The agile enterprise is described as the highest level of the SOA Maturity Model introduced in Chapter 1. Many factors must converge to achieve the agile enterprise level. SOA provides the basic architectural pattern for defining, integrating, and managing the capabilities of the enterprise—the service units. Other factors support integration, optimization, security, accountability, and adaptation of those service units. Governance is what pulls it all together and directs the efforts to achieve enterprise agility and stockholder value.
Enterprise governance is defined as: The set of responsibilities and practices exercised by the board and executive management with the goal of providing strategic direction, ensuring that objectives are achieved, ascertaining that risks are managed appropriately and verifying that the organization's resources are used responsibly.(Board Briefing on IT Governance, second edition, ©2003 ITGI; all rights reserved.)
This definition has also been adopted by the International Federation of Accountants (IFA) as expressed in Enterprise Governance—Getting the Balance Right, 2004.
This definition applies to the agile enterprise. Agile enterprise governance is enterprise governance extended to exploit the consistency, visibility, accountability, and flexibility provided by the architecture and disciplines described in the previous chapters. This chapter describes a governance framework as a basis for governance of an agile enterprise. Alignment with the governance framework gives the board of directors and executive management improved oversight and control of the enterprise operation—and much greater assurance that the enterprise is doing the right thing and doing it well. It also empowers individual service units to optimize the value they contribute from an enterprise perspective while holding them accountable for meeting performance objectives and complying with enterprise directives and government regulations.
Governance of IT is a byproduct of agile enterprise governance. Enhancement of IT governance for SOA, as proposed by some vendors, by itself does not meet the needs of the agile enterprise. The IT organization must be driven by the business needs of the enterprise in the context of the business architecture, process optimization, integration and support of service unit capabilities, and delivery of customer value. In this chapter we include discussion of how the IT organization contributes to agile enterprise governance and supports optimization of enterprise operations.
The following sections highlight the benefits of agile enterprise governance.
Management of the enterprise as a composition of service units enables value chains to be quickly adapted or developed to address new business pursuits. This adaptation exploits existing capabilities while minimizing operational disruption and the need to develop new information systems.
Distributed responsibilities are explicitly defined for front-line responses to disruptive events. For events of strategic significance, assessment of impact and risk as well as relevant strengths and weaknesses drive consideration of appropriate actions.
Strategic planning is integrated into monitoring the enterprise ecosystem and managing enterprise operations so that strategic planning is not a periodic exercise but part of the continuous evolution of the enterprise. The changing competitive landscape, new challenges and opportunities, and product plans should be considered in the strategic planning context to ensure that the course is right and the enterprise is on the right course.
Responsibility is defined for ensuring accessibility of timely and consistent data, information, knowledge, and expertise for monitoring, analysis, planning, and decision making throughout the enterprise.
Specification of service unit interfaces as distinct from the details of service unit implementation enables service unit managers and employees to use their creativity and initiative to improve their operations within the limits defined by interface specifications, budgets, enterprise directives, and government regulations.
Well-defined service unit capabilities and interfaces, along with cost-based billing for service units, consolidation of redundant capabilities, and defined contributions to value chains, provide clarity of responsibility for performance and accountability for deviations.
Service units provide a structure, focus, and accountability for the application of regulations and assessment of compliance.
The governance framework defines responsibility for risk assessment, including security, business continuity, and cost control concerns as well as accountability for elimination, mitigation, or acceptance of risks.
Consolidation of similar capabilities, along with incentives for continuous improvement, can achieve economies of scale, not only in terms of improved service unit efficiency but in the ability to develop and sustain higher levels of expertise.
Centralized design of service unit requirements and the resulting organization structure reflect consideration of enterprise objectives through a disciplined analysis, design, configuration, and optimization of the enterprise value chains and management structure.
Transformation planning and management reflect consideration of the full impact of changes to the business architecture, and the architecture enables a measured and orderly transformation with steps that provide incremental business value.
Service units and their management chains have responsibility for optimizing their operations, whereas enterprise governance addresses performance from a value-chain perspective and ensures optimization at an enterprise level.
The IT organization, like other service groups, is responsible for optimization of its operations. In addition, at an enterprise level, standards and technology preferences minimize the diversity of information technology and enable optimization of application development and data-processing operations.
