applicable to each dimension in the model. This means that labels that specifically describe a level of interoperability between information systems, are not suitable to describe the level of collaboration on the organizational level and will therefore be discarded. With the use of five stages in the model, the first stage should furthermore describe a situation where there is little to no interorganizational collaboration, the final stage represents complete interorganizational collaboration and the stages between them should be labeled to describe the gradual shift between the two extremes. When looking at the naming of the stages and what these names entail, there are several similarities and differences noticeable in existing models. Aside from the Gottschalk and Solli-Sæther (2008) model, which focuses on collaborative achievements, the other models all describe a level of connectedness. These three models all start with describing a stage in which organizations in the network mainly operate independently. However they all end with different stages. When looking at the transitions between the stages, some models seem to have labeled their stages as generically as possible and the naming implies a gradual shift between the different stages. Other models mentioned specific applications in their stage naming, such as a nationwide portal (Klievink & Janssen, 2009) and reengineering through information technology (Peristeras, Tsekos & Tarabanis, 2002), making the naming quite specific. As discussed in the requirements before, these specific labels are less suitable to apply to the three different dimensions. With regard to the first stage, a label that represents a state in which all organizations in the network mostly operate independently without the presence of interorganizational collaboration. Some models such as the one by Peristeras et al. (2002) describe this situation quite specific by labeling it Islands of automation or Stovepiped applications. Others are more generic such as isolated or independent by Sarantis, Charalabidis & Psarras (2008). Independent seems to describe this situated clearly and can be applied to every dimension and is therefore most suitable to use in the artifact. In the second stage there are also some specific labels such as integrated organizations by Klievink and Janssen (2009). Others use connected or transaction, but it is unclear to what degree this connection or transaction takes place. The use of ad hoc is also used in several models and gives a more clear view of the degree of interorganizational collaboration at this point and can also be applied to every dimension. 41

42 The name for the third stage should resemble the transition from ad hoc collaboration in the network to a more coordinated form of collaboration. In the existing models there are diverse labels to describe this degree of collaboration. Some call it collaborative or integrated (Sarantis, Charalabidis, & Psarras, 2008), but with these names the level of integration still remains unknown and they might therefore be too broad. One model describes a stage of coordinated collaboration (Janssen, 2010), which is a more logical stage after the ad hoc collaboration and will therefore be adopted. In the fourth stage domain specific attributes can be added to the coordinated collaboration in the network. Several models such as Sarantis et al. (2008) name this stage integrated, however as argued before there can still be several degrees of integration. Same goes for transformation which also does not describe a certain degree of collaboration but rather a certain effect on this collaboration. Since the network in this phase will combine various organizations in a domain, this stage can best be labeled domain, which is also used in the LISI model. There are various labels possible to describe the final stage. When looking at the labels that are generic, several candidates are available, such as fully integrated, seamless, joined up government and unified. The label unified, will be adopted from Sarantis, Charalabidis & Psarras (2008) because it better describes how the individual organizations are connected to a degree that they can be seen as a uniform organization. Shifting between the stages In order for the networks to reach a higher level of interoperability, agreements have to be reached on standards for the system, information and process dimension. Archmann (2007) mentions several phases in defining such standards. In the preliminary phase there are no mutually agreed standards between organizations. A phase of drafting/agreeing on standardization follows, leading to initial standards in each of the dimensions. The next stage consists of applying these initial standards in practice. An evolving phase follows, in which adaptions will be made to certain standards while other will be maintained. The final stage will be reached when all standards are fully developed and evolved into stabilized, flexible standards and agreements. This process of evolving standards and agreements was applied as a foundation for the transition between stages. 42

43 3.3 New stage model for interoperability in government networks Now that the dimensions and stages are selected, their characteristics will be further described in the following section, an overview of the model is depicted in Figure 12. The properties will be described for each stage, in combination with the processes that lead to the following stage. An overview of the complete model with all stages and capabilities is shown in Table 5. Independent Ad hoc Coordinated Domain Unified System Information Process System Information Process System Information Process System Information Process System Information Process Creating opportunities Drafting/Agreeing on Evolving standards standards Figure 12: Interoperability maturity model for government networks Maintaining standards Independent stage This stage is characterized by a network of government organizations that share a common goal. This common goal mostly consists of providing a certain service to the public, and each of the organizations delivers a part of the solution. In the independent stage however, the majority of the organizations in the network mostly operates independently. The level of interorganizational collaboration to efficiently achieve the common goal and ensuring a high level of quality is minimal. Similar stages were identified in Klievink and Janssen (2009), Sarantis, Charalabidis and Psarras (2008) and in Peristeras, Tsekos and Tarabanis (2002). From the information system perspective every organization uses its own applications and data formats and is not able to exchange information that is created within the organization with other organizations or even with other departments in the same organization. These legacy systems were designed to be tightly coupled internally and therefore have little options to interact with external systems, resulting in inflexibility in the exchange of data (European Commision, 2010). These systems can also be described as islands of automation or silo s (Peristeras, Tsekos, & Tarabanis, 2002). The meaning of concepts differ between organizations and make information exchange difficult due to the inability to compare and integrate information coming from other organizations or even departments within the same organization. Since the technical abilities to exchange information are absent in this stage, there is no focus on developing semantic interoperability. Interoperability problems in the information dimension are closely related to efforts on the system level (European Commision, 2010). Duplicate information is stored in several different organizations and departments, due to a lack of information overview. The focus is only on the internal workflow and not on processes that go across the borders of the organization. There is no overview of the business processes between different organizations in the network. The focus on internal workflow prevents organizations in the network to make initial agreements on documenting interorganizational processes in the network. Documenting the business processes between the organizations in the network represents the first step in analyzing and aligning them. Shift towards the ad hoc stage: initial contact In this stage there is little to no progress made towards an interoperable network of organizations. The different stakeholders are often unaware of the information that is available in other organizations and that could be of use to improve the quality and efficiency of their service. The shift towards the following stage is made by creating a more outward faced mentality and starting small scale collaborations of organizations with other organizations in their immediate surroundings. 43

44 Ad hoc stage This stage is characterized by interorganizational collaboration in the network on ad hoc basis. Several organizations in the network start to collaborate on specific occasions, mostly with partners in their immediate surroundings and when collaboration is unavoidable in achieving the common goal. This form of interorganizational collaboration does not cover the entire network, is often temporary and is only supported by limited organizational frameworks and standards. From an information system perspective the ad hoc stage consists of peer-to-peer connections between a few organizations in the network. However, each organization still uses its own information systems and applications and there is no standardized format for data exchange. The state of information systems in the ad hoc stage is equal to that of the independent stage. The type and amount of data that can be exchanged is therefore also limited to the possibilities that the current systems offer. Viewed from the information perspective, different organizations still use their own concepts and attributes and connect these concepts where possible and when needed. A dynamic meta-model for ad hoc information exchange must be used to be able to effectively exchange information. Workflow and business processes from different organizations are aligned on the fly to support the ad hoc collaboration, similar to what Janssen (2010) describes in his ad hoc quadrant. This creates a first phase of awareness of business processes of the other partners in the network. Due to the ad hoc nature of the collaboration at this point there is no motive for the organizations to further align their processes. Small parts of the network are starting to become apparent, however initial agreements to document, analyze and align process across the whole network are still to be agreed on. Shift towards the coordinated stage: agreeing on initial standards In order to shift towards the next level of interorganizational collaboration, agreements on initial standards should be reached between the current organizations in the network. In the beginning of the ad hoc phase, there will only be interaction between several organizations in the network and there won t be any coordinated meetings between all the parties in the network. The first part of reaching the coordinated phase is identifying all parties and creating a platform for parties to exchange information and to ventilate their ideas. The second part of the process includes agreeing on initial standards for each of the dimensions as can be viewed in Table 5. Once initial standards for each of the dimensions are agreed upon and these standards apply within the entire network, the coordinated stage will be reached. Coordinated stage This stage is characterized by the presence of initial agreements and standards that enable and stimulate interorganizational collaboration throughout the network. The common goal that binds the organizations as a network is recognized and roles and responsibilities that lead to this goal are made clear. This starts an initial phase of coordinated collaboration between organizations across the network that will gradually evolve towards the end. Information systems of the different organizations are adapted to accommodate information exchange between different partners. Issues that should be agreed on are among others open interfaces, interconnection services, data integration and data presentation and exchange (European Commision, 2010). These initial standards apply to every organization in the network. Since information exchange with other organizations takes regularly place in this stage, several agreements on the structure and meaning of what is exchange has to be made. An important first step in the information dimension is made by creating a listing of information that is available at the various organizations in the network and which can be shared throughout the network. Furthermore agreed basic principles for the management of information in the network exist, including protocols for sharing and reusing network specific information. A low level ontology of the basic vocabulary used between the different organizations is agreed on, containing common sector-specific sets of data structures, data elements and protocols (European Commision, 2010). From a process perspective organizations need to reach detailed agreements on how their processes will interact (synchronize and cooperate), such as described in (European Commision, 44

45 2010). Shared goals are recognized between organizations and the accompanying roles and responsibilities are understood by all parties. Organizations in the network agreed on a standard to document the processes in the network. In order to align processes in the network, they first have to be properly understood by all parties in the network. Using this standard, a repository of business processes and best practices should be created to enable reuse of best practices throughout the network. Shift towards the domain stage: Evolving initial standards Shift towards the domain stage: Evolving initial standards. In this third transition phase, formalization of the collaboration takes place, the initial standards will be used in the daily work routine and its advantages and disadvantages will be encountered in practice. The informal consultation between the organizations in the network will evolve into formal consultations such as described by Grijpink (2009). At the same the initial standards will evolve and thereby create a better suit with the network. Where the initial standards were still quite generic in nature, they will evolve by incorporating domain specific attributes ad thereby enabling the shift to the next stage. Domain stage This stage is characterized by the addition of domain specific attributes to the initial standards that were present in the coordinated stage. Where initial standards and agreements provided the basic means for interorganizational collaboration throughout the network, in the domain stage these will have evolved to better suit the context and nature of the collaboration. In the fourth stage domain specific attributes are added to the initial standards in the network. The core technical interoperability in the network is expanded with supportive technical interoperability, such as described in Archmann (2007). Initial standards and services are often provided by standard industry solutions, however in the domain stage, network specific solutions are often designed from scratch. The use of open standards and open source software is in this stage an important success factor for interoperability (Archmann, 2007). Information lifecycle management established in the coordinated stage has led to evolved protocols and concepts. Concepts are standardized and a domain ontology is agreed upon. Next to the ontologies that exist on low level, there now also exists a domain ontology that covers all relevant domain specific concepts that are used in the network. This means that there are no longer different descriptions used for the same concept and there is a uniform structure for information storage. All information that is required in the work processes is available for all partners that need it and can be further processed in their own systems without semantic problems. Processes throughout the network are documented and aligned to not only create an overview of the processes throughout the network, but to make a seamless connection between them as well. Each organization in the network is aware of both its own processes and the other main processes in the network. The establishment of a repository for business processes has led to a reuse of best practices throughout the network. Business process management principles can be applied to redesign processes to create a more efficient workflow through the network. Business processes that support domain specific processes are developed in this stage. Shift towards the unified stage: Maintaining standards Shift towards the unified stage: Maintaining standards and flexibility. A network of organizations will likely spend a considerable time on just developing and adjusting standards due to stakeholders having each their own interests and demands. The development and adjustment will therefore never be completely finished. However when these standards reach a point where all parties in the network agree on the current functionality and the standards have the flexibility to adapt to changes, the network will reach a stage in which all dimensions are fully developed. The level of collaboration is at a sufficient level to develop a shared information system if necessary. Unified stage This stage is characterized by a network of government organizations that collaborate with each other in a highly efficient manner and thereby providing a high level of quality. Agreements and standards 45

