Migration Strategies for Multi-Channel Service Provisioning in Public Agencies

Abstract

Many public agencies aim to realize better service delivery for citizens and businesses by implementing multi-channel service provisioning (MCSP). MCSP coordinates service delivery channels to enable citizens and businesses to choose their preferred channel for their contact with the government. As the introduction of MCSP requires substantial organizational and technical change, organizations are in search of the right migration strategies. This study explores migration strategies for MCSP deployed in practice. We found that migration strategies for MCSP can be divided into two types: design and change management strategies. By looking at four case studies and evaluating the findings during an expert workshop, we found that organizations prefer a gradual migration process, as both step-wise migration and the celebration of quick-wins were often mentioned strategies. Furthermore, the strategy to implement MCSP starting in the front office was observed more often than starting in the back office, as the client needs were usually taken as a starting point for the migration. And finally we found that the public agencies under study consider having the right change management strategy in place more important than deploying the right design strategy. [PUBLICATION ABSTRACT]

Headnote

Abstract: Many public agencies aim to realize better service delivery for citizens and businesses by implementing multi-channel service provisioning (MCSP). MCSP coordinates service delivery channels to enable citizens and businesses to choose their preferred channel for their contact with the government. As the introduction of MCSP requires substantial organizational and technical change, organizations are in search of the right migration strategies. This study explores migration strategies for MCSP deployed in practice. We found that migration strategies for MCSP can be divided into two types: design and change management strategies. By looking at four case studies and evaluating the findings during an expert workshop, we found that organizations prefer a gradual migration process, as both step-wise migration and the celebration of quick-wins were often mentioned strategies. Furthermore, the strategy to implement MCSP starting in the front office was observed more often than starting in the back office, as the client needs were usually taken as a starting point for the migration. And finally we found that the public agencies under study consider having the right change management strategy in place more important than deploying the right design strategy.

Keywords: multi-channel service provisioning (MCSP), integrated service delivery, multi-channel management, migration strategies, change strategies, eGovernment

1. Introduction

eGovernment is concerned with creating better service delivery to citizens and businesses (Curtin and al. 2004). Citizens and businesses are demanding faster delivery of public services and much better information (Ongaro 2004). One of the elements that can improve public service delivery is implementation of multi-channel service provisioning (Parisopoulos et al. 2009). Multi-channel service provisioning (MCSP) is the use of multiple service delivery channels as part of one public service delivery process. The rationale of MCSP is that citizens or business can use their preferred channel to interact with the government, or change their preferred channel depending on their needs and circumstances while interacting with the government. While the service desk may be preferred for complex services, the internet may be preferred for tracking and tracing. MCSP contributes to the creation of integrated service delivery, as service delivery across multiple channels is integrated (Kernaghan 2005). From the point of view of the organization new electronic channels can be used to lower costs of service delivery (Ebbers et al. 2008). Service channels are often viewed as complementary to each other. Yet new channels may substitute the old - often expensive - channels to lower costs. The main complexity is that these new channels need to be synchronized with the traditional and other new channels such as the telephone, twitter and the service desk.

Implementing eGovernment is considered a challenging process as it requires substantial organizational and technical change (e.g. Bannister 2001; Beynon-Davies and Williams 2003). For years it has been acknowledged that silos should be dismantled (Bannister 2001); with MCSP it is necessary that these silos are coordinated and synchronized. Given the complexities involved, many organizations struggle to implement MCSP and are in search of the right migration strategy. Determining the right migration strategy likely depends on the situation within an organization and on its environment, but not much is known about which strategies are effective in specific circumstances (Van Veenstra et al 2009). This study investigates which migration strategies are deployed in practice and explores which migration strategies for MCSP are effective in the specific circumstances of an organization and its environment. First, migration strategies were identified in an expert workshop. We found that migration strategies for MCSP can be divided into two categories: design strategies that describe the migration process from the perspective of the order of implementation and change management strategies that describe the migration from a perspective of managing the changes in an organization. For each of the two categories we identified four different strategies.

Then, to find out which of these strategies are deployed in practice and in which specific organizational context, four cases studies in the Netherlands were examined: two executive agencies and a large and a mid-sized municipality. The organizations are in different stages of MCSP implementation. The findings from the case studies were evaluated during an expert meeting consisting of government officials that are involved in the implementation of MCSP. After investigating the migration strategies used in practice, a comparison is made between these two categories to investigate whether organizational characteristics influence the migration strategies used. This paper first presents an overview of the migration strategies identified. Then, the four case studies are described followed by the findings from the cases and the evaluation of the expert meeting. Finally, we present conclusions and recommendations.