We now turn to consideration of the way governance of the enterprise should change to implement, continuously evolve, and exploit an agile enterprise architecture. Figure 9.1 depicts an agile governance framework. An actual organization structure may have additional activities and variations in structure, depending on the industry and characteristics of the specific enterprise. The point of this diagram is to identify the governance functions required and affected by transformation to an agile enterprise.
It is important to note that agile governance is not intended to be something separate from governance of the enterprise nor governance of IT; rather, it becomes a new approach to governance that adapts and extends traditional governance to align with a new business paradigm. That paradigm structures the enterprise as a composition of service units, exploits information technology, and optimizes business processes and the utilization of enterprise resources and capabilities.
In a fully mature agile enterprise, the entire enterprise is composed of service units and supports a consistent design and governance discipline. The primary organizational change for governance is the addition of service units at the executive staff level. Though a strategic planning activity is typical of current enterprises, strategic planning is extended for the agile enterprise. The other five executive staff service units shown in Figure 9.1 are, for the most part, new, or at least clarified. Information technology services are peer to other support service units such as finance and accounting, purchasing, and human resources. Each of these contributes in some way to agile governance. Finally, the product value chains represent the business operations that deliver customer value and are primary targets for business change.
Enterprises should not expect to achieve agility in a single transformation. The capability must be developed in steps, and those steps should each be designed to realize business value. The SOA Maturity Model discussed in Chapter 1 provides guidance on the improvements needed to advance to each level of maturity. An industry framework such as eTOM, discussed in Chapter 2, provides guidance on the identification and relationships of service units in a particular industry. Business objectives guide the selection of service units to be developed.
Many incremental steps achieve the goal, probably as quickly as would a major transformation, and the cost and risk of the incremental approach are much lower. As the enterprise matures, executive management can grow the executive staff service units that are needed to guide and manage the transformation and improve enterprise governance.
The following sections discuss each of the elements of this governance framework, focusing on the changes introduced by SOA and required for an agile enterprise.
Strategic planning is a conventional executive management activity, but it is extended for the agile enterprise to achieve a clear linkage to the operation and future design of the enterprise.
Figure 9.2 depicts a high-level view of the Business Motivation Model (BMM), a standard strategic planning model adopted by the Object Management Group. This is a foundation for strategic planning for an agile enterprise. The concepts are typical of common strategic planning practices.
Strategic planning is primarily the responsibility of the CEO with the participation of his or her direct reports. At the same time, there is a need for a support staff to gather information, support the analysis, develop the work products of the strategic planning effort, and coordinate with related activities. The leader of the strategic planning service unit typically reports to the CEO. The following points briefly describe each of the elements in Figure 9.2:
Figure 9.3 depicts the BMM structure modified to support strategic planning for an agile enterprise.
We have made five changes: (1) changed Influencers to Enterprise Intelligence, (2) added Business Architecture, (3) replaced Potential Impact with Risks within Assessment, (4) added Transformation to Means, and (5) added Standards and Technology to Means.
These five changes not only connect strategic planning to the operation of the enterprise; they provide the means for insight and participation in governance by the board of directors. Along with Strategic Planning, these changes align with the executive staff service units in the governance framework described in the previous section and in Figure 9.1. We discuss their roles in strategic planning activities in the following list, and their broader service unit capabilities and responsibilities in the sections that follow.
These additions to the strategic planning model are supported by corresponding executive staff service units depicted in the Agile Governance Framework illustrated in Figure 9.1. We will discuss each of these executive staff service units as well as the other units in the agile governance framework of Figure 9.1 in the sections that follow.
We define enterprise intelligence as “obtaining, interpreting, and presenting relevant information about the enterprise and its ecosystem.” Here, the Enterprise Intelligence (EI) service unit addresses the need for consistent information across the enterprise and regarding the enterprise and its environment. EI is responsible for management of information resources that support enterprise governance as well as other activities related to analysis, planning, collaboration, and decision making from an enterprise perspective.
EI is not the only place in the enterprise where these capabilities should be considered or developed, but EI should ensure that they are addressed from an enterprise perspective and applied consistently throughout the enterprise. Much of the actual development and support of intelligence capabilities are provided by the IT service units.