46 that were created in the coordinated stage and were further evolved in the domain stage are now fully adapted to the context of the network. Not only are these standards fully adapted to the context, they furthermore provide sufficient flexibility to be able to adapt to changes in the environment. This creates a network of government organizations that can be perceived as a unified organization. The individual information systems are replaced or adjusted to enable information exchange throughout the whole network. There is a shared infrastructure in the network and protocols and syntax for data exchange are fully standardized, adapted to the context of the network while remaining a high amount of flexibility. Data exchange can thereby take place between all the partners in the network without having to agree on standards. Information can be freely distributed between all partners in the network. Organizations know which information is available and where it can be found. Duplicate storage of information and information sources that are out of date are thereby prevented. All concepts and attributes are documented in a high level ontology that is adapted to the network. This ensures that there is no ambiguity in the meaning of concept and enables information to be shared without loss of meaning. Processes within the network are documented and transparent for all organizations. Processes between organizations are aligned with the help of business process management solutions and can adapt to changes from the environment. Processes are dynamically linked to each other, this creates a network of organizations that is flexible and can rapidly respond to changes in the environment when needed. 3.4 Context & use of the model Type of government networks The new model focuses on networks of government organizations that share a common goal. This common goal mostly consists of providing a certain service to the public, and each of the organizations delivers a part of the solution, similar to the Public Service Networks described by Janssen (2010). The nature of the collaboration in these networks is comparable to the environment that is described in the interorganizational collaboration section of the systematic literature review. Although different organizations have to collaborate to jointly provide a service and partly depend on each other, they all maintain a large degree of autonomy. Conflicting interests play an important role between different organizations that have to collaborate and this process is further complicated by conflicting political interests. This creates a situation in which negotiation over ownership and control of information decides if and how information will be shared, the so called information politicking (Bekkers, 2005). This has a large influence on the shift between the different stages in the model. Agreeing on standards is an important process to enable a shift towards a higher level of interoperability in the network. Especially in the first stages it is important that enough support is created from the different organizations, starting with creating awareness of their common goal, followed by the necessity to transform the network into a coordinated entity. So although the transition from one stage to another in the model can seem fairly easy to make, in reality it can take several years to accomplish this. There will first have to be a sufficient level of joint consultation before any successful progress can be made in any of the dimensions. The model can be applied to both describe and prescribe government networks. The current level of interoperability maturity in a network can be determined based on the stage descriptions and capabilities. Based on the current maturity level, a higher level of collaboration in the network can be reached by prescribing a roadmap of the processes that lead to a shift between stages. Size of networks These networks operate on different levels, both on national, regional and local level. Since such networks on a national level tend to be very large and are already covered in literature as discussed in chapter 2, this model focuses on the regional and local level. Regional and local levels are defined 46

47 based on the Dutch governmental bodies, where regional collaboration corresponds with a network of collaborating organizations typically found in a province and local collaboration corresponds with collaboration found in municipalities. In some cases such networks operate on both regional and local level. The consequence of focusing on regional and local government networks lies mainly in the specificity of the model. It enables the model to describe the different stages of collaboration and the processes that lead to higher stages more precisely. This is due to the smaller scale at which the collaboration takes place. Although the legal and political dimensions are not incorporated into the model, they do have influence on how networks move through the model. Some forms of interorganizational collaboration are enforced by legislation and political motives, which differentiate the collaboration in the public sector from that in the private sector. Whereas Grijpink (2009) for example is wary of such top-down interventions, Janssen on the other hand finds that such top-down interventions can stimulate development of interorganizational collaboration and integrated service delivery. The exact conditions under which networks reach a higher level of maturity will be further discussed based on the results of the case study. Developing capabilities Creating initial standards and agreements and evolving these into context specific agreements and solutions plays an important role throughout the model. These standards are seen as important capabilities that are needed to propel change in the network maturity level. For each of the three dimensions in the network, three main capabilities can be developed, as can be seen in Table 5. These capabilities are distilled from the EIF descriptions (European Commision, 2010) and evolve throughout the stages and can be assessed on their current maturity level by mapping them to the capability descriptions in Table 5. This leads to a maturity level for each of the dimensions. The maturity level of a dimension is restricted to the lowest developed capability, a certain maturity level is only reached when all three capabilities in the dimension have reached that specific level. The overall maturity level is restricted to the lowest level of the individual dimensions. A level of overall maturity is therefore only reached when all dimensions reside on that level. The capabilities that are discussed in the model always apply on a network level. They are agreed upon by all organizations in the network and can be used by all of them. It is possible that there are some standards agreed upon between some of the organizations in the network in the early stages of the model, but this should not be seen as a development on the network level. When assessing the current level of maturity, this should be taken into account. Caution is advised when assessing networks, since the level of maturity can be easily overrated. This especially applies when the assessment is used to plan the further development of collaboration in the network, since overambitious plans have a greater chance of failing. 47

48 Table 5: Maturity model for interorganizational collaboration in government networks with capabilities I. System dimension 1. Independent 2. Ad hoc 3. Coordinated 4. Domain 5. Unified A. Data structure No data structure standards standards defined. B. File type and document formats C. File and message transfer protocols & services File types and document formats differ per organization. No file and message transfer protocols & services present. Common data structures used for incidental data exchange. Ad hoc conversion between file types and document formats necessary. File and message transfer protocols agreed upon on-thefly, no services available. Agreements on initial data structure standards in the network. Initial agreements on standard file type and document formats in network. Initial agreements on standards for file and message transfer protocols & services. Creation of additional domain specific data structure standards in the network. Creation of additional domain specific file types and document formats. Domain specific protocols and (web) services added. Data structure standards are fully developed and used throughout the network. File types and document formats are fully standardized. All file and message protocols and services are fully developed. II. Information dimension 1. Independent 2. Ad hoc 3. Coordinated 4. Domain 5. Unified A. Protocol and standards for No protocol for information exchange information exchange. B. Managing concepts (ontology) C. Managing information sources No listing of concepts available. No listing of information sources available. Ad hoc agreements on how information should be structured for exchange. Merely concepts used in ad hoc collaboration are listed. Information sources used in ad hoc collaboration are known. Initial protocol and standards for information exchange throughout the network. Low level ontology of the main concepts in the network. Standard list of all information sources throughout the network available. Evolved protocols and concepts for information exchange. Domain ontology of all relevant concepts in the network. Alignment of information sources preventing duplicates and ambiguity. Protocol and standards for information exchange are known and used throughout whole network. Complete ontology, overview and understanding of all concepts in network. All information sources are connected and available in the network. III. Process dimension 1. Independent 2. Ad hoc 3. Coordinated 4. Domain 5. Unified A. Documentation of (shared) business processes No documentation of (shared) business processes. Ad hoc sharing of information on business processes, no documentation Initial documentation of primary processes in the network. Expanded documentation of processes and adaption to Processes in network are fully documented and transparent for all organizations in the 48

49 B. (Re)design of business processes C. Repository of business processes and best practices No (re)design of business processes. No repository of business processes. available. No (re)design of business processes. Ad hoc exchange of knowledge on processes. Design initial network specific processes. Initial repository of business processes and best practices. redesigned processes. Redesign of processes to create better alignment with existing processes. Reuse of best practices in business process repository. network. Processes are fully aligned and have the flexibility to adapt to changes in network. Complete repository of business processes in the network. 49

50 4 Case study 4.1 Objectives Dul and Hak (2008) state that the majority of the authors on case study research considers it as a useful research strategy when the topic is broad and highly complex, when there is little theory available or when context is of high importance. When a case study is used as a research strategy it can contribute to the knowledge of individual, group, organizational, social, political and related phenomena (Yin, 2003). In the case of studying the development of interorganizational collaboration in government networks that apply e- government, the topic is broad and highly complex due to the large amount of parties involved and the issues that have to be addressed. The context is furthermore of high importance to be able to determine the factors that influence the development. There are several objectives in performing the case study in this thesis. The primary objective is to validate the conceptual model by applying it to actual networks of interorganizational collaboration. This is done by analyzing these networks on their current level of collaboration, and by looking at the path they took to get to this stage, concluded by studying their future ambitions. The outcome of the case study will be mapped to the stages that are described in the model and can be used to improve the model where necessary. This will be done by applying the model on different cases in the Dutch public sector. The secondary objective is to provide an insight in the current state of interorganizational collaboration maturity in Dutch government networks. By applying the model to several government networks, the current status of their development can be showed including the barriers they encounter and the opportunities they see for future development. 4.2 Case study design Yin (2003) makes a distinction between single-case designs and multiple-case designs. He states that although all designs can lead to successful case studies, multiple-case can be preferred over single-case designs. Analytic conclusions that arise independently from several cases are more powerful than those coming from a single case. The context from several cases will furthermore likely differ from each other. When common conclusions can still be drawn under these varied circumstances the external generalizability will be greater than when drawing these conclusions from a single case (Yin, 2003). The case study in this thesis will therefore also be performed with a multiple-case design. A data collection protocol will be created and three cases will be selected to perform the case study on. For each of these cases, an individual case study report will be written. Based on these individual reports, several cross case conclusions will be drawn. The theory will be modified based on these conclusions where needed and the results will be described in the final section of this chapter. Figure 13 shows the complete process of the case study. The following sections of this chapter will further describe the scenario of the case study, the data collection protocol, the selection of the cases and the analysis and conclusions that were drawn from the case study. 50

51 Figure 13: Case study design (based on Yin, 2003) Case scenario Selecting the scenario Before actual cases can be selected, the scenario of the case study must be defined first. A suitable scenario for the case study has to contain several elements. This thesis describes interorganizational collaboration in government networks that apply e-government. The cases should therefore contain networks that operate within the public sector and where several organizations collaborate to achieve a common goal. It is furthermore important that these cases also contain an e-government element. With these requirements in mind, several interviews were held with employees of PwC to gain an overview of the current developments in the public sector. The implementation of a new permit application procedure at municipalities with the use of e- government was mentioned several times. This case furthermore contains several organizations and departments that collaborate in a network. This case scenario was therefore found suitable to use in the case study and will be further described in the following section. Environmental permit The ministry of public housing, spatial planning and environmental management ( Ministerie van Volkshuisvesting, Ruimtelijke Ordening en Milieubeheer or VROM in Dutch) decided that as part of modernization, several permits should be combined into one permit, called the All-in-one Permit for Physical Aspects ( omgevingsvergunning in Dutch). By doing so, a single request procedure can be started for up to 25 permits, ensuring that the applicant only has to address one authority. This enables the government to present itself as a unified organization. By combining several permits, the administrative burden is expected to decrease as well as the lead time of the assignment procedure. The costs savings for citizens and businesses as a result of this new procedure are estimated at 130 million euro (Rijksoverheid, 2010). To enforce effective processing of these new unified permit requests a law was introduced that specifies demands to which municipalities must comply. This is called the Wabo law ( Wet algemene bepaling omgevingsrecht in Dutch), which among others specifies the deadline for processing a permit request. The permits that are combined into the All-in-one permit cover matters such as construction, demolition, spatial planning, listed buildings and the environment. The more simple permits such as a permit for cutting a tree have a relatively standard procedure and do not require the advice of many external advisors. For other projects however, several permits are necessary, requiring the advice of several organizations next to the internal municipality departments. These projects make the procedure considerably more complex and deliver more time pressure to the competent authority. During the procedure, the competent authority checks if the applicant is entitled to receive the requested permit. This authority can consist of different governmental bodies, often a municipality and sometimes a 51