2. Migration strategies

Many public organizations are in search of a migration strategy for MCSP. During an expert meeting with practitioners implementing MCSP we found that migration strategies can be broadly divided into two types: strategies that are concerned with the design (or the 'product' of implementation) and strategies that guide the implementation process (or the process of change). Design strategies are those strategies that determine the order of the migration to MCSP. These are called design strategies as they are usually defined - or designed - at the start of the implementation process, e.g. by designing enterprise architecture (Armour et al. 1999; Janssen and Van Veenstra 2005). They are usually based on the vision for multi-channel service delivery in an organization.

Change management strategies are those strategies that guide the implementation project. They are not concerned with choices of how to design MCSP but with the process of migration. In contrast with the design strategies that set out progress beforehand, they mainly deal with the challenges that come up during the migration process. They are meant for managers that have to bring about organizational change. During the migration process the influence of managers and stakeholders is considered important, as a solution is dependent on their support (De Bruijn and Ten Heuvelhof 2000; Chisholm 1989; Fernandez and Rainey 2006; Powell 1991) and the migration strategy has to be flexible enough to mitigate any problems that come up during the process (Van der Voort et al. 2009).

During a workshop with practitioners involved in implementing MCSP in the public sector we identified eight common decisions that need to be taken when implementing MCSP. Although these decisions are presented separately, they may influence each other. The first decisions refer to the design strategies (1-4), whereas the latter refer to decisions concerning the change management strategies (5-8) for migration to MCSP:

* 1. Starting by changing the front office or the back office?

* 2. Starting by changing the technology infrastructure or the organization?

* 3. Changing the channels one by one or all at the same time?

* 4. First implementing quick-wins or radically redesigning the whole organization?

* 5. Big-bang migration or step-by-step change?

* 6. Managing migration as a 'project' or as a 'process'?

* 7. Implementing change top-down or bottom-up?

* 8. Making middle management or the project team responsible for the migration?

In order to research the occurrence of these strategies in practice, we use 'ideal types'; the strategies are defined by presenting the extremes. The strategies are identified in such a way that they describe the migration from scratch; as migration takes place in an organization without MCSP. In such a situation all service delivery channels are developed and maintained separately. After migration MCSP is fully implemented: service channels are coordinated and communication strategies are in place. In order to get from the first situation to the second situation, the two extremes of the strategy can be followed during the migration process. Although we do not expect to find any of these 'ideal types' in practice, but rather a combination of both extremes, ideal types have a heuristic use to explain a specific situation. The migration strategies are elaborated in the next section.

2.1 Design strategies

2.1.1 'Starting by changing the front office or the back office?'

The one extreme of the first design strategy is to start with changing the front office and the other extreme is to start by changing the back office. In case the decision is made to start by changing the front end, the migration often starts by making changes that are quickly beneficial for customers, such as identifying how the service delivery can be set up in a more client-oriented manner. After these decisions and reorganizations have been made, the content of the channels is harmonized and finally changes are made in the back office to ensure that the different channels in the front office use the same information provided by the back office systems. The disadvantage of this strategy is that it may take long before the changes in the back end have followed the changes in the front end. The risk associated with starting with changing the front end is, thus, that the back end cannot 'keep up' with a MCSP front end, leading to a suboptimal performance of service delivery.

The second option is to start at the back office and harmonize processes, information and systems first, to ensure that after these changes have been made the front office can be easily migrated using the new back office systems to improve service delivery. This strategy focuses on creating the technological and organizational foundations before moving forward. Although this path is more likely to result in a coherent design of MCSP, this path is also found to be very costly. Large investments in information systems will have to be made before any improvement of service delivery can be noticed by the customer. This strategy may not be aligned with the objectives of elected politicians to show visible results quickly, or at least within their term of office. The main risk of this strategy is, thus, that high investments in a 'heavy' back office will not lead to improvements for customers in the short-term, causing the migration to be labeled a failure before benefits have been realized.

2.1.2 'Starting by changing the technology infrastructure or the organization?'