EI should address intelligence as represented in the framework of Figure 9.5, sometimes called the DIKW hierarchy. The origin of the information-knowledge-wisdom hierarchy is attributed to a 1934 poem by T.S. Elliot. Addition of the data layer and association with information technology is attributed to Russell Ackoff in 1988. In the long term, the agile enterprise must take this broad view of enterprise intelligence to optimize enterprise operation and agility and to make most effective use of its intellectual assets.
We briefly discuss each of the levels from an agile enterprise perspective in the following subsections.
Data are the records stored in databases and files and communicated over networks. Data represent facts—attributes and relationships—about people, places, and things. Data have no meaning to humans until they are put into a context and presented in a form that associates the facts with the real-world entities they represent and describe.
Data are the foundation of the intelligence hierarchy. A critical challenge of SOA is to synthesize data from sources in many service units—to realize relevant information, knowledge, and wisdom for the success of the enterprise.
To ensure access to consistent and timely data about the enterprise and its environment, EI should have ownership responsibility for the following:
Information is data in context that has meaning to humans. Humans require data to be presented in a way that reflects the meaning of the numbers and letters. Information is captured, communicated, and prepared by computers for interaction with humans. A human expresses information when writing an email message or submitting an order. The computer provides information when it produces a display or prints a report that presents the context and associated descriptions with the data elements being displayed. So, though the computer captures, stores, and communicates data, the systems are designed to capture and present the data to humans as information—meaningful information.
EI is responsible for ensuring that data are transformed and presented as meaningful information with consistent semantics (meaning) to support analysis, planning, and decision making. This may have a variety of forms. For example:
Knowledge is the expression of patterns, dependencies, and constraints occurring in the enterprise and the world in which it operates. Knowledge can be applied to understanding a situation or developing a plan. EI should support the management of enterprise knowledge, which may take a variety of forms:
Knowledge can be codified and applied by computers but within narrow domains. Rule-based systems, often called knowledge-based systems, operate on human knowledge in the form of rules and apply the knowledge to specific problems.
Techniques such as case-based reasoning and neural networks enable computers to “learn” from experience and apply that learning to specific situations. So a neural network (an artificial intelligence model that simulates neurons receiving signals and adapting to feedback) can learn which credit card charges are “normal” and which are questionable, to reduce credit-card fraud. The underlying data structures of a neural network link characteristics of the transaction through various weighted relationships to determine whether there is a significant risk of fraud.
The knowledge and the applications for such solutions are narrowly focused and require specialized development skills. On the other hand, computers deal with massive amounts of knowledge on diverse topics. This knowledge is produced and applied by humans for humans but stored and communicated as computer data, most often in the form of messages and documents. Unlike the records described in the preceding “Data” section, these messages and documents are not defined or structured for use by computers, so they are often called unstructured data. They are typically the primary subject matter of knowledge management.
Wisdom reflects consideration of values with an understanding of complex causal relationships, behavior, and consequences. Wisdom yields optimal solutions for design, problem-solving, planning and decision making that goes beyond the application of computational algorithms. Wisdom is usually viewed as a human quality. People apply wisdom to problems and opportunities. Wisdom also implies the ability to consider solutions “outside the box.” Computers don't apply wisdom; they only do what they are programmed to do, although computer scientists continue to work toward giving computers more human capabilities.
Wisdom is managed by engaging the right people in analysis, planning, and decision making and ensuring that they have appropriate access to available knowledge, information, and data. The expertise directory, created to support knowledge sharing, can also support engaging the right people for their wisdom.
Though EI may not be directly responsible for bringing wisdom to enterprise management, modeling tools help managers develop more effective mental models of the business so that they can be better prepared to exhibit wisdom in their plans and decisions. In particular models of organization structure, costs, business processes, value chains, and other enterprise design viewpoints will ensure consistency, support consensus building, and enable management of complexity.