52 province. Depending on the kind of permit, the authority can decide to consult (external) advisors, before deciding whether or not to assign the permit. This new permit application procedure has several consequences for the collaboration between government organizations. By combining the application of several permits into one large permit application, different organizations in the back-office have to collaborate. After the permit is applied for, the application is first globally checked on feasibility and completeness by the competent authority. After this check the actual decision making process starts. During this process several parties deliver advice to the competent authority, which uses this advice to decide on whether or not to assign the permit to the applicant. The more complex the application, the more organizations will be involved in the decision making process. The different organizations have to collaborate to adjust their advice to each other. These can be other governmental bodies, such as municipalities with their various departments, provinces and district water boards, but also external advisors such as welfare committees, fire departments and environmental advisors. The front office of the permit application is being digitalized with this new procedure, by using a web service that indicates if a permit application is needed in a specific situation or not. This web application is called the All-in-one-permit Online ( Omgevingsloket online in Dutch). The applicant can then fill in the application for the permit right away through this web service. The process then continues in the back office, where several organizations collaborate to reach a decision. Unit of analysis In the research approach, the unit of analysis was briefly described. The main unit of analysis is the chain of government organizations and in order to analyze the chain of government organizations, individual organizations that are part of the chain have to be analyzed as well. In the case scenario that will be applied in this case study, the network of organizations that will be analyzed are the organizations that are involved in the procedure of assigning an environmental permit. These include the competent authority, other government agencies and several external advisors. In this scenario municipalities are most often indicated as the competent authority and their different departments furthermore play an important role in the decision making. Because of their central role in the network, municipalities will also be used as the central point for the data collection. Site selection It was decided earlier that the case study would be performed with a multiple-case design. Three cases will be selected to compare the earlier described scenario between different municipalities. The ministry of Infrastructure and environment ( Ministerie van infrastructuur & milieu in Dutch) is concerned with the implementation of the new environmental permit at the municipalities. A preliminary interview at the ministry of infrastructure and environment was therefore conducted to identify several interesting sites on which to perform the case study. Based on the interviews it could be concluded that while both provinces and municipalities can be the competent authority in assigning permits, municipalities were most interesting to look at. They process more permit requests and on more diverse issues than the province. During this preliminary interview it was asked which municipalities could be interesting to include in a case study. The following were suggested for further analysis: Alkmaar, Arnhem, Ede, Enschede, Groningen, Haarlem, Helmond, Leidschendam and Purmerend. These were suggested because of their prominent status with regard to their adaption to the new all-in-onpermit application. It was also mentioned that the majority of the municipalities in the Netherlands has a small population and therefore processes only a small amount of permits. The level of digitization in the processes and the interorganizational collaboration might possibly be too little to be of use. It was therefore decided to only focus on the mid to large sized municipalities in the site selection. Next to the suggestions from the interview, an online search on the implementation of the all-in-one-permit at municipalities was performed. The criteria for selecting additional candidates were the size of the municipality, which was required to be mid to large sized in population. Another criterion was the proactive approach that was taken with regard to the implementation of the new all-in-one-permit. Several municipalities had described the IT initiatives to support the interorganizational collaboration in online reports. These municipalities could therefore be of worth to analyze in the case study. This led to the selection of the municipalities of Haarlem, Ede and Leiden. Haarlem and Ede were both suggested by the ministry during the preliminary interview and 52

53 are of comparable size with regard to their population. To be able to compare the differences of the interorganizational collaboration between municipalities that were equal in size, a third municipality of comparable size was added. This was the municipality of Leiden, which was found during the online search and had described some of their ICT initiatives to support the all-in-one-permit. Data collection methods Multiple data collection methods will be employed in this case study. The main data collection method consists of conducting interviews with the involved organizations based on the site and scenario selection. Other methods rely on the analysis of available documentation, which can range from written material that describes the collaboration between different organizations to process descriptions and technical architecture. Acquiring these documents was partially done through the interviews, but a lot of documentation was already publicly available through the internet. The interviews were divided into preliminary interviews and depth interviews. The preliminary interviews were conducted to create an understanding of the case in general and to explore possible sites and scenario s to further investigate. The preliminary interviews were mainly held at the ministry of infrastructure and environment. The in depth interviews were the actual interviews with subjects at the site. To conduct these interviews a protocol was created for both the preliminary and in depth interviews. The protocol that was used during these interviews can be found in the appendix. With permission of the interviewees the interviews were recorded with a digital voice recorder and these recordings were summarized afterwards. A list of the interviewees can also be found in the appendix. Other data collection methods that are mentioned in Benbasat et al. (1987), such as direct observation and physical artifacts will not be employed in this case study. They do not play an important role in validating the model or answering other research questions. The following section will describe the data analysis and results of the case study. 4.3 Data analysis In this section the results of the three cases will be discussed. The structure is based on the dimensions in the model. The dimensions will be discussed for each case separately. Cross-case conclusions will be drawn followed by analyzing and discussing the current level of maturity in each of the cases. Case 1: Municipality of Haarlem Information systems dimension The municipality of Haarlem uses a back-office application called COSA. This is a system that is jointly developed with the municipality of Haarlemmermeer to support the permit application procedure. Through the front office application for permit applications, called the Omgevingsloket online (OLO) in Dutch, citizens and businesses can apply for various permits. They hereby supply the competent authority with the information that is required to process the permit. The information that the applicant provides in OLO is further processed in the backoffice system COSA. The majority of the modules that are used to support the most common permit applications are integrated into this system, however there are still some modules missing. These missing modules are increasingly being added based on the new Wabo law. At this moment the feature of automatically transferring all the data from OLO to the own backoffice system is missing. The links to enable the data exchange between the systems are not available yet. This causes the inability to automatically transfer all data to COSA and forces them to do this manually at the moment. For the same reason, information that is taken from OLO and processed in COSA cannot automatically be updated in OLO. This prevents the applicant from checking the current status of his permit application through OLO. Next to the information exchange between the backoffice system COSA and the OLO there are also other parties who deliver information to the municipality. An example of one of these parties is the local fire department, who advices the municipality in certain cases. The local fire department however, prefers to deliver its information through the OLO instead of directly to the COSA system because of an incompatibility between their systems. 53

54 From a technological point of view one can notice that there is no agreement on the central point of information exchange. The ministry of VROM developed the OLO to create a central point where customer and municipality could meet to handle permit requests. The majority of the municipalities however, uses their own back office systems to process the permit requests. In case of the municipality of Haarlem, the OLO is only used in the first phase of the permit application procedure, but other partners like the local fire department would also like to use it later on in the procedure, to exchange information. The OLO however lacks the support for the processes that take place in the backoffice, resulting in the development of custom backoffice applications that do support these processes. The municipality of Haarlem is one of these municipalities that only use the OLO for retrieving information from permit requests. Before the all-in-one-permit application procedure the municipality of Haarlem used a different system to support the processes in the backoffice of their permit application procedure. This system was called Geobars and dated from the year This system knew little process steering and was mainly useful for tailor work on permits. It supported the procedure of a permit application primarily by being an administrative storage system, basically consisting of a database for information related to a certain permit request. This forerunner of the current system was limited in its functions compared to the current system but did support the organization in their basic needs. One of the shortcomings in the current system however is the inability to share information on the current status of a permit request with the customer. Because there is no technical link with the OLO system, the customer needs to make a call to the municipality to check on the current status. However limited the old system was, there was indeed a web based application available for customers to check their status directly online. The new system therefore did not only receive additional functionalities following the development towards the all-in-one-permits, but has had to compromise on certain functionalities as well. The municipality of Haarlem attempts to fully work digitally in processing permits, there are however still some interruptions in the digitalized work process. Sending out the advice request to various departments and external partners is done manually by mail at the moment. The system cannot automatically send out these s to the required advisor. Whenever an advisor finishes his part of the permit advice and needs to send this back to be processed in the system, this is done manually as well. Furthermore there is currently an interruption in the digital chain due to the inability to provide digital signatures. This requires that documents that need to be signed, must first be printed, then signed and finally made digital again to further process in the system. The main objective of the municipality of Haarlem is to finish the permit application procedures in time with relatively little interruptions in the digital process. When this process is running smoothly, the municipality will aim at additional fine tuning of the process. Information The backoffice system of the municipality provides standardized information exchange between the different parties. The various forms and advices generated in the system are standardized. This ensures that every advisor can process this information accordingly and prevents ambiguity. There is furthermore a standardized way of storing the information in the backoffice system, so that every unit within the organization can easily access this source of information. There are however some limitations in the information exchange that follow from the lack of links between different systems within the municipality and those of external partners. Information is not always up to date by the absence of a link with the OLO. In case of construction permits, applicants have to deliver blueprints of the construction site through OLO. The municipality then manually transfers these blueprints to the COSA system and starts processing the permit. In the meantime however it is possible that the applicant has made some changes to the blueprints and updates them in OLO. By the lack of a link with the OLO, the backoffice system COSA is never signaled of this change. This results in old information being used in the application procedure without notice. This also implies that information can t flow in the opposite direction and that the applicant can therefore not be updated on the current status of the permit through OLO. Only when the applicant contacts the municipality himself, this information can be exchanged. Within the municipality there are specific standard for storing documents, the format that is currently used is the pdf-format. The municipality should however also take the regional archival service (Noord Hollands Archief in Dutch) into account. The municipality has to comply with the public records act (archiefwet in Dutch) and has to comply with certain standards for information storage. The application that they currently use for archiving the documents created in the backoffice system may not comply with these standards. They already use a different application that does comply with these standards, but this application is not connected to COSA yet. 54

55 Next to the department that handles all permits falling under the all-in-one-permit category, there are several other departments who handle other permits in the public domain such as permits for festivals, markets and parking. They all use their own systems and information between these different systems cannot be exchanged. This can lead to different permits that conflict with each other. An example of such a conflict is that a permit is granted for an event on a certain square in the city, while at the same day construction activities are scheduled at that square to replace power lines. A good overview of all activities in a certain area lacks at the moment because this pool of information is not managed at the moment. Although the information need is available in different places in the organization there is no way of connecting it. The municipality of Haarlem is aware of this shortcoming and has formed a project group to focus on it. This group reviews all information systems used throughout the municipality and will formulate ideas on how to align them better. Processes The permit application procedure starts at the service unit of the municipality or through a request in OLO. For complex applications an account manager is appointed to the application. He will have an initial conversation with the applicant and tries to identify the exact plans. To increase the chances of a successful permit application, a quick scan is made in the case of a larger project. In a short amount of time a group of advisors within different units of the municipality will be consulted to determine the chances of success. Especially in the case of larger applications one can benefit from this. Once a permit application procedure is started, there are namely only two option left, either granting the permit or rejecting the permit request. The process subsequently continues in the backoffice, where the procedure that leads to granting or rejecting a permit is directed by the backoffice system COSA. Based on the information available in OLO a request for advice is send out to different parties within and outside the municipality. This does not happen automatically from COSA, instead a senior assessor sends these requests out by mail. The various advisors work these requests out and then return their request by mail. These advices are then manually entered into COSA again. When a certain procedure is started in COSA, this procedure cannot be undone and has to be completed first. There is also no possibility to return to previous procedures that are already completed, this creates a strict control of the system over the different processes in the whole application procedure. In the predecessor of the current system there was a less strict control on the processes. This system was less sequentially and thereby more flexible for the user. An advantage over the current system was that there was room for customization in the process. The new system on the other hand is described as being more rigid. After all advice requests are received and processed in COSA, it is clear whether or not a permit can be granted. The permit is than printed, checked by a senior assessor, signed and finally sent to the administrational unit to be scanned. The applicant then receives the notice and his permit either by or on a portable storage device (in case of large files). Relation between dimensions The municipality of Haarlem uses an integrated way of working in the permit application procedure. Their way of working is furthermore focused on complete digitization in their work processes, which has already resulted in a majority of these processes being digitalized. Several steps in the work process are not fully digitalized yet. As a result, the processes that are developed from a viewpoint of complete digitalization, cannot yet be run fully efficient. The municipality of Haarlem already started focusing on developing these processes from a viewpoint of complete digitalization before the arrival of the Wabo law, but the technical implementation still lies behind on their vision of complete digitalization. The decision to use an integrated way of working on permit application is said to be of more importance than the arrival of new technology, such as the COSA system. The focus has been mainly on developing and streamlining the processes. These processes are only partly adjusted to the technology because they were already developed from a digital point of view. ICT is mainly used in a supportive role, but can be seen in a steering role as well by signaling the deadlines for different permit requests and thereby encouraging to work more efficiently. The municipality of Haarlem uses a mentality of learning by doing, enabling them to already perform a large part of the permit application procedure by use of ICT and increasingly expand this. 55