This strategy is concerned with starting to migrate to MCSP by implementing new technological infrastructure for MCSP or by re-organizing the organization. For migration to MCSP, both need to be changed as both work processes and information technology need to be adjusted to the new situation. The technological infrastructure provides the foundations and facilities for creating MCSP or eGovernment (Janssen et al. 2009). The organization needs to harmonize its activities to be able to serve clients in the same way though every service delivery channel, while the software applications may need to be changed to support this harmonization. Both changes can be the starting point of the migration process.

When the choice is made to start with changing the organization and its work practices first, activities such as training employees are the starting point of the migration process. The organization will be changed from a functional to a customer-oriented organization structure. Also re-organizing the business processes and work practices are part of this adjustment process. The main pitfall is, however, that the information systems that are subsequently implemented do not follow the changes in the organization, thereby undoing part of the organizational changes. This means, that while the organization is ready for the new situation, service delivery may be still reliant on old information and systems, causing employees to fall back on their old habits.

Starting by implementing new information technology could mean that an application integration infrastructure is introduced. An organization can be divided in a front office and a back office (Richardson 1994). A technical infrastructure that can be used as an information broker for sharing information between front office and back office is called mid office. The advantage of a mid office implementation is that harmonization between front office and back office is taken care of. The pitfall of starting with the technical implementation is that it may be costly to implement new technology first without training employees to use the IT, which may result in a slow process with high costs, in which clients do not notice any of the changes. In this case, the main risk is that the potential of information technology is not realized or that it takes very long before it is realized.

2.1.3 'Changing the channels one by one or all at the same time?'

The choice between the strategies of changing all channels at the same time or rather one by one is represented by the third migration strategy. Changes can be made according to topic, by changing this topic in all channels at the same time, or they can be made by changing all topics at the same time within one channel. The main advantage of starting with the former, is that harmonization of the content is likely to happen smoothly. There is, however, a low tolerance for mistakes, as service delivery cannot be shut down altogether. In case channels are migrated by attuning them to the other channels one by one, risk of this happening is lower, because at least one channel will be available for clients at any time. The reliability of the service provisioning is, thus, higher. Synchronization of channels, will, however, be more difficult in this case. The main risk associated with a migration process changing channels one by one is that it may resulting a situation in which there are discrepancies between service channels, which is undesirable for MCSP.

2.1.4 'First implementing quick-wins or radically redesigning the whole organization?'

The fourth design strategy represents the choice between first identifying and implementing quick-wins as visible intermediary results or starting out with a radical redesign of the organization as well as the new information technology. The latter strategy requires designing a whole new situation and getting a precise overview of everything that needs to be altered, such as work processes, technology and organizational changes. Its greatest asset as well as its greatest risk is that it is all-encompassing. The complexity attached to this situation may lead to large failures, whereas making smaller changes may result in a less coherent redesign but it is also less risk-prone. When a choice is made to first identify and implement quick-wins, the first results are visible quickly. This my help in convincing people who resist changing the organization for MCSP. Overall, there is a risk associated with this strategy that the redesign will be less coherent, which may result in a longer lead time, as additional changes need to be made constantly to be able to fully migrate to MCSP.

2.2 Change management strategies

2.2.1 'Big-bang migration or step-by-step implementation?'

A first change management strategy is represented by the choice between big-bang implementation of all necessary changes or a change process that takes place step by step (Van der Voort et al. 2009). Whereas a big-bang migration aims to quickly implement all necessary changes, step-by-step change results in a longer-term gradual implementation process. The former strategy, however, requires much more control than the latter, as the changes are more likely to influence each other and delays need to be overcome quickly. The risk of failure, however, is also larger because of the interrelations between changes. Furthermore, there is a risk involved that employees of are resistant to a big-bang implementation of organizational changes. Step-by-step implementation aims at a migration in a gradual manner that can be managed easily. An associated risk is, however, that discrepancies in the design emerge as the process evolves over a longer period of time.

2.2.2 'Managing migration as a 'project' or as a 'process'?'

A second choice that needs to be made while implementing MCSP is whether to manage the migration as a 'project' or as a 'process' (De Bruijn et al. 1998). The former means that the migration process is tightly managed by steering towards clearly defined results, whereas during the process of migration in the way of the latter strategy happens in a more organic manner. In this case there is more room for building commitment by the various stakeholders during the project. Whereas lack of consensus for the new situation is a major risk of the former strategy, the latter strategy has a slow implementation process as a risk as well as the risk of incoherency emerging as a result of focusing on building consensus instead of steering towards clear results (De Bruijn et al. 1998).