Creation and refinement of the models are as important as the resulting representation. Design of the model requires the development of insights on the problem to produce a model that behaves in a way that is consistent with the real world. Simulation provides a validation of the model and can expose behavior that is not otherwise expected. Business analytics involves computational and mathematical techniques for discovery of correlations and behavioral characteristics in the enterprise ecosystem. These insights promote wisdom. Wisdom is essential in an agile enterprise if decisions and business changes are to be made quickly with appropriate results. Business models are discussed in greater detail in Chapter 10.
The Business Architecture (BA) service unit is the focal point for design of the enterprise. It is responsible for maintaining much of the Enterprise Business Model (EBM) that integrates models representing different viewpoints of the enterprise. The EBM is discussed in Chapter 10.
We have used the term business architecture rather than enterprise architecture because the latter has been taken by the IT industry to refer to the architecture of the information systems of an enterprise. Business architecture refers to the architecture of the enterprise from a business perspective—the way the components of the enterprise fit together to fulfill the business purpose and strategic plans.
BA develops and evolves the design of the organization structure, service units, and value chains as input to the strategic planning activity and receives direction from the strategic planning activity for further alignment with leadership objectives and insights.
BA should be a center of expertise for the design of the service-oriented architecture. At the same time, BA must be a center for collaboration with specialists from various segments of the business, to incorporate understanding of the effects of potential changes and achieve appropriate enterprise optimization.
Phased improvements should be defined for the business architecture, each of which realize business value and lead toward the desired strategic architecture. Each phase of architecture improvement is input to enterprise transformation.
BA is separate from the Strategic Planning service unit because the participants are different, the workload is substantial, and special skills are required. Development of the business architecture should reflect consideration of operating capacity, economies of scale, business continuity, security, outsourcing, compatibility of service units, optimization of performance, efficiency, and agility. At the same time, BA should have liaisons with at least the major department heads of the enterprise to ensure that the business architecture and transformation roadmap are realistic and feasible. In addition, Information Technology service units provide key capabilities in support of the BA efforts.
Figure 9.6 illustrates the BA responsibilities that produce and manage the business architecture. The flow of these activities is generally counter-clockwise, but it is not a strict sequence. Rather, the evolution of the business architecture is driven by business needs and opportunities that are continually changing.
With support from IT, BA should leverage technology in development of the business architecture. We discuss each of the areas of responsibility of Figure 9.6 in the following sections.
The current architecture is a depiction of the current state of the enterprise in a form that can be compared to the strategic architecture. For the conventional enterprise, this focuses on current business applications and processes. In the more agile enterprise, it reflects the service units, organization structure, and value chains. This defines the starting point for transformation.
Evolution of the architecture is influenced by current applications, the capabilities of current technology, and the skills of IT staff. Consequently, the IT organization has a significant role in these activities.
As part of this area of responsibility, BA is responsible for maintaining the current applications portfolio, service unit portfolio, and value chain models that provide insight into current enterprise operations.
The strategic architecture is a potential future state, as currently envisioned. It is an ideal state under current business circumstances. Of course, business circumstances and thus the strategic architecture will likely change before the ideal state can be achieved, but the strategic architecture provides a valuable strategic perspective for implementation of plans and for consideration of operational and tactical changes.
Initially, the strategic architecture can be developed through a rigorous service-oriented analysis, as described in Chapter 2, but more likely it will be based on an industry best-practices framework that defines service units, a logical data model, and business processes. The fully developed agile architecture reflects the desired service units, organization structure, and enterprise value chain.
The purpose of the strategic architecture is to assist in the identification and development of sharable service units that will likely persist as the enterprise continues to evolve. It also provides insight on opportunities to improve cost or performance. In the long term, changes to the strategic architecture are driven in large measure by the service unit requirements and business objectives of value chains for multiple products.
The gap analysis is a comparison of the current architecture with the strategic architecture. The purpose of the gap analysis is to identify the variances and support analysis of changes that could yield incremental business value. The gap analysis should provide a general assessment of the costs, benefits, risks, and time required to transform current operations to conform to the requirements of service units defined in the strategic architecture. Emphasis should be placed on the transformations that have high potential business value.
A significant part of the gap analysis, particularly during the early stages of transformation to SOA, is the mapping of existing business applications to service units of the future architecture. Gap analysis helps identify both service units that are embedded in other activities and service units to be considered for consolidation. This information is an important factor in planning for incremental improvements.