56 Case 2: Municipality of Ede Information systems The municipality of Ede uses a backoffice system to support the processes in the permit application procedure. This system is called Wave and the municipality of Ede is the principal client and largest user of the package. The Wave system is an open source package that is largely developed based on the permit application procedure that is applied in the municipality of Ede. This package is therefore largely tailor made for the municipality of Ede. The system is furthermore developed based on the Wabo law, which implies that various modules for construction, environment and cutting permits are integrated in the software package. The advantages of developing a system for all-in-one-permits from scratch contrary to buying a license on an existing software package are next to cost savings also the influence that one has on the functionalities of the system and the alignment with existing work processes in the organization. There are furthermore several smaller municipalities that use the package, who can be supported during the implementation phase of the system, leading to new collaborating networks. The Wave system is connected to the online front office OLO, where the function of OLO can be seen as an intermediary between the applicant and the municipality backoffice system Wave. The further processing of a permit application takes place in Wave. This has both advantages and disadvantages, a disadvantage of this construction is formed by the inability to automatically process status information back to the applicant. This is not possible at the moment since there is no technical link from the Wave system back to the OLO. All files that the municipality of Ede needs to process the permit are located on the OLO. These files are automatically transferred to an FTP server and can be further transferred to a local network drive. From Wave there is a link to the external files (using hyperlinks) needed to process the permit. With every permit application, a standard folder structure is automatically created to host the different application files. External organizations such as the local fire department have their information systems linked to the Wave system and can thereby use the information they need to advise on a specific permit application. Other organizations, like the province for instance, would rather work through OLO. However, the municipality of Ede believes that for a working link with the OLO system, they are dependent on the developers of the OLO and are therefore not able to provide all partners with access to the information on their Wave system. Links to other systems within the municipality take partially place. There is still no link between the Wave system and the key register for addresses and buildings (abbreviated to BAG ). An alternative for this missing link is formed by a different package called Stroomlijn, which provides air pictures of different locations within the municipality. This package on the contrary is linked to the Wave system, enabling access to this information from Wave. Other systems that are often used are among others the document management system called Version, which is not connected to the Wave system. The consequence of this missing link is that information from other departments within the municipality is not available. Information from different departments that may affect a permit application procedure is thereby not automatically available. In the future an addition to the current situation can be formed by implementing a case system, which is available throughout the entire municipality, enabling the combination of different files in an overarching system. Cost savings are mentioned as one of the reason for not having already implemented such a system. The development towards digital working is started by the municipality of Ede in an early stage. From the perspective of interorganizational collaboration, the focus was not just on the municipality itself, but external partners such as the local fire department were also involved in this process. The municipality of Ede has invested in technological support throughout the whole chain. This has improved the collaboration between the different organizations by creating links between their systems resulting in more efficient communication. Before the current Wave system there were already several systems that supported the permit application procedure, but they did not integrate the different kinds of permits as well as in the current package. Construction and environmental permits used to have their own packages and are now integrated in Wave to form an all-in-one-permit package. The new Wabo law has therefore also played an important role in the development of a system that supports these different kinds of permits. 56

57 Information To enable efficient information processing further on in the permit application procedure, the completeness of the application is assessed first. The municipality of Ede made a conscious decision by first assessing the completeness of an application before further assessing the content of an application. Experience has taught Ede that when the permit application procedure is entered with an incomplete application, this will inevitably lead to an incomplete result. There are therefore specific demands set to the information that is provided by the applicant, to which they have to comply within two weeks. Only when an application is guaranteed to be complete it will be judged based on its content. Next to this first check on completeness, there is also a work description created that describes how the different parties and organizations should process the application information. This implicates among others that the advices that are delivered by different parties are standardized on content. Next to the standardization that applies to different units within the municipality there are also several submission requirements that apply to external parties. Since the development of the Wabo law, all addresses are registered by address, this will prevent conflicts between different permit applications on the same address. Because the Wave system contains all permit applications and makes this information available to the municipality, one can easily see if there are any conflicting permit applications. An exception to this situation can be caused when a so called gray route is chosen, in this case certain information is for example exchanged by telephone calls and this information is subsequently not processed in the system. In general these situations are not often encountered and when they do they do not cause large conflicts in the information exchange. Next to maintaining certain standards for information exchange there is also room for flexibility. In the Wave system for example, a memo field is used to provide additional commentary next to the standard forms and process steps. The municipality of Ede indicates that this is a welcome addition to the standard information exchange provides by the system since this provides a way of adding contextual information. Processes The municipality of Ede uses a process screen to keep track of the current status of a permit application. The first status of a permit application is created in the system when an application arrives in the Wave system through the OLO. Where this information used to be inputted manually into Wave, this now happens automatically. The interaction between different units and sectors within the municipality of Ede that are involved in a permit application runs through the process screen in Wave. This is the place in the system where new advices are created and where can be decided which unit should be requested to provide the advice. An is then used to send out a request for advice to the units and external organizations involved. Team leaders at those units further divide the work under advisors. There is little steering in the work processes by the back office system Wave. The larger steps are standardized in the system but the user will not be forced to finish a certain process before further steps can be taken. In contrary to integrating the entire work process into the system, the municipality of Ede uses a work description to steer the different advisors in the application process. This has led to a more supportive than steering role of the system. The municipality of Ede has invested in screening the whole work process by using a logistic advisor, several processes are streamlined and adjusted based on this screening. The work description that is set up and used is the result of the focus that the municipality has on its processes. Ede admittedly adds that this work description often changes, because the processes change quite often and should therefore be adjusted accordingly. By the supportive role of the system there is a lack of a strict workflow in the system which steers the processes. The lack of freedom in systems that have a strict steering workflow prevent de municipality of Ede of using such a system. The organization and its processes in that case have to adapt to the system instead of the other way around. The Wave system has a more open structure enabling the organization to define the processes in a way that suits them best. The municipality of Ede realizes that this way of working requires more responsibility and that one should put some effort in to it. The amount of flexibility on the contrary, increases as well and outweighs the disadvantages. Using an open system such as Wave furthermore prevents the creation of workarounds when the system prevents them of taking certain steps and thereby losing efficiency. With regard to the collaboration and alignment of processes with other partners in the chain, a mixed picture is given. On one side there is good collaboration with local partners such as the fire department, with whom there is consultation and there is alignment of processes. On the other side it is more difficult to reach this kind of 57

58 collaboration with other partners such as the province. Despite several consultations regarding the collaboration, there is not enough influence from the municipality on the processes of the province, making it more difficult to align with this partner. The increasing focus on work processes, both internally and in collaboration with other organizations is largely the result of the new Wabo law. Before the Wabo law the deadlines for finishing a permit application procedure were less tight than now. Currently such a procedure must be finished within 8 weeks. By streamlining the processes that lead to granting or denying a permit application throughout the chain and using the system in a supportive role, the municipality of Ede hopes to achieve better service provision in their permit application procedure. Case 3: Municipality of Leiden Information systems The municipality of Leiden also uses their own system to support the processes in the backoffice of the permit application procedure. This system is called BARS (an abbreviation of Bouw en Aanvraag Registratie Systeem in Dutch) This system supports and steers the users in the steps that should be taken to come to granting or rejecting a permit application. The system is built based on the processes that are applied within the application procedure of the municipality of Leiden and is therefore only used within the municipality of Leiden. The majority of the process steps have been implemented into the system and there are little steps that have to be performed outside of this system. The system has also been developed in such a way that this is not possible. The current system derived from an older system exclusively for construction permits and is expanded towards a system that supports all permits within the all-in-on-permit category. There have been made several direct links to other information systems and databases within the municipality. This implies that there is no storage of duplicate data within the BARS system, external data is selected at the source and deleted after usage. The municipality of Leiden also uses data from several key registers such as the key register for addresses and buildings (abbreviated to BAG in Dutch). The internal advisors can access these information sources from within the BARS information system. External advisors such as the local fire department, are not directly linked to the BARS information system, but instead use the OLO to exchange information with the municipality. The OLO can in this situation be seen as an intermediary between the information system of the municipality and its partners in the network. It forms a central point for information exchange. This is also the reason why there are not direct links between the systems of the different partners. The only link is the one from OLO to the BARS system, by making use of the XML standard for data exchange. The whole permit application procedure is digitalized, which implies that everything that is still provided on paper, is digitalized and entered into the system. In that case the customer is asked whether the result of the application can also be sent in a digitalized form. For the small percentage of the applicants that still want the permit application on paper, the whole process is performed digitalized, but only the decision will be printed. The municipality of Leiden already used an information system to support the permit application procedure. In about twenty to twenty-five years, this system has evolved into the current system. Before the Wabo law and the OLO, the system was mainly used to monitor the process. The current system has been expanded to not only monitor the process, but to steer it as well. In the past it were forms and drawings that went through the organization and the system only monitored the location of these objects. Now the drawing go past the different units and advisors through the system and the system ensures that this process is steered in the right direction. Information When exchanging information between different parties, the quality and structure of the information is monitored. Several agreements have been made on how information should be formulated and structured in an advice and this has resulted in several standards for information exchange. By agreeing on these standards, the information has to be adjusted as little as possible for further processing in the different units. Within the municipality the focus is mainly on assuring that these standards are complied with. These standards are mainly described in guidelines, but they are not described in a strict contract. These standards are not only agreed on and applied at the communication within the municipality, but are also agreed on with different external parties such as the local fire department. 58

59 The system furthermore delivers information on the different process steps by signaling the deadlines of the major steps. When certain advices are not delivered within the deadline, the system signals this to a supervisor who can act accordingly. These signals will be expanded even further in the future by also signaling at deadlines of smaller processes. The aim of integrating the majority of the processes into the system is to ensure that information that is processed and produced in these processes can be made available to other departments, organizations and even the public more easily. Processes The procedure of a permit application starts with a request from the customer, this can be done both through a front office and through OLO. In case of a permit request on paper, the administrative unit enters this request into the system. What follows is a check on the completeness of the request and the applicant is asked to provide additional information where needed. A coordinator then checks which advice is needed and who has to deliver these advices before contacting the corresponding internal and external parties. When every advice is collected and processed, the coordinator will finally finish and sign the outcome of the permit request. When collaborating with external parties, there is regularly contact with the local fire department and in case of environmental issues the regional environmental department is involved. The civil service and the province are involved in the permit application procedure to a lesser extent, the province for example is only consulted once or twice a year. For every advice that is required in a permit request, a deadline is set. Every step in the process is bound to these deadlines and the system monitors these deadlines. When a certain advice is submitted after the deadline, the system signals the respective supervisor. Since the introduction of the Wabo law, the deadlines for permit application procedures are shortened and this influences the deadlines for the internal processes within the municipality as well. Strictly defining the process steps and the available time in the system however, has as advantage that the procedure becomes less dependent on individuals. Should there be a loss of a certain link in the chain, it will be easier to cope with this loss and replace the missing link because the processes are clearly defined in the system. The relatively strict steering on processes can sometimes be experienced as oppressive, but on the other hand a system with a clear structure at the same time ensures calm when the workload is high. Not everyone within the organization has a complete overview of all the processes in the permit application procedure. Since these processes are integrated into the system, the different departments and advisors are mostly aware of the part of which they are responsible. The administrative organization (AO), existing of a schematic description of the work processes in an organization are not currently documented. The municipality of Leiden indicates that this should actually be done, but the focus has been predominantly been on optimizing the processes in the system. The view on the processes over the whole chain, including the external partners, is limited at the moment. There is coordination with external organizations but there is relatively little focus on the improvement of the collaboration. There is a covenant with the external organizations with regard to the quality of the information they supply to each other en within which terms they do this. How external organizations gather this information is seen as an internal affair and no attention is paid to addressing this issue. Priority is that information should be delivered according to the quality norms agreed and within the agreed time frame. The external organizations should arrange the information exchange through the OLO itself and no collaboration takes place to improve this process. This applies to all external organizations dealing with the municipality, with the local fire department and environmental department as most common organizations. This view on the collaboration has been mainly caused by a change of mentality, which in its turn has been caused by the introduction of the Wabo law. This has led to an integrated working mentality in which shorter delivery terms apply. To realize this way of working, clear agreements have to be made between the involved internal and external parties and more focus is put on responsibilities. This new mentality mainly applies to the internal departments, but is also reflected in the collaboration with external organizations. Although a public service is provided, there are still clear quality and time demands connected to the delivery of that service. This furthermore implicates that certain shortcomings and complications in a permit request are already pointed out to the permit applicant in an early consultation session. By somewhat steering the applicant in its permit 59