2.2.3 'Implementing change top-down or bottom-up?'

The third change strategy represents the choice between top-down and bottom-up migration. Whereas top-down migration requires a strict planning by the management, bottom-up migration originates at the bottom of the organization, within the operational layer. Top-down enforcement has similarities with radical redesign as well as with a project implementation style as both need careful management setting clear guidelines and deadlines. Bottom-up migration on the other hand, needs careful coordination as it is likely to lead to quick-wins for customers and for operations, but it may also lead to a slow implementation process that halts half-way without truly changing the whole organization. It also requires involvement of the employees on the operational level, who need to drive the change. Although this presents a risk, it also provides an opportunity to involve them in the change process. The main risk of top-down migration is resistance to change that is enforced on the operational level and a lack of support for MCSP.

2.2.4 'Making middle management or the project team responsible for the migration?'

The final change management strategy concerns the responsibility for the migration process. A choice can be made to make the middle management responsible or a specialized project team can be set up. In case the project team is responsible, the team has more freedom to be innovative as they operate at a greater distance from everyday operations. Its risk is, however, that the organization may not accept a solution designed by a project team that is seen as an 'outsider'. When the middle management is responsible for the migration, the process is more likely to become easily accepted as part of day-to-day operations. The risk is, however, that the current way of doing business is the dominant factor in determining the migration process. This could mean that innovations do not find their way in the organization, implementing MCSP in a suboptimal way.

3. Case studies

In order to find out which of the eight migration strategies that were identified in the previous section are used in practice, we carried out four case studies in government organizations implementing MCSP in the Netherlands. The organizations were selected as they were in different stadia of development and we explored which migration strategies they employed as well as the reason for choosing a specific migration strategy. Furthermore, we investigated situational factors that were considered influential in determining the choice for a specific migration strategy for MCSP. Two cases are large executive agencies and the other two cases are local government organizations, one large municipality and one medium-sized. We used a combination of studying policy documents and semi-structured interviews for gathering information. Six people were interviewed in total, including four project leaders responsible for implementing MCSP as well as an architect and a manager of the service delivery department. The interviews were held in June and July 2009. Each interview lasted around one hour.

3.1 Mid-sized municipality

The first municipality under study is a local government with a little over 200.000 inhabitants. A few years ago the municipality set out to improve its service delivery after a very critical article appeared in a local newspaper. Implementation of MCSP was one among a number of initiatives to improve service delivery to citizens. The department driving these improvements of service delivery was the citizens' services department, as this department was the front runner in the area of service delivery within the municipal organization.

For implementation of MCSP an organization-wide action plan was set up by a project team within the citizens' services department and all other municipal departments were persuaded to get on board. The migration, thereby, effectively started in the front office of the organization and was initiated in a bottom-up manner. Simultaneously, MCSP was also enforced top-down as the management team of the municipality announced that it would oblige departments to migrate to MCSP. Until this enforcement was effected, a strategy was pursued aiming to achieve efforts bottom-up within those departments willing to change their way of working. Another top-down strategy deployed by the project team was to make MCSP a political issue ensuring that a sense of urgency is created in the organization. For political reasons, the potential efficiency gains of MCSP were stressed within the implementation program.

The migration took place stepwise by first engaging those departments that are willing to collaborate and change their work practices. Furthermore, the migration took place channel by channel by making 'smart combinations' and finding quick-wins in the existing information systems. The main reason for this was that it was believed to be important to celebrate the achievements during the project. These achievements were then used to create commitment among managers, who were ultimately made responsible for the migration to MCSP within their department. In time, a more top-down enforcement became effective for those departments in the organization not willing to change on a voluntary basis. Until then, fragmentation of the organization was considered a major obstacle as budgets are tied up with departments. It remained difficult, therefore, to develop a plan for the organization as a whole.

3.2 Large municipality

In 2005 this municipality started a program to improve its service delivery across the organization. Shortly afterwards, the program merged with the municipality's citizen's service department. The first phase of the program, between 2005 and 2008, aimed to develop the different service delivery channels, separately. Since 2008, joining up of services using life events and implementation of MCSP were aimed for. For the implementation of MCSP, an action plan was set up to develop the internet channel further and to make this channel leading in all service delivery. An important reason for this was the financial situation of the municipality that made setting up service delivery in a more cost-efficient way even more important than before. As the front office initiated the necessary changes, back office and mid office began to lag behind and their information supply and content management needed to be made up-to-date to enable further improvements in service delivery. Although it was considered important that the back office was sufficiently developed to support MCSP, it was decided that the program could not wait for this. Therefore, development of the channels and coordinating the channels continued at the same time.