The gap analysis identifies potential transformations such as consolidation of capabilities that could provide direct business value through economies of scale, regulatory compliance, or improved product quality. These opportunities, along with current business priorities and the cost, risk, and delay of service unit implementation, must be weighed to determine an optimal course of action for the enterprise. Each transformation should be designed to yield business benefit. Transformations that are large and complex increase risks and may not have the anticipated business value by the time they are completed. If benefits are not realized on a regular basis, there is a risk that the commitment to SOA could be abandoned in favor of more immediate gains.
The next-generation architecture is the next incremental improvement of the current architecture that realizes meaningful business benefit and is consistent with current business priorities. The planning horizon for such an architecture might be on the order of a year or two, so it represents significant change, but a target that is not likely to change significantly before it is reached. In particular, a next-generation architecture might reflect adaptations to support a new product or line of business. The next-generation architecture should not only include service units, it should define associated organizational changes and implement incremental infrastructure capabilities that support the development, integration, and management of future service units.
BA must maintain specifications for next and future generations of the business architecture. These align to a transformation roadmap that defines phases of transformation that provide incremental value and move the enterprise in the desired direction. Each phase of the roadmap must achieve a fully operational business architecture that realizes some level of return for the enterprise or supports a new business strategy. Essentially, BA is charting the course of the enterprise to transition from the current state to a strategic future state.
The transformation roadmap may include replacement or modernization of current applications to support adaptation of service units or to reduce operating costs. Information Technology should provide capabilities to support the planning and implementation of the technical transformation. The Transformation Management service unit has responsibility for management of the business transformation, ensuring a coordinated effort.
An approach to defining service units is outlined in Chapter 2, and an approach to organization design is outlined in Chapter 7. The service units and organization design activity is focused on the specification of requirements for service unit capabilities, interfaces, and associated service levels, along with placement of service units in the organization structure.
Since the incremental transformation is based in part on expected service unit implementation, BA must consider the alternatives of adaptation of legacy systems, the acquisition of commercial software, and service unit outsourcing to determine the cost, risk, duration, and business benefit of a service unit implementation and associated transformation.
Implementation of service units also requires consideration of the organizational context in which the service units are managed. A consolidated service unit must not favor one user over another as a result of organizational history or proximity. Users of a new service unit must be able to trust that their performance will not be jeopardized because they have delegated responsibility to an unresponsive shared service unit.
Role-based access control (RBAC), discussed in Chapter 6, is an important aspect of effective management of access control. Individuals require role assignments for access to service units and information from across the enterprise. Managers cannot be expected to know or specify the technical details behind an authorization they grant. They must be able to grant authorization in terms of roles and responsibilities that are related to the purpose of the authorization. At the same time, there should be consistency in the semantics and authorization of the roles from an enterprise perspective.
Thus the specification of role semantics should be the responsibility of BA, whereas grants of authorization to individuals should be performed by their managers and details of the access authorization of elementary roles should be defined by the managers of service units responsible for protecting the resources. Changes to role definitions, assignments to participants, and authorization specifications must be controlled by appropriate specifications of grantor roles for authorization of role assignments, along with role assignment approval processes.
Enterprise rules express constraints on the operation of the enterprise based on enterprise policies and regulations (see Chapter 4). The business architecture provides a framework for the appropriate application of these rules. BA should manage the rule specifications and define the assignment of rules to service units, thus delegating responsibility for implementation of the rules to the specific service units. A rules repository must track the deployment of rules, to support verification of their application and rapid deployment of future changes.
BA requires a different discipline and skills from those required for enterprise transformation. Consequently, the role of BA in transformation is one of specification of requirements and oversight of implementation to ensure adherence to the business architecture and make adjustments as necessary. BA should be represented as part of a transformation steering committee and should participate in progress reviews.
BA must support performance assessment of service units and lines of business and support evaluation of performance of the enterprise. The Enterprise Value Chain is key to this evaluation. Operational relationships and performance metrics requirements for the service units must be maintained by BA to support understanding and analysis of value chains. Whereas individual service units are charged with responsibility for continuous improvement of their internal operations, BA must address issues that go beyond the scope of individual service units. Access to much of the data for this analysis should be supported by the Enterprise Intelligence service unit.