60 request, problems that might otherwise arise later on in the permit application procedure can be solved in an early stage. This saves time further on in the procedure and ensures the timely handling of the request. With regard to working digitally, the introduction of the Wabo law is indicated as a turning point. A good working system is seen as sixty to seventy percent of the success. The largest amount of the information needed in the procedure is automatically delivered to the user. This creates an increased amount of available information and provides a better view on all the processes. By thinking about the development of the system, one is also forced to think about the processes that the system has influences. Working digitally also means that many agreements have to be reached between various parties and the side effect is that the processes become clearer as well. Although the system steers the processes heavily, the system is increasingly more seen as supportive than steering. In the beginning the system was experienced as very tight, where people felt forced to follow certain steps. People are however so used to the system that it is increasingly more experienced as a supportive tool. 4.4 Cross-case conclusions Information systems The setup of information systems differs per municipality. The first setup is characterized by developing an information system in collaboration with other municipalities. In these cases new networks of collaborating municipalities arise to jointly develop a system that supports their needs in the permit application procedure. Other municipalities have systems that are specifically developed and tailored to their processes. A final category of back office systems can be distinguished, in which a standardized package is used. Several municipalities can obtain a license that gains them the permission to use this standardized information system. In view of the new all-in-one-permit, the information systems at the municipalities had to be redeveloped. In all cases, the municipalities have developed their own back office system to support the permit application procedure. Existing systems were depreciated or radically redeveloped to fit this new situation. Several context factors played an important role in this development. The new all-in-on-permit and the law that was introduced to enforce correct use, forced the municipalities to not only change their work processes, but also the systems that support these processes. These factors did not only influence the municipalities, but also their collaborating partners, which had to redevelop their systems as well in order to keep the ability to exchange data. The municipality takes a prominent and central role in the network and sends out a lot of information requests to other organizations. When the information system which is used by the municipality has sufficient possibilities for external partners to connect with, the level of technical connectedness within the whole network is higher than when one of the other partners is technically more developed. Although the municipality does not have enough power over the other organizations to steer the technical developments in the whole chain, it is still the most influential partner in the network. The OLO system is in this regard used in different ways by the three municipalities. Two of the municipalities and their partners use the OLO only in the role of an intermediary between the applicant and the municipality. Further processing and storage of data is done in the backoffice. Other organizations in the network exchange information with the back office system of the municipality, instead of with the OLO system. This situation is depicted in Figure 14. In the figure there is a distinction made between information exchanges that takes place manually, for example by , portable storage devices or even by mail and automatic information exchange between information systems. 60

61 Figure 14: Setup one Other municipalities however, use the OLO as the central point of information exchange. This is mainly due to the inability of external partners to make a direct connection to the system of the municipality. In this case every organization only links to the OLO and stored or requests its information there. This difference in set up of information systems is illustrated in Figure 15. Figure 15: Setup two 61

62 Information In the information exchange between the different collaborating parties, the focus is mainly on the completeness of the information. Especially at the start of the chain, the standards that are set for the completeness of application forms are very high. When organizations further down the chain have to work with incomplete permit applications, this will result in an incomplete permit decision as well. This focus on completeness is furthermore driven by the fact that once an application procedure is started, it should be finished within a predefined period of time. Completeness in the information exchange results in more efficient collaboration and enables the organizations to meet their deadlines. All the municipalities have standards for processing the application information and processing these in different forms. Standard document structures are agreed on and are encouraged to be used by both internal and external parties. In most cases however these standards do not cover the entire network. Next to the standardization on the structure of the documentation exchanged, there was little standardization on semantics. There was no common vocabulary defined between the different organizations in the networks. An explanation for the lack of a common vocabulary can be given by the context of the cases. The information that is exchanged mostly takes the form of reports being exchanged. These reports are standardized on predefined chapters and sections, but the actual content of these reports are not further standardized. Processes Standardization of the business processes was encountered in every case. In the alignment of different business processes, the municipalities apply an internal perspective. The business processes were redesigned to enable an efficient permit application procedure. Two different perspectives on how these business processes could be managed most effectively were identified. The first perspective heavily relies on flexibility in the business processes so that these can be steered where necessary. The second perspective that was identified relies on the stability of the business processes and maximal support of these processes from the back office system. Aside from the differences in perspective on how to manage the business processes and how these are actually supported by their systems, in all these cases a similar workflow could be identified. The permit requests are assembled at a central point and are checked on completeness by the municipality. After this check on municipality, the application is further processed on content. The municipality then identifies the need for several advices from internal and external partners. The requests for these advices are then sent out to the relevant advisors. These advisors further process these advices and communication may take place between the advisors, the municipality and the applicant. This communication will mainly address missing or additional information that is needed to finish the advice. All advices are eventually sent back to the municipality, who processed them and decides whether or not the permits can be granted. There was a clear overview of processes within this procedure which took place within the different departments of the municipalities. These processes were often also aligned and adjusted to the information system that was used. There was however little knowledge of the processes that took place in other parts of the network. Since these processes were not clear within the municipalities, they were also not aligned with the internal processes. There was no view of all the processes over the whole network and little discussion or initiatives took place to apply an interorganizational perspective. The lack of such initiatives originates partially from a difference in attitude towards external advisors. They are no longer just seen as partners of the municipalities. They now view them more in a role of service providers, the municipality uses them to provide a certain service that adheres to specific quality standards and expects to receive these services before a certain deadline. This attitude in turn changed due to the stricter legislation which forces municipalities to finish these procedures within a certain time limit. Relations between dimensions The tree dimensions did not always seem to develop evenly. A difference in the case studies was noticeable between networks that focused more on the development of the technological dimension and networks that focused more on the process dimension. The focus on the technological dimension was expressed by implementing all possible business processes into the system and relying on the system for an effective workflow. In these cases the system heavily steers on the business processes and there is little room for organizational flexibility. Processes cannot easily be adapted but the ones that are implemented in the system 62

63 can be effectively executed. For the networks that focused on the process dimension however, the description, development and adjustment of different business processes in the procedure played a dominant role. The development in the technological dimension was present, but the network did not heavily depend on the system. By clearly describing all the work processes in the organization and aligning them where possible, not only internally but also with external parties, the application procedure could be effectively performed. The flexibility in the processes remained high, due to the absence of a strict steering role of the system. Reasons that were given for focusing on developing the collaboration further in one dimension, were flexibility and on the other hand efficiency. Flexibility was given as a reason to not focus too much on implementing all business processes into the system. Efficiency on the other hand was given as a reason why one should focus on technical maturity. Even though these approaches are quite different from each other, both approaches were perceived as a successful strategy to reach an efficient permit application procedure. Since the networks in these cases operate within the public sector, there were some additional factors influencing the interorganizational collaboration. Especially political and legal influences were noticeable in these cases. The all-in-one-permit is the result of the new Wabo law that forces municipalities to follow a permit application procedure that complies with certain quality standards and time limits. This puts additional pressure on the different organizations that are collaborating to achieve these goals. Since the municipality is the main responsible authority when these demands are not met, this pressure is also reflected in the collaboration with other organizations. It was noticeable that the municipality did no longer take on a passive role in the collaboration, but shifted its responsibility by also demanding a specific time limit and level of quality of the delivered information from other organizations in the network. Maturity levels cases Assigning the maturity level Based on the information gained in the interviews and the available documentation, the current level of interorganizational collaboration in the government networks will be determined. For each of the three dimensions in the network, three main capabilities can be developed as can be seen in Table 5. These capabilities evolve throughout the stages. For each case, these capabilities have been assessed on their current maturity level by mapping them to the capability descriptions in Table 5. This leads to a maturity level for each of the dimensions. The maturity level of a dimension is restricted to the lowest developed capability, a certain maturity level is only reached when all three capabilities in the dimension have reached that specific level. The overall maturity level is restricted to the lowest level of the individual dimensions. A level of overall maturity is therefore only reached when all dimensions reside on that specific level. This analysis has resulted in the creation of Table 6 in which for each dimension, the capabilities are assessed on their current maturity level. Table 6: Comparison maturity levels on three dimensions I. System dimension Case 1: Haarlem Case 2: Ede Case 3: Leiden A. Data structure standards Common data structures used for incidental data exchange (2). Common data structures used for incidental data exchange (2). B. File type and document formats C. File and message transfer protocols & services Ad hoc conversion between file types and document formats (2). File and message transfer by (2). No standard file types and documents used, differs per partner (2). Several organizations can automatically exchange files and messages (2). Agreements on initial data structure by using XML standard (3). Initial agreements on standard file type and document formats in network (3). Initial agreements on standards for file and message transfer through OLO (3). System maturity level Ad hoc (2) Ad hoc (2) Coordinated (3) II. Information dimension Case 1: Haarlem Case 2: Ede Case 3: Leiden A. Protocol for information Forms and advices are Forms and advices are Agreements on exchange standardized (3). standardized (3) formulation and 63

64 B. Managing concepts (ontology) Main concepts are listed through forms (3). C. Managing information sources Only the information sources of direct partners are known, no complete overview (2). Main concepts are listed through forms (3) Only the information sources of direct partners are known, no complete overview (2) structure of information (3) Main concepts are listed through forms (3) Municipality maintains a list of information sources which are available throughout the network. (3) Information maturity level Ad hoc (2) Ad hoc (2) Coordinated (3) III. Process dimension Case 1: Haarlem Case 2: Ede Case 3: Leiden A. Documentation of (shared) business processes B. (Re)design of business processes C. Repository of business processes and best practices Main processes are documented in the system, little to no alignment (3). Business processes are (re)designed and incorporated in the system (3) No repository of business processes available on network level, only internally (2). Processes are partially described, but change often, not strictly incorporated in system (2) Screening of work processes throughout the network, basic alignment (3) Experiences and best practices are shared with several other municipalities (2). Primary processes in the network are documented and incorporated into the system (3). Network specific processes are being designed and adjusted in the work flow system (3). Initial repository of business processes, best practices not yet applied (3). Process maturity level Ad hoc (2) Ad hoc (2) Coordinated (3) Overall maturity level Ad hoc (2) Ad hoc (2) Coordinated (3) Although some of the capabilities in case 1 and 2 reside on the coordinated level (IIA and IIB for example), the majority resides on the ad hoc level. The main explanation for this is that the different organizations in the network still have not reached an agreement on the central place for data exchange. Some organizations would like to supply data directly to the back office system of the municipality, where other organizations would rather use the OLO as central point for information exchange. As a result, ad hoc measures such as the use of are used to exchange information between different partners. The system dimension seems to be under developed compared to the other two dimensions and limits the networks to further develop. Their overall maturity level is therefore also limited to the ad hoc level. In the third case the system dimension is further developed. Each organization used its own information system, but information exchange is ensured by agreeing on a standard platform for information exchange. All systems use the OLO as a central point of data exchange and comply with the standards that were agreed in the OLO. This enables the other dimensions to reside on the coordinated level as well and the overall maturity level is therefore assessed on the coordinated stage. In Figure 16 the maturity levels per dimension in each of the cases are graphically represented. In case 1 for example the system dimension resides on the level of ad hoc collaboration (2). The information and organization dimension also reside on the level of ad hoc collaboration, but since two out of three capabilities in these dimensions already reside on the coordinated level, the graph shows them between level two and three. Note that this is purely to show that the information and organizational dimension are developing towards the coordinated level but are presently limited to the ad hoc level. 64