The first steps of migrating to MCSP were made by identifying quick-wins, for example by making the migration of the top-10 of the most used products (e.g. parking license issuance and complaint filing) a priority. In this way, the first benefits would be visible to citizens soon. The employees working in the service delivering front office were involved in the changes that were made, as they possess a lot of knowledge on how to improve their daily operations. Purely enforcing things top-down was not believed to achieve the right things, while stakeholder involvement was considered very important for showing the organization how MCSP can be used to improve operations. The topic by topic migration was, thus, undertaken by employing a process management strategy, engaging stakeholders from the different sub-departments for every step that was taken. The organization was convinced that this was the only way to commit all its different departments to such a large-scale changes.

Implementing MSCP in this large municipality was even more complex than in a smaller municipality because of the large, highly autonomous departments within the organization. Additionally, the administration of this municipality consists of a number of elected area councils for specific neighbourhoods in addition to a city council for the whole municipality. Service delivery to citizens is undertaken by these offices instead of centrally. As a result, a number of different websites and offices existed throughout the city with a relative high degree of autonomy. To get all departments and organizations on board was a difficult process, especially as many local politicians had different targets for the service delivery in their area. To mitigate some of the fragmentation, the implementation program deliberately sought top management support. Another factor considered very important was to include management and steering information in the information systems. Otherwise, management would not be able to understand the implications of implementation.

3.3 Inland Revenue Service

The Inland Revenue Service (IRS) already had well-developed channels, but its channels were hardly coordinated. Instead, all channels were supported by separate information systems and different departments were responsible for the various service channels, which made it difficult to harmonize the service delivery of channels. Still, the first steps towards MCSP were taken as the organization saw the need for coordinating its service delivery channels. MCSP migration would be driven by client demands, as this was considered to be the only way to achieve the necessary consistency of service delivery. Segmentation of clients was used as a way to distinguish channels as different client groups have different needs and wishes for their contact with the organization.

Before migration started, to support the front end, the information used in the back office for the different service channels needed to be harmonized. The first steps towards this harmonization were made by sharing the data used by the telephone (the leading channel) with the internet channel and the service desks. But before any coordination could take place, the back office supporting the channels with essential data needed to be updated first. If the necessary data were not available for the different channels in a synchronized manner, MCSP could not be achieved. Therefore, work process change was considered as the only possible starting point for the migration. Furthermore, the organization started out by migrating the processes that have the largest scale as this would have most impact on the organization.

The interviewees indicated that the most important strategies for migration to MCSP were related to the organizational change that needed to be made to ensure that the whole organization works with a consistent image of the client in mind. This change took place step by step as making large-scale changes at once was not feasible in such a large organization. Also, to manage change, middle management remained responsible for the migration as they were ultimately responsible for the service delivery of the organization. The front office changes preceded the changes in the back office as this allowed all process changes to be undertaken in a client-centred manner.

3.4 Student loan organization

Service delivery of the student loan organization used to be organized very badly. So badly, that total redesign of the organization was considered necessary; low ratings of its services required radical change. Coordinating the service channels was considered necessary to solve the problems of the telephone channel that was unable to handle all service requests. Therefore, one front office department was set up to support MCSP. Its main objective was to relieve the phone channel and attract more people to the website. This front office department harmonized the client image in order for all channels to treat the clients in the same manner using the same information. Service delivery, thereby, became a second primary process (in addition to handling student loans) instead of a secondary process. By now, the organization has successfully attracted many clients to its website.

This migration took place incrementally over the course of five years, and currently, the organization successfully employs MCSP as one of the first organizations in the Netherlands. However, as the front office was the starting point of implementing MCSP, the back office started to represent a major problem for multi-channel service delivery, as it could not keep up with the re-organized front office. Back office systems, thus, were subsequently migrated to support the new situation. To achieve the migration top management support has been paramount. The drive for efficiency ran into resistance at other departments that felt threatened by the developments. Therefore, it was necessary to have a formalized mandate for MCSP to be implemented, supported by the management levels. This mandate functioned as a 'stick' forcing departments to change according to the requirements of the service delivery department for MCSP.