Service unit improvements are changes to the interface, capabilities, and levels of service of service units to more effectively meet service unit users' needs. Some of this should be done by the individual service units, but there are barriers or trade-offs that require an enterprise perspective. For example, a potential improvement may require an investment. The investment would impact the cost of the service unit, and improvement might benefit some users and not others. The change in internal processes might improve timeliness but increase costs, and timeliness might not be a concern for all users of the service unit. BA should provide an enterprise perspective to resolve these issues and support investment where appropriate. BA should also review and approve all changes to service unit interfaces so that their impact on the architecture and other service units is fully evaluated and so that business architecture records are kept current.
The Audit and Risk Assessment service unit, discussed later, is responsible for identification and evaluation of risks. BA is responsible for mitigation of risks from an architectural perspective.
Mitigation of risks resulting from practices within individual service units is primarily the responsibility of the managers of those service units. For example, in an application development service unit, developers might be allowed to informally accept new or modified requirements from service unit managers to maintain good relationships, but such practices increase the risk of cost and schedule overruns. This risk would be identified by the Audit and Risk Assessment service unit, but resolution would be the responsibility of the application development management.
On the other hand, reliance on application testing performed by the same application development team creates a risk that the test results may not be objective. BA should consider separation of an application-testing service unit from application development to eliminate this potential conflict of interest.
BA should also define requirements for event resolution services (i.e., process initiation) for each service unit that is identified as responsible for resolving disruptive events.
When new products are being planned, an expected product value chain should be modeled. This future product value chain may be a change to the value chain of a current product or line of business or a significantly new product value chain. The product value chain should be composed of current and potentially new service units. Measurements for current service units should be obtained from Enterprise Intelligence services. New service units may be variations on the use of existing capabilities or capabilities that must be acquired. BA must support the line of business manager to configure the new product value chain and perform an appropriate evaluation of the potential cost, quality, and timeliness of the new product or service unit in collaboration with other affected service units.
In addition, the new product or service may require changes in existing service units such as increased capacity, alteration of service unit interfaces, or changes in level of service—possibly resulting in cost increases. These must be evaluated in support of strategic planning. Implications to other lines of business may require evaluation to anticipate the implications to the enterprise as a whole.
Risk management is a topic of considerable concern given the increased risks associated with government regulations, potential liabilities associated with product defects, and breaches of security. SOA has the potential to increase risks through broader exposure of systems and the creation of consolidated service units that could be single points of failure. Understanding and either eliminating, mitigating, or deciding to tolerate risks is an important concern for executive management as well as the board of directors.
Audit is combined with Risk Assessment because both are concerned with identifying risks. Audits generally focus on compliance with policies, regulations, or standard practices. This group may not perform all levels of audit, such as security and recoverability of systems, but it should ensure that such audits are performed. Risk assessment goes beyond consideration of deviations from expressed requirements and considers circumstances, practices, events, trends, and enterprise design that create notable risks to the enterprise.
Audits and risk assessments are not intended to resolve risks. This separation of duties is important so that those assessing the risks do not become invested in the solutions. There are always risks. The purpose of this service unit is to ensure that executive management and the board of directors are aware of the risks and that they are at least mitigated to an acceptable level.
The following are examples of potential risks of interest:
Each risk situation should be documented along with the level of risk. Resulting corrective action or mitigation efforts should be documented and followed by reassessment to ensure that an acceptable level of risk is achieved.
Enterprise Transformation (ET) is responsible for the implementation of change. ET does not focus on the development of a new product per se, but the ET responsibility includes coordination of enterprise changes that are needed to develop and deliver a new product or otherwise improve the operation of the enterprise.
Some changes occur within the scope of individual service units and remain the responsibility of managers of those service units. Some changes occur through the collaboration of service a unit with the users of the service unit—particularly changes that would affect the service unit interfaces, costs, or levels of service of the changing service unit. Changes that have a potentially significant impact on the enterprise, involve a number of service units, or have a risk of adverse impact on customers should be reviewed by EA and managed by ET.