65 Case 1 Case 2 System Information Organization Case Figure 16: Maturity levels per dimension for each case Several remarks should be made when discussing the current maturity levels. There has to be made a distinction between the highest level of collaboration possible and the most effective level of collaboration. The most effective level of collaboration depends on the context of the case. Some networks can operate effectively on a level of ad hoc collaboration without having to strive for a shared information system. The advantages of achieving a stage of unified collaboration do not outweigh the costs and effort that have to be put into that achievement. In order to be able to assess the current level of interorganizational collaboration in government networks that apply e-government, the first step should be to determine which level is sufficient in that specific context. In this specific context for example, the municipality of Haarlem stated that they are satisfied with their current level of collaboration since they can handle all permit applications within the required time frame. They however acknowledge that to be able to execute their work processes more effectively, development on the system and process dimension is needed. The same remark can be made for the development of the individual dimensions, the business process dimension may be further developed than the information systems dimension. This does not automatically imply that the current level of collaboration on information systems is insufficient. In light of the maturity levels in the three cases that were discussed, there were different approaches as to which of the dimensions had to be developed predominantly. In one case the business process dimension was focused on, while in another case the information systems dimension was most important to develop. In both cases it was a deliberate decision to take either one of these approaches and both were found to be successful. 4.5 Discussion The different levels of interoperability seem to be applicable to government networks in the Dutch public sector. The three dimensions in the model form a good starting point for discussing and measuring interoperability in government networks. The most effective level of collaboration possible seems to depend on the context. Networks can effectively operate on a certain level without the need to strive for a higher stage. The first step should therefore be to determine which level of collaboration is sufficient in a specific context, the model can then be applied to close the gap between the current level and the desired level. In the case of permit applications, the network is in theory relatively big, since the large amount of different possible permit requests. However, in practice there are mostly two or three main partners in the network to collaborate with. The collaboration therefore shows many features of ad hoc collaboration. This level of collaboration, although not the most advanced, was found sufficient for daily operation. 65

66 Furthermore, the three dimensions may not always develop evenly. Different strategies seem to exist, in which the focus is predominantly on one of the dimensions. This focus might change after (several) stages. Further research could therefore focus on using the model to analyze new cases that could uncover additional growth strategies. Identifying additional growth strategies can contribute to creating several successful roadmaps through the model. Since each case is influenced by a set of environmental factors, some strategies may prove to be more successful in some areas than others. By combining concepts of interoperability and interorganizational collaboration, not only were the different stages of interoperability in government networks described, but the processes that initiate shifts between these stages as well. This process oriented approach is of great importance when applying the model in practice, since it supports government networks that wish to improve their collaboration and uses concepts that apply to a government network context. Limitations and future research The current maturity model was tested and found applicable in three cases within the public sector. The cases were however limited to the domain of the permit request procedure. As mentioned before, when applying the model, the context has to be taken into account. Other government networks within different domains were not included in this study and could be further researched. The similarities or differences between growth strategies in different domains are worth researching in particular. This research also uncovered several political and legal influences on the development of the interorganizational collaboration. To limit the scope of the research and first create a solid foundation for the research, these factors were not included in the initial model, but were rather discussed in the context of the model. They are however unmistakably of interest to this research and could be integrated in a later version of the current model. Next to the capabilities that are currently used in the model to determine the current maturity level, additional capabilities may need to be identified. The relationship between capabilities and the current maturity level could also be further researched, in order to identify capabilities that have a greater influence on overall maturity. In the current model, the most crucial capabilities in each dimension were identified, but they can be further divided into sub-divisions of capabilities in future work. The current research assessed the maturity level based on interviews, whereas a future version could take a more quantitative approach in measuring the current maturity level. 66

67 5 Conclusions The aim of this thesis was to study the development of interorganizational collaboration in government networks that apply e-government. In order to study this phenomenon several research questions were developed. The main research question was formulated as follows: How does interorganizational collaboration develop in government networks that apply e- government? Several sub research questions were developed as well to help answering the main research question. Since this interorganizational collaboration is studied in networks of government organizations that use of ICT as an important tool in their collaboration, the concept of e-government had to be clarified first. The first sub research question was dedicated to this: What is e-government and why is it relevant? To answer the first part of the question, a fitting definition was selected from literature that covered all parties involved and refrains from stating explicit technologies that change over time. The following definition was identified from literature: The relationships between governments, their customers and suppliers (businesses, other governments, and citizens) by the use of electronic means. New goals for the use of IT in governments were increasing the convenience of government service provision, facilitating administrative reform and furthering democratic participation. The use of ICT became an important factor in the transformation of governments. Organizations needed to collaborate with each other to be able to achieve these goals and e-government is an important prerequisite to enable successful interorganizational collaboration. These are the main reasons why e-government is relevant when studying interorganizational collaboration in government networks. Since this thesis focuses on describing the development of interorganizational collaboration in government networks that apply e-government, current models that describe the development of these concepts had to be identified. The second sub research question addressed this issue: Which current models exist to describe the maturity of interorganizational collaboration in government networks that apply e-government? In a systematic literature review, models that described these developments were searched and analyzed. A total of 19 models were found and further analyzed on the dimensions they used to describe the collaboration, the stages they distinguished and the preconditions that were required to shift towards higher levels. Many of the models that were identified in literature were created in the early 2000s and focused on broadly describing e-government development. The majority of these models and especially the early ones describe these developments from a customer perspective (G2C/G2B). These describe the different service levels that the government provides. Only four models were identified that focus on collaboration between government organizations (G2G) and thereby described the back office of e-government. These were predominantly created in recent years and can be seen as a reaction on the large amount of G2C/G2B models. The G2C/G2B models were largely outcome oriented, meaning that they describe the outcome or characteristics of a certain stage in e-government without describing the processes and preconditions that lead to that stage. Process oriented models are predominantly found among the G2G models. The process oriented models describe the boundaries between stages and how transitions take place. Capabilities are mentioned to describe prerequisites that are required to make a transition between different stages. The recent rise in G2G stage models showed a need to describe the development of interorganizational collaboration using ICT between government agencies. The currently available G2G stage models give an idea of the issues that arise when dealing with interorganizational collaboration and its development. 67

68 Which dimensions can be used to describe the maturity level of interorganizational collaboration in government networks that apply e-government? The maturity level of interorganizational collaboration can be viewed from different perspectives or dimensions. Several of these dimensions were identified through analyzing the existing models of growth. The majority of the models only measure the collaboration by looking at one dimension. The technical dimension was found to be applied most often in these models. This is not surprising since it is an important dimension to consider when realizing e-government. A dimension that is recognized more and more to be important as an addition to the traditional technical dimension. This dimension is part of a majority of the interoperability frameworks and describes the collaboration on the information level. Organizations realize that it is no longer a matter of using technology to connect to each other, but agreements on the structure and meaning of what is exchanged have to be made as well. In the design of the dimensions of the new model, the decision was made to follow these three main dimensions. By deciding to select the technical, semantic and organizational dimension, the conceptual model adheres to the dimensions mentioned in the European Interoperability Framework (EIF). Because of their importance to interoperability (Kubicek & Cimander, 2009) these dimensions will be incorporated into the model first. Other dimensions were considered as well such as a political, governance or legal dimension. These dimensions are not selected to incorporate into the design of the maturity model itself, since they can be better used to describe the context of the model. Which stages, their characteristics and preconditions can be distinguished to describe a maturity model of interorganizational collaboration in government networks that apply e-government? Several stages were defined that described the development of the three dimensions discussed before. These stages represent levels in interorganizational collaboration that intensify with each additional stage. The division of the stages in the available maturity models was analyzed first. This showed that the majority of the models maintain a number of four or five stages. No literature was found that states that a division of a stage model in a certain amount of stages is required and the division into five stages was found to be the most practical. By defining five stages, two extremes could be defined, balanced by the third stage. The second and fourth stage, make a gradual shift between these extremes possible. Furthermore, for statistical purposes five stages can be practical when applying a five point Likert scale to determine certain capabilities. Several requirements were set for the selection of the names for the stages. The name should represent a level of connectedness between organizations in an interorganizational network. The name should furthermore be applicable to each dimension in the model. With the use of five stages in the model, the first stage had to describe a situation where there is little to no interoperability, the final stage represents complete interoperability and the stages between them should be labeled to describe the gradual shift between the two extremes. Defining these stages led to the following labeling of the stages: independent, ad hoc, coordinated, domain and unified. For each of these stages, the characteristics in the three dimensions were described, as well as the factors that enabled a transition between the stages. Each dimension was further divided into capabilities that evolve throughout the different stages. This model was then applied to several cases of government networks. The following research question addressed how the model could be used in practice: How can this maturity model be used in describing and guiding the development in interorganizational collaboration in government networks that apply e-government? To be able describe how interorganizational collaboration in government networks that apply e-government develops a multiple-case study was performed. The first goal of the case study was to select a suitable case scenario. This scenario had to meet certain requirements. The cases had to contain networks that operate within the public sector and where several organizations collaborate to achieve a common goal. These cases furthermore had to contain an e-government element. The implementation of a new permit application procedure at municipalities with the use of e-government was found suitable to use in the case study and was used to select three cases. 68

69 While both provinces and municipalities can be the competent authority in assigning permits, municipalities were most interesting to look at. They process more permit requests and on more diverse issues than the province. It was furthermore decided to only focus on the mid to large sized municipalities in the site selection. This led to the selection of the municipalities of Haarlem, Ede and Leiden. To be able to compare the differences of the interorganizational collaboration between municipalities these municipalities were partially selected due to their equality in size. Several conclusions could be drawn from the case study. First of all the current state of the interorganizational collaboration in these government networks that apply e-government was discussed. This led to several conclusions per dimensions which will be summarized below. Information systems The setup of information systems differs per municipality. The first setup is characterized by developing an information system in collaboration with other municipalities. Other municipalities have systems that are specifically developed and tailored to their processes. A final category of back office systems can be distinguished, in which a standardized package is used. Several municipalities can obtain a license that gains them the permission to use this standardized information system. Several context factors played an important role in the redevelopment of information systems over the previous years. The new all-in-on-permit and the law that was introduced to enforce correct use, forced the municipalities to not only change their work processes, but also the systems that support these processes. These factors did not only influence the municipalities, but also their collaborating partners, which had to redevelop their systems as well in order to keep the ability to exchange data. The legal dimension played a role in this case to stimulate the information systems dimension to develop. The connection between different partners in the network is closely related to the information systems that are used by the municipality. The municipality takes a prominent and central role in the network and sends out many information requests to other organizations. When the information system that is used by the municipality has sufficient possibilities for external partners to connect with, the level of technical connectedness within the whole network is higher than when one of the other partners is technically more developed than the municipality. Although the municipality does not have enough power over the other organizations to steer the technical developments in the whole chain, it is still the most influential partner in the network. Information All the municipalities have standards for processing the application information and processing these in different forms. Standard document structures are agreed on and are encouraged to be used by both internal and external parties. In most cases however there are guidelines for these standards but no strict agreements and documentation to ensure the correct use of these standards. Next to the standardization on the structure of the documentation exchanged, there was little standardization on semantics. There was no common vocabulary defined between the different organizations in the networks. An explanation for the lack of a common vocabulary can be given from the context of the cases. The information that is exchanged mostly takes the form of reports being exchanged. These reports are standardized on predefined chapters and sections, but the actual content of these reports are not further standardized. This makes it less practical to implement an ontology of concepts or other semantics. Processes Two different perspectives on how business process could be managed most effectively were identified. The first perspective heavily relies on flexibility in the business processes so that these can be steered where necessary. The second perspective that was identified relies on the stability of the business processes and maximal support of these processes from the back office system. There was a clear overview of processes that took place within the different departments of the municipalities. These processes were often also aligned and adjusted to the information system that was used. There was however little knowledge of the processes that took place at the external advising organizations. Since these processes were not known within the municipalities, they were not aligned with the internal processes either. There was no view of all the processes over the whole network and no discussions or initiatives took place to apply an interorganizational perspective. 69