4. Findings

After the strategies for implementing MCSP deployed in the case studies were identified, they were evaluated in an expert workshop held in December 2009. Around twenty-five officials from government agencies implementing or planning to implement MCSP took part in this workshop.

4.1 Findings from the case studies

The findings from the case studies per strategy are summarized in Table 1. This table shows that a variety of strategies is employed. Still, some recurring patterns can be observed. The design strategies that had a clear tendency towards one of the extremes were the first (front office or back office) and the fourth (quick-wins or radical redesign). To be able to improve the service delivery to clients quickly, changing the front office first was considered to be the best option in all cases. Also the need to create 'a consistent image of the client' throughout the organization was mentioned as a reason to change the front office first. Our interviewees remarked that perhaps in case the back office has considerable legacy it may need to be updated first to be able to achieve a sufficient degree of information quality that is necessary before the front office can be updated. The second design strategy that was widely considered more suitable for migration to MCSP was a migration process implementing quick-wins first before moving on to organization-wide redesign. The celebration of quick-wins to create support among 'non-believers' was found to be very important, while sense of urgency was mentioned as the one incentive for the choice to radically redesign the organization first. The two strategies that were not clearly indicated to entail a best option were said to depend on the specific circumstances. For instance, the choice for changing all channels at the same time or one-by-one was said to depend on the degree to which the channels already use the same information.

The preferred change trajectory was incremental to ensure a high level of involvement and participation. The tendency to migrate gradually became apparent as this was considered better manageable than taking a 'big-bang' approach. This can be explained by the Dutch organizational culture which is characterized by negotiation and seeking consensus. Similarly, three of the four case studies favoured managing the migration as a 'process' giving greater attention to creating support for the change process. Still, some portions of projects were managed as a 'project', as this was found to be more efficient. Furthermore, these three cases emphasized the need for process management and for gaining support of all (important) stakeholders. MCSP is considered to be a change that needs to have a broad support within the organization and process management is believed to be more suitable for such a large-scale migration process. Top-down implementation was more often mentioned than bottom-up because vision and leadership were considered very important in the migration process. Still, some cases did carry out parts of the migration bottom-up to involve the operational level in an early stage. And finally, responsibility tends to be in the hands of the middle management more often, although a project team is responsible for implementation in some cases. The reason for this was that the case studies favoured management support over innovation.

4.2 Evaluation of findings in expert workshop

After the strategies employed by the case studies were identified, we evaluated the findings in a second expert meeting. We set up a survey in which we asked the participants, firstly, which type of strategies they consider to be more important for achieving successful migration: design strategies or change management strategies. We asked them to rank the importance of design strategies in relation to change strategies on a seven-point scale ranging from (1) only the former strategy important to (4) both strategies equally important to (7) only the latter strategy important. Secondly, we asked the respondents which strategies they employ in their own organizations by asking them to score the extent to which each of the two axes of the eight migration strategies are deployed within their organizations on a comparable seven-point scale. We deliberately evaluated the findings from the case studies by asking the participates to express a preference for a type of strategy or to indicate which of the two extremes most resembled the situation in their organization. Although some of the participants indicated that they found it hard to make this decision as they maintained that both extremes go hand in hand, some patterns can be recognized giving insight in the preferences for certain migration strategies in practice. Thirteen people filled out this survey and its findings are summarized in Tables 2 and 3. The numbers in the tables refer to the amount of experts that chose the specific answer.

The findings of the expert workshop confirm the findings from the case studies. Nevertheless, some small differences can be observed. First of all, the workshop confirms the relative importance of having a change management strategy in place, which was observed in the case studies. Secondly, also the tendency to start with changing the front office was confirmed. A third observation is that while the implementation of IT was seen to be leading in some cases (even though all case studies considered organizational change to be very important), during the evaluation in the expert workshop a clear tendency to change the organization first can be observed. The interviewees from the case studies indicated that there are some situations in which it is useful to change the IT first. For example, this can be the case when the complexity of the technology to be implemented is low or in case there is it will take too long to wait for everyone in the organization to be ready for implementation. As was the case in the cases, also in the evaluation workshop the choice between changing all channels at the same time and changing the channels one-by-one is spread out more evenly.