Transformation to an agile enterprise is a major, enterprise-level endeavor that may take a number of years. This transformation must be managed by ET through a number of projects and phases. After the transformation, the agile enterprise should anticipate continuous change to adapt to new business requirements or advances in technology. In an agile enterprise, business operations no longer exist in silos but are interconnected through a network of services. There is a continuing need for enterprise-level management of transformation.
These transformations typically involve substantial investment as well as risk for the enterprise. Plans, progress, and associated risks should be recognized as important aspects of enterprise governance.
The ET service unit should have a permanent staff responsible for program and project management as well as expertise in organizational transformation. Though transformations to address new or changed products would require collaboration with the LOB management organization, with SOA, changes for one LOB will likely affect service units used by other LOBs. At the same time, most of the work of a transformation is performed by way of other activities, particularly those of the Information Technology organization. Planning and coordination require participation by representatives of all the affected organizations. This can take various forms, such as steering committees, task forces to address particular issues, communities to contribute input, and project teams to develop and deploy solutions.
For example, suppose there is an insurance enterprise initiative to consolidate four claims-processing activities into one shared service unit to achieve economies of scale. A new service unit will be formed and positioned in the management hierarchy to be relatively independent of the organizations currently managing claims processing. The requirements for the new service unit, organizational implications, the impact on the related service units and lines of business, and the expected investment and return on investment are developed by BA in collaboration with affected organizations. Each of the four organizations that have processed claims in the past are expected to use the shared service unit.
The IT organization is required to adapt, acquire, or develop the automated business processes and application software for the consolidated service unit. A steering committee chaired by ET might be formed with representatives of the affected organizations, including IT, ET, and BA. ET is responsible for a program plan that coordinates project plans, including an application development project and an organization transformation project. ET, along with the steering committee, is responsible for requirements change control. Each project has a project team.
ET must also address, directly or indirectly, issues of training, cultural change, alignment of goals and incentives, business process management, and standards compliance. A community of affected personnel might be formed to provide a forum for discussion and resolution of personnel issues and concerns.
The role of the Standards and Technology service unit is to develop consensus on adoption of enterprise technical standards and identify preferred purchased products for both products of the enterprise and its internal operations. This includes leadership of a standards and technology committee representing diverse enterprise interests, which approves standards and preferred technology and authorizes deviations when appropriate. The purpose is to achieve consistency, economies of scale, adaptability, extensibility, manageability, and compatibility of enterprise capabilities.
For information technology, the objective is to (1) ensure compatibility for exchange of information between service units, (2) ensure compatibility of data for cross-enterprise access and integration of information, and (3) promote economies of scale in information technology service units by avoiding unnecessary technology diversity.
Standards and Technology should have an enterprise perspective on the needs of the enterprise for standards and may be the focal point for participation of the enterprise in the development of industry standards. Standards and Technology should fund initiatives to evaluate and select preferred products, using the expertise of other service units, particularly the IT organization, to perform evaluations.
Standards and technology selections for information systems should include:
Figure 9.7 depicts potential technical diversity in a service-oriented architecture. As long as the service units comply with the interface standards for the integration infrastructure and the data exchange is compatible with the enterprise logical data model or can be translated to be compatible, the service units can interoperate.
However, diversity of technology within the service units increases costs and risks. Development and support of applications using diverse technologies require that a staff be maintained with skills in each of the technologies. The internal technologies of outsourced service providers and business partners are of minimal concern. The primary concerns are legacy systems and commercial off-the-shelf (COTS) software products that must be supported by the IT organization. Each of the technologies has its own problems and risks as well as different solutions to shared risks.
In the long term, the availability of skilled personnel for legacy systems diminishes, resulting in a limited ability to resolve problems and adapt to new requirements for those legacy systems. Transformation of the applications to shared technologies should be considered in enterprise transformation planning to reduce the costs and risks of diversity.
Product acquisition and application development processes must include review by the Standards and Technology service unit, and deviations from standards should be authorized only if there is a clear business case for the deviation. This consideration must reflect the likelihood that the loss of economies of scale experienced by the IT service units will persist many years into the future, whereas the benefits of using a noncompliant solution may be only temporary, since equivalent, compliant solutions may become available in the future.