70 Relations between dimensions The tree dimensions did not always seem to develop evenly. A difference in the case studies was noticeable between networks that focused more on the development of the technological dimension and networks that focused more on the process dimension. The focus on the technological dimension was expressed by implementing all possible business processes into the system and relying on the system for an effective workflow. In these cases the system heavily steers the business processes and there is little room for organizational flexibility. Processes cannot easily be adapted but the ones that are implemented in the system can be effectively executed. For the networks that focused on the process dimension however, the description, development and adjustment of different business processes in the procedure played a dominant role. The development in the technological dimension was present, but the network did not heavily depend on the system. By clearly describing all the work processes in the organization and aligning them where possible, not only internally but also with some external parties, the application procedure could be effective be performed. The flexibility in the processes remained high, due to the absence of a strict steering role of the system. Reasons that were given for focusing on developing the collaboration further in one dimension, were flexibility and on the other hand efficiency. Flexibility was given as a reason to not focus too much on implementing all business processes into the information system. Efficiency on the other hand was given as a reason why one should focus on the information system maturity. Even though these approaches are quite different from each other, both approaches were perceived as a successful strategy to reach an efficient permit application procedure. This shows that multiple strategies can be developed through the model, depending on the context, which although differing in focus can both lead to successful growth in the network. Since the networks in these cases operate within the public sector, there were some additional factors influencing the interorganizational collaboration. Especially political and legal influences were noticeable in these cases. The all-in-one-permit is the result of the new Wabo law, which forces municipalities to follow a permit application procedure that complies with certain quality standards and time limits. This puts additional pressure on the different organizations that are collaborating to achieve these goals. Since the municipality is the main responsible authority when these demands are not met, this pressure is also reflected in the collaboration with other organizations. Maturity levels Several conclusions can be made when discussing the current maturity levels. The different levels of interoperability seem to be applicable to government networks in the Dutch public sector. The three dimensions in the model form a good starting point for discussing and measuring interoperability in government networks. Two out of the three cases resided on the ad hoc level of collaboration and the third one resided on the coordinated level. The most effective level of collaboration possible depends on the context. Networks can effectively operate on a certain level without the need to strive for a higher stage. The first step should therefore be to determine which level of collaboration is sufficient in a specific context, the model can then be applied to close the gap between the current level and the desired level. Furthermore, the three dimensions may not always develop evenly. Different strategies seem to exist, in which the focus is predominantly on one of the dimensions. In light of the maturity levels in the three cases that were discussed, there were different approaches as to which of the dimensions had to be developed predominantly. By combining concepts of interoperability and interorganizational collaboration, not only were the different stages of interoperability in government networks described, but the processes that initiate shifts between these stages as well. This process oriented approach is of great importance when applying the model in practice, since it supports government networks that wish to improve their collaboration and uses concepts that apply to a government network context. The current model can therefore be applied to measure the current maturity level in a network and plan a roadmap to develop future collaboration. 70

71 References Abrahamse, M., Wisse, P., & Oude Luttighuis, P. (2009). Semantiek op stelselschaal: issues en oplossingsrichtingen. Forum Standaardisatie, Den Haag. Abramowicz, W., Bassara, A., Wisniewski, M., & Zebrowski, P. (2008). Interoperability Governance for e- Government. In R. Kaschek, C. Kop, C. Steinberger, & G. Fliedl, Information Systems and e-business Technologies (pp ). Berlin Heidelberg: Springer. Andersen, K. V., & Henriksen, H. Z. (2006). E-government maturity models: Extension of the Layne and Lee model. Government Information Quarterly, 23(2), Archmann, S. (2007). egovernment Services: Interoperability Success Factors, Inclusions and Accessibility. European Review of Political Technologies, 4, Baum, C., & Di Maio, A. (2002). Gartner s Four Phases of E-government Model. Bekkers, V., & Homburg, V. (2002). The Back-Office of E-Government (Managing Information Domains as Political Economies). 35th Annual Hawaii International Conference on System Sciences (HICSS'02), 5, pp Big Island, Hawaii. Benbasat, I., Dexter, A., Drury, D., & Goldstein, R. (1984). A critque of the stage hypothesis: theory and empirical evidence. Communications of the ACM, 27(5), Benbasat, I., Goldstein, D. K., & Mead, M. (1987). The Case Research Strategy in Studies of Information Systems. MIS Quarterly, 11(3), Burn, J., & Robins, G. (2003). Moving towards e-governmen: a case study. Logistics Information Management, 16(1), C4ISR Architectures Working Group. (1998). Levels of Information Systems Interoperability. Chandler, S., & Emanuels, S. (2002). Transformation not Automation. In D. Remenyi, 2nd european conference on e-government. St. Catherine's College Oxford. Clark, T., & Jones, R. (1999). Organizational Interoperability Maturity Model for C2. Command & Control Research & Technology Symposium, (pp. 1-13). Newport. Coursey, D., & Norris, D. F. (2008). Models of E-Government: Are They Correct? An Empirical Assessment. Public Administration Review, 68(3), Dul, J., & Hak, T. (2008). Case Study Methodology in Business Research. Oxford: Elsevier. European Commision. (2004). European Interoperability Framework for pan-european egovernment Services v1.0. Opgehaald van European Commision. (2010). European Interoperability Framework (EIF) for European public services. Opgehaald van Gil-Garcia, J. R., & Martinez-Moyano, I. J. (2007). Understanding the evolution of e-government: The influence of systems of rules on public sector dynamics. Government Information Quarterly, 24(2), Gottschalk, P., & Solli-Sæther, H. (2008). Stages of e-government interoperability. Electronic Government, An International Journal, 5(3), Greiner, L. E. (1972). Evolution and revolution as organizations grow. Harvard Business Review, 50(4),

72 Grijpink, J. H.A.M. (2009). Kijk op Ketens. Het Ketenlandschap van Nederland. (J. H.A.M. Grijpink, & M. G.A. Plomp, Red.) Den Haag: Grijpink. Grijpink, J.H.A.M, & Plomp, M.G.A. (2011). Combating Identity Fraud in the Public Domain: Information Strategies for Healthcare and Criminal Justice. The Proceedings of the 11th European Conference on egovernment (pp ). Reading: Academic Publishing Limited. Guijarro, L. (2007). Semantic interoperability in egovernment initiatives. Computer Standards & Interfaces, 31(1), Hart, C. (1998). Doing a Literature Review: Releasing the Social Science Research Imagination. London: SAGE Publications Ltd. Heeks, R., & Bailur, S. (2006). Analyzing e-government research: Perspectives, philosophies, theories, methods, and practice. Government Information Quarterly, 24(2), Hevner, A. R., March, S. T., & Park, J. (2004). Design research in information systems research. MIS Quarterly, 28(1), Hiller, J. S., & Bélanger, F. (2001). Privacy Strategies for Electronic Government. Washington DC: IBM Center for the Business of Government. IEEE. (1990). IEEE standard computer dictionary: a compilation of IEEE standard computer glossaries. New York, NY: Inst of Elect & Electronic. Janssen, M. (2010). Governing and Integrating Public Service Networks. tgov Workshop '10 (tgov10), (pp. 1-11). West London. Kazanjian, R. K., & Drazin, R. (1989). An Empirical Test of a Stage of Growth Progression Model. Management Science, 35(12), King, W. R., & Teo, T. S. (1997). Integration between Business Planning and Information Systems Planning. Journal of Management Information Systems, 14(1), Klievink, B., & Janssen, M. (2009). Realizing joined-up government Dynamic capabilities and stage models for transformation. Government Information Quarterly, 26(2), Kraemer, K. L., & Dedrick, J. (1997). Computing and Public Organizations. Journal of Public Administration Research and Theory, 7(1), Kubicek, H., & Cimander, R. (2009). Three dimensions of organizational interoperability: Insights from recent studies for improving interoperability frame-works. European Journal of epractice, 6, Layne, K., & Lee, J. (2001). Developing fully functional E-government: A four stage model. Government Information Quarterly, 18(2), Lee, J. (2010). 10 year retrospect on stage models of e-government: A qualitative meta-synthesis. Government Information Quarterly, 27(3), Leem, C. S., Kim, B. W., & Yu, E. J. (2008). Information technology maturity stages and enterprise benchmarking: an empirical study. Industrial Management & Data Systems, 108(9), Meijer, A. (2007). Why don t they listen to us? Reasserting the role of ICT in Public Administration. Information Polity, 12(4), Ministry of Infrastructure and the Environment. (2010). All-in-one Permit for Physical Aspects in a nutshell. The Hague: Ministry of Infrastructure and the Environment. 72

73 Netchaeva, I. (2002). E-government and E-Democracy: A Comparison of Opportunities in the North and South. International Communication Gazette, 64(5), Nolan, R. L. (1973). Managing the Computer Resource A Stage Hypothesis. Communications of the ACM, 16(7), Okoli, C., & Schabram, K. (2010). A Guide to Conducting a Systematic Literature Review of Information Systems Research. Sprouts: Working Papers on Information Systems, 10(26), Orgun, B., Dras, M., & Nayak, A. (2008). Approaches for semantic interoperability between domain ontologies. Expert Systems, 25(3), Papantoniou, A., Hattab, E., Afrati, F., Kayafas, E., & Loumos, V. (2001). Change Management, a Critical Success Factor for e-government. Proceedings of the 12th International Workshop on Database and Expert Systems Applications, (pp ). Munich. Pardo, T. A., & Jiang, Y. (2007). Electronic Governance and Organizational Transformation. Proceedings of the 1st international conference on Theory and practice of electronic governance (pp ). New York: ACM. Peffers, K., Tuunanen, T., Rothenberger, M. A., & Chatterjee, S. (2008). A Design Science Research Methodology for Information Systems Research. Journal of Management Information Systems, 24(3), Peristeras, V., Tsekos, T., & Tarabanis, K. (2002). Analyzing E-government as a Paradigm Shift. UNTC Occasional Papers Series(1), Plomp, M. G.A., & Batenburg, R. S. (2010). Measuring chain digitisation maturity: an assessment of Dutch retail branches. Supply Chain Management: An International Journal, 15(3), Rijksoverheid. (2010, October 1). Wabo van start. Retrieved August 18, 2011, from Web site Rijksoverheid: Ronaghan, S. (2001). Benchmarking E-Government: A Global Perspective. New York: United Nations Division for Public Economics and Public Administration and American Society for Public Administration. Sarantis, D., Charalabidis, Y., & Psarras, J. (2008). Towards standardising interoperability levels for information systems of public administration. The Electronic Journal for e-commerce Tools & Applications (ejeta), 2(3), Schneider, D., & Means, G. (2000). Meta-capitalism: The e-business revolution and the design of 21st century companies and markets. New York: John Wiley & Sons Inc. Scholl, H. (2005). Interoperability in e-government: More than Just Smart Middleware. Proceedings of the 38th Hawaii International Conference on System Sciences, (pp. 1-10). Scholl, H. J., & Klischewski, R. (2007). E-Government Integration and Interoperability: Framing the Research Agenda. International Journal of Public Administration, 30(8), Silcock, R. (2001). What is e-government? Parliamentary Affairs(54), Soares, D., & Amaral, L. (2011). Information Systems Interoperability in Public Administration: Identifying the Major Acting Forces through a Delphi Study. Journal of Theoretical and Applied Electronic Commerce Research, 6(1), Solli-Sæther, H. (2011). A framework for analysing interoperability in electronic government. International Journal of Electronic Finance, 5(1),