The findings of the expert workshop confirm the findings from the case studies. Nevertheless, some small differences can be observed. First of all, the workshop confirms the relative importance of having a change management strategy in place, which was observed in the case studies. Secondly, also the tendency to start with changing the front office was confirmed. A third observation is that while the implementation of IT was seen to be leading in some cases (even though all case studies considered organizational change to be very important), during the evaluation in the expert workshop a clear tendency to change the organization first can be observed. The interviewees from the case studies indicated that there are some situations in which it is useful to change the IT first. For example, this can be the case when the complexity of the technology to be implemented is low or in case there is it will take too long to wait for everyone in the organization to be ready for implementation. As was the case in the cases, also in the evaluation workshop the choice between changing all channels at the same time and changing the channels one-by-one is spread out more evenly.

During the expert workshop celebrating quick-wins was considered more important but not much more important as was observed in the case studies, thereby implying that having a vision for the redesign is also valued highly. For the change management strategies, the importance of step-by-step change and managing the migration as a process are confirmed. And top-down implementation was more often mentioned than bottom-up because vision and leadership were considered very important in the migration process, while during the expert workshop a tendency towards bottom-up implementation can be observed. Finally, the importance of making the middle management responsible for the migration process is clearly confirmed by the expert workshop.

5. Discussion

In both the case studies and the expert workshop the importance of having a clear change management strategy in place was stressed, indicating that having the right design strategy is considered less important. All interviewees and respondents during the expert workshops indicated that undertaking such a complex organizational change requires most attention in guiding the changes within the organization. The exact order of changes was considered to be of secondary importance, although, naturally, having a plan for migration in a certain order was valued highly too. While this research makes use of ideal types representing two 'extremes' for analytical purposes, both the interviewees in the case studies and the experts in the workshop indicated that in reality often a mix between the two extremes can be found. Still, among the design strategies as well as the change strategies some strategies were considered more effective than others and patterns were identified.

Which factors influence effective migration likely depends on the organizational characteristics as well as on the organizational environment. Between the two categories of case studies examined, some fundamental differences could be observed potentially influencing the migration strategies deployed. Therefore, we investigated the major differences between the two types of organizations, which are summarized in Table 5.

Some of the major differences between the two types of organizations are the client base, the number and variety of products and the number of applications of a certain product and the political involvement. While the client base of an executive agency that deals with a specific government service is often more homogeneous, such as in the case of the student loan organization that deals with students, municipalities as well as the IRS have to deal with all inhabitants. They may, thus, encounter much greater differences in clients' needs and wishes than most executive agencies. The same is the case for the number and the variety of products on offer. While municipalities deal with a great number of different services and products, executive agencies can usually focus on less and more homogeneous services and products. The number of applications for a certain service, however, is usually much larger for executive agencies (the IRS and the student loan organization process millions of applications of a specific product every year) than for municipalities, who may, in certain cases, only process a few applications of a product per year. A final major difference between the two types of organizations is the political involvement. While every municipality has an elected local government involved in its operations, executive agencies fall usually under the responsibility of a Ministry. Only when things go wrong Parliament may decide to interfere in its business, but usually they have more discretionary space for action.

Which choice for a migration strategy is most successful likely depends on the situation within the organization that undertakes migration as well as on its environment. Situational factors that were found to be of influence in the case studies on the choice for a specific strategy include: coherence among the channels, complexity of the new situation, lead time, client-centeredness, acceptance within the organization, innovation, support for the change process and speed of first success. While looking at the situational factors and the difference between the different types of organization, factors of influence on the choice for a specific strategy include the scale of the processes and the number of applications for a single product, the degree of autonomy of the different departments having to cooperate realizing MCSP and the level of political involvement. Further research should look into how these factors influence the effectiveness of the migration strategies more closely. The impact and relative importance of these situational factors and organizational characteristics should also be looked at more closely by deriving critical success factors (CSF's) for migration to MCSP.

6. Conclusion

Migration to MCSP is a complex undertaking, requiring changes to occur in the organization, work processes and the technology as well as in the culture of the organization. Public organizations are in search of the right migration strategy for their specific circumstances. This paper explores which migration strategies are deployed by government organizations in the Netherlands and the circumstantial factors influential in choosing a specific change strategy. Eight decisions are identified influencing the strategy and subsequently divided into two types: design strategies and change management strategies. By looking at four case studies we have indications that practitioners consider having a change management strategy in place guiding migration more important than choosing a design strategy determining the order of migration. The findings from the case studies were confirmed by an expert panel evaluating the findings, although some minor differences were observed, mainly depending on differing situational factors.