74 Solli-Sæther, H., & Gottschalk, P. (2010). The Modeling Process for Stage Models. Journal of Organizational Computing and Electronic Commerce, 20, Strejcek, G., & Theil, M. (2002). Technology push, legislation pull? E-government in the European Union. Decision Support Systems, 34(1), Wescott, C. (2001). E Government in the Asia pacific region. Asian Journal of Political Science, 9(2), West, D. (2004). E-Government and the Transformation of Service Delivery and Citizen Attitudes. Public Administration Review, 64(1), Yildiz, M. (2007). E-government research: Reviewing the literature, limitations, and ways forward. Government Information Quarterly, 24(3), Yin, R. (2003). Case Study Research: Design and Methods. Washington, DC: SAGE Publications. 74

75 Appendices 1. Overview of models of growth found in the systematic literature review Author(s) Year Main focus Process oriented Grijpink 1999 G2G Yes 5 Informal talks Formal (1999) consultation Gartner (Baum and Di Maio) Layne & Lee Hiller & Belanger Deloitte (e.g. in Silcock) 2000 G2C/ G2B 2001 G2C/ G2B 2001 G2C/ G2B 2001 G2C/ G2B Ronaghan (UN) 2001 G2C/ G2B Wescott 2001 G2C/ G2B Netchaeva 2002 G2C/ G2B Chandler and Emanuels Peristeras, Tsekos & Tarabanis 2002 G2C/ G2B Nr. Stage 1 Stage 2 Stage 3 Stage 4 Stage 5 Stage 6 Joint decision making Ad hoc joint action No 4 Presence Interaction Transaction Transformation Yes 4 Catalogue Transaction Vertical Integration No 5 Info Two-way Service and dissemination communication financial and catalogue transaction No 6 Information publishing No 5 Emerging presence No 6 and internal network No 5 Specific information presented Official two-way transactions Enhanced presence Enable interorganization al and public access to information Interactive websites Multi-purpose portals Interactive Two-way communication Forums and opinion polls Horizontal Integration Vertical and horizontal integration Portal personalization Transactional government Exchange of value Some services online No 4 Information Interaction Transaction Integration 2002 G2G No 4 Islands of Automation Automated Process Chains Reengineering through Information Technology Total Reinvention Common network organisation Political participation Clustering of common services Seamless/Fu lly integrated Digital democracy Unified government portal Full integration and enterprise transformati on Joined-up government 75

76 West 2004 G2C/ G2B Siau & 2005 G2C/ Long G2B EU 2006 G2C/ Commision G2B (Wauters) Anderson & Henriksen Papantonio u et al. Gottschalk & Solli- Sæther Sarantis, Charalabidi s & Psarras Klievink & Janssen Klievink & Janssen 2006 G2C/ G2B 2007 G2C/ G2B No 4 Billboard Partial-Service- Portal Stage Interactive Stage Delivery stage democracy No 5 Web presence Interaction Transaction Transformation e-democracy No 4 Information available online One way interaction Two way interaction Electronic case handling (transactions) No 4 Cultivation Extension Maturity Revolution Yes (change managem ent) 2008 G2G Yes (developi ng capabilitie s) 2008 G2G Yes (developi ng capabilitie s) 2009 G2G Yes (dynamic capabilitie s) 2010 G2G Yes (dyn. capabilitie s) 4 Static Information 4 Aligning Work Processes Citizen interaction Sharing Knowledge Knowing the citizen Joining Value Creation Full e- government Aligning Strategies 5 Independent Ad hoc Collaborative Integrated Unified 5 Stovepiped applications Integrated organizations Nationwide portal 4 Ad hoc Coordinated Shared infrastructure Interorganizatio nal integration Orchestrated Demanddriven joined up government 76

77 2. Protocol preliminary interview case study (Dutch) Introductie Zoals ik in mijn al heb aangegeven, ben ik op dit moment aan het afstuderen in het kader van de opleiding informatiekunde aan de Universiteit van Utrecht. Dit is een studie die theorie vanuit de bedrijfskunde en organisatiewetenschappen combineert met kennis op het gebied van informatie en computerwetenschappen. In mijn afstudeeronderzoek probeer ik inzicht te krijgen in de samenwerking tussen overheidsorganisaties die binnen een bepaalde keten een gemeenschappelijk doel nastreven. Ik onderzoek hierbij of er verschillende niveaus in deze samenwerking zijn te onderscheiden op technologisch en organisationeel vlak. Om de verschillende niveaus van samenwerking in deze ketens te kunnen beschrijven, heb ik een model gecreëerd bestaande uit vijf fasen van samenwerking en verdeeld in een technologische, semantische en organisationele dimensie. In mijn casusonderzoek wil ik dit model toepassen op een aantal casussen om het te testen en eventueel te verbeteren. Ik heb hiervoor de omgevingsvergunning als casus geselecteerd omdat er verschillende organisaties betrokken zijn bij het proces van een vergunningsverlening. Ik heb u benaderd voor dit interview omdat u bent betrokken als bij de omgevingsvergunning en daarom meer informatie kunt verschaffen over dit onderwerp. Doel van dit interview is enerzijds om meer informatie over de omgevingsvergunning te verkrijgen en anderzijds om doelprovincies/gemeenten te selecteren die verder geanalyseerd zullen worden. Casus afbakenen Algemeen 1. In welke rol bent u betrokken bij de omgevingsvergunning? 2. Bij welke provincies/gemeenten bent u betrokken m.b.t. de omgevingsvergunning? 3. Kunt u een algemeen beeld geven van de partijen die betrokken zijn bij het afgeven van een omgevingsvergunning? Oriënteren op scenario 4. Kunt u voorbeelden geven van situaties die de invoering en aanvraag van de omgevingsvergunning complex maken? Bijvoorbeeld door: a. Het aantal partijen dat betrokken is? b. De hoeveelheid informatie die uitgewisseld dient te worden? 5. Komen dergelijke situaties bij elke provincie/gemeente voor? Locatie bepalen 6. Welke provincies/gemeenten lopen achter/voorop in de invoering van de omgevingsvergunning? 7. Welke provincies/gemeenten onderscheiden zich in de aanpak van de omgevingsvergunning van andere provincies/gemeenten? Waarin onderscheiden zij zich? 8. Zijn er grote verschillen in de invoering van de omgevingsvergunning tussen de verschillende provincies/gemeenten? 77

78 Doorverwijzing vervolg interviews 9. Heeft u contactgegevens van personen die werkzaam zijn bij de provincies/gemeenten die u beschreven heeft? 10. Heeft u suggesties voor andere personen die een bijdrage zouden kunnen leveren aan dit onderzoek? 78

79 3. Protocol in-depth interview case study (Dutch) Introductie Zoals ik in mijn al heb aangegeven, ben ik op dit moment aan het afstuderen in het kader van de opleiding informatiekunde aan de Universiteit van Utrecht. Dit is een studie die theorie vanuit de bedrijfskunde en organisatiewetenschappen combineert met kennis op het gebied van informatie en computerwetenschappen. In mijn afstudeeronderzoek probeer ik inzicht te krijgen in de samenwerking tussen overheidsorganisaties die binnen een bepaalde keten een gemeenschappelijk doel nastreven. Ik onderzoek hierbij of er verschillende niveaus in deze samenwerking zijn te onderscheiden op technologisch en organisationeel vlak. Om de verschillende niveaus van samenwerking in deze ketens te kunnen beschrijven, heb ik een model gecreëerd bestaande uit vijf fasen van samenwerking en verdeeld in een technologische, semantische en organisationele dimensie. In mijn casusonderzoek wil ik dit model toepassen op een aantal casussen om het te testen en eventueel te verbeteren. Ik heb hiervoor de omgevingsvergunning als casus geselecteerd omdat er verschillende organisaties betrokken zijn bij het proces van een vergunningsverlening. Ik heb u benaderd voor dit interview omdat u bent betrokken als bij de omgevingsvergunning en daarom meer informatie kunt verschaffen over dit onderwerp. Het doel van dit interview is om meer inzicht te krijgen in de samenwerking tussen verschillende overheidsorganisaties en andere betrokken organisaties bij de totstandkoming van de omgevingsvergunning. Deze informatie zal ik vervolgens gebruiken om mijn model te toetsen. Vragen Organisatie & keten 1. Welke functie heeft uw organisatie binnen deze keten? 2. Welke functie heeft u binnen deze organisatie? 3. Met welke andere instanties werkt uw organisatie regelmatig samen in deze keten? Scenario 4. Komt dit scenario voor binnen uw organisatie? 5. Welke organisaties spelen een rol in dit scenario? De volgende vragen hebben betrekking op het eerder besproken scenario: Technische volwassenheid* 6. Hoe ziet de infrastructuur waarover gegevensuitwisseling plaatsvindt eruit in de keten? 7. Welke standaarden worden toegepast voor gegevensuitwisseling? 8. Welke standaarden zijn specifiek ontwikkeld voor deze keten? 9. Welke applicaties/systemen zijn specifiek ontwikkeld voor deze keten? *Voor elke vraag nagaan voor welk gedeelte van de keten dit geldt. Semantische volwassenheid* 10. Welke informatie komt de organisatie binnen? 11. Welke informatie wordt verwerkt binnen de organisatie? 79

80 12. Welke informatie komt uit de organisatie? 13. Welke standaarden met betrekking tot semantiek zijn aanwezig binnen de keten? (standaard vocabulaire etc.) 14. Welke domeinkennis is gestandaardiseerd en gedeeld in de keten? *Voor elke vraag nagaan voor welk gedeelte van de keten dit geldt. Organisationele volwassenheid 15. Zijn de bedrijfsprocessen in andere organisaties bekend bij uw organisatie? 16. Zijn de bedrijfsprocessen van alle organisaties in de keten gedocumenteerd? 17. Sluiten de bedrijfsprocessen van uw organisatie aan op die van andere organisaties in de keten? 18. Is er overeenstemming bereikt over standaardisatie van processen in de keten? Volwassenheidsmodel 19. Welke specifieke veranderingen hebben gezorgd voor een verbetering van de samenwerking op: a. Technologisch gebied? b. Semantisch gebied? c. Organisationeel gebied? 20. Wat weerhoudt volgens u op dit moment de verbetering van de samenwerking op: a. Technologisch gebied? b. Semantisch gebied? c. Organisationeel gebied? Samenhang dimensies 21. Zijn de organisationele ontwikkelingen gestuwd door de technologische ontwikkelingen? 22. Zijn de semantische ontwikkelingen gestuwd door de technologische ontwikkelingen? 23. Zijn de technologische ontwikkelingen gestuwd door de organisationele ontwikkelingen? 80

81 4. Contacts & interviewees Melior Hoekstra - PwC Neal Muusze - PwC Tjitske Visser - PwC Emile Weesie - Advisor Change management organizations Marc de Rooij - Ministerie van VROM Marthijn Antonisse - Municipality of Haarlem Hans van Everdingen - Municipality of Haarlem Aad van den Ende - Municipality of Ede Jorrit ten Böhmer - Municipality of Ede Nol Karbet - Municipality of Leiden 81

82 5. Simplified process model All-in-one-permit Figure 17: Simplified process model of the all-in-one-permit application procedure 82