Although our findings should be generalized with care, we observed that public agencies foremost manage migration to MCSP as a re-organization process. This is reflected in the higher occurrence of the migration strategy managing the migration as a 'process' carefully making decisions, than as an implementation 'project' with clear implementation deadlines. Furthermore, we found that organizations prefer a gradual migration process, as both a step-wise migration and the celebration of quick-wins are often mentioned strategies. Many organizations also consider the client to be the leading factor in reorganizing the organization, by choosing an order for implementing MCSP starting in the front office more often than in the back office. A final observation is that responsibility for the migration process is more often in the hands of the middle management than the responsibility of a separate project team. Further research should look into deeper the matter of which strategies are most suitable under which circumstances and determine critical success factors.

References

References

Armour, F.J., Kaisler, S.H. and Liu, S.Y. (1999) "A big-picture look at Enterprise Architecture," IEEE IT Professional, Vol. 1, No. 1, pp. 35-42.

Bannister, F. (2001) "Dismantling the silos: extracting new value from IT investments in public administration," Information Systems Journal, Vol. 11, No. 1, pp. 65-84.

Beynon-Davies, P. and Williams, M.D. (2003) "Evaluating electronic local government in the UK," Journal of Information Technology, Vol. 18, No. 2, pp. 137-149.

Bruijn, J. A. de, ten Heuvelhof, E.F. and in 't Veld. R.J. (1998) Procesmanagement. Over procesontwerp en besluitvorming, Academic service, Schoonhoven.

Bruijn, J.A. de and ten Heuvelhof, E.F. (2000) Networks and Decision Making, Lemma, Utrecht.

Chisholm, D. (1989) Coordination without Hierarchy. Informal structures in multi-organizational systems, University of California Press, Berkeley.

Curtin, G.C. and Sommer, M.H. et al. (2004) The world of eGovernment, Haworth Press, Binghampton.

Ebbers, W.A., Pieterson W.J. and Noordman, H.N. (2008) "Electronic government: Rethinking channel management strategies," Government Information Quarterly, Vol. 25, No. 2, pp. 181-201.

Fernandez, S. and Rainey H.G. (2006) "Managing Successful Organizational Change in the Public Sector," Public Administration Review, Vol. 66, No. 2, pp. 168-176.

Janssen, M. and van Veenstra, A.F. (2005) "Stages of Growth in EGovernment: An Architectural Approach," Electronic Journal of eGovernment, Vol. 3, No. 4, pp. 193-200.

Janssen, M., Chun S.A. and Gil-Garcia, J.R. (2009) "Building the Next Generation of Digital Government Infrastructures," Government Information Quarterly, Vol. 26, No. 2, pp. 233-237.

Kernaghan, K. (2005) "Moving towards the virtual state: integrating services and service channels for citizen-centred delivery," International Review of Administrative Sciences, Vol. 71, No. 1, pp. 119-131.

Ongaro, E. (2004) "Process management in the public sector: The experience of one-stop shops in Italy," International Journal of Public Sector Management, Vol. 17, No. 1, pp. 81-107.

Parisopoulos, K., Tambouris E. and Tarabanis, K. (2009) "Transformational Government in Europe: A Survey of National Policies," In: Lecture Notes in Artificial Intelligence 5736, Lytras, M.D. et al. (Eds.), Springer, Berlin Heidelberg.

Powell, W.W. (Ed.) (1991) Neither Market nor Hierarchy: Network Forms of Organization, Sage Publications, London.

Richardson, R. (1994) "Back-officing front office functions - organizational and locational implications of new telemediated services," in: Management of Information and Communication Technologies, Emerging Patterns of Control, ASLIB, London, pp. 309-335.

Veenstra, A.F. van, Janssen, M. and Klievink, B. (2009) "Strategies for Orchestrating and Managing Supply Chains in Public Service Networks," Electronic Journal of eGovernment, Vol. 7, No. 4, pp. 425-432.

Voort, H. van der, de Bruijn, H. and Janssen, M. (2009) "Shared Service Centers in the public sector: Benefits, risks and transformation strategies," in: Handbook of Research on ICT-Enabled Transformational Government: A Global Perspective, Weerakkody, V., Janssen, M. and Dwivedi, Y.K. (Eds.), Information Science Reference, Hershey, NY.