The IT services organization has primary responsibility for recommending standards and preferred products. The Standards and Technology service unit is responsible for determining which standards and product preferences are appropriate from an enterprise perspective. Representatives from the various departments of the enterprise must be involved in these decisions as well as decisions regarding deviations from the standards and preferred products.
Information Technology (IT) is responsible for management of the service units that support effective exploitation of information technology by the enterprise. We view the CIO as the top management leader of Information Technology services. The CIO is responsible for optimizing the use of information technology by the enterprise. IT services fulfill three primary roles:
IT brings technical expertise and insights to the executive staff. It provides development and support of modeling capabilities and may provide modeling expertise. At the same time it has different roles in support of each of the executive staff service units:
IT service units support automation of business service units. This includes development of automated business processes, development of supporting computer applications, implementation of commercial software, transformation of legacy systems, systems integration, problem resolution, and technical support. These are solutions owned and funded by the service units. Consequently, the basic requirements are determined by the service unit interface, the interfaces of other service units used, and capability requirements of the service unit.
IT is responsible for management and operation of the technical infrastructure, including computers, communications, and data storage. The SOA infrastructure, described in Chapter 2, includes the following:
In addition to these capabilities, IT must provide shared facilities for human communication, collaboration, and knowledge management. This includes telephone, email, teleconferencing facilities, group/community servers, and other technical capabilities that support information sharing and collaboration. EI should be viewed as the business owner of these enterprise facilities.
Within the Finance and Accounting services, SOA has a significant impact on cost accounting. The cost of every service unit must be determined both with and without the costs of services it uses. The total cost must be allocated to the units of service it provides for cost recovery in a way that achieves a reasonable representation of the actual cost of each unit of service.
Accurate cost accounting is essential for four purposes: pricing, performance evaluation, billing for services, and enterprise design. Cost determines the profit margin on products and services. Without accurate costing, it is difficult to determine an appropriate price or even whether a product or option should be continued.
Cost is an indicator of the efficiency of a service unit. Costs provide a basis for accountability of service unit managers, planning for process improvements, investment in new methods, service unit redesign, organizational changes, consolidations, outsourcing, and technology upgrades. Billing can influence users with respect to the utilization of a service, and it may influence the behavior of the service unit personnel in attempts to reduce costs.
Determination of the cost of a unit of service is not trivial, since there are both costs directly attributable to the particular service and costs that are shared. For example, the service unit incurs the cost of an employee even if the total work of providing all services does not require the employee 100% of the time. Since much of the work of a service unit may be automated, considerable employee time may be allocated to problem resolution and process improvement. From time to time, these employees may engage in projects funded by outside initiatives so that the service unit cost may go down, but then local projects may be delayed.
The costs of support services and facilities may not be associated with the delivery of specific services, but the costs are, nevertheless, a necessary part of providing the services. Consequently, the costs for each unit of service are always approximations and vary as a result of the mix of services provided during a particular period of time.
Table 9.1 illustrates a hypothetical cost allocation for the Assembly Service applied to Product 123. This example illustrates the nature of cost accounting and some of the difficulty in defining reasonable cost for individual services. The example service provides three operations, A, B, and C. Operation A is the primary service. The rows represent variations in the request options where the first row, Base, represents the product without options. The Fixed Cost column is allocation of the total fixed cost of $8,000 based on the variable costs in the column to the left. Different ways of allocating fixed costs may be more appropriate for different types of service units; for example, the option cost variances may be only associated with the cost of purchased material and not the costs incurred in performing the operations in this service unit. It may also be important to divide costs between labor and material so that sources of costs of a service and total product cost can be better understood.
Assembly Service | Operations | |||||||||
---|---|---|---|---|---|---|---|---|---|---|
Product 123 | A | B | C | All Operations Total | Fixed Cost | Net Cost | ||||
Request Attribute | Volume | Cost | Volume | Cost | Volume | Cost | Volume | Variable Cost | ||
Base | 400 | 4000 | 10 | 100 | 4 | 40 | 414 | 4140 | $7,419.35 | $11,559.35 |
V | 200 | 40 | 3 | 30 | 1 | 10 | 204 | 80 | $143.37 | $223.37 |
W | 150 | 45 | 2 | 40 | 2 | 2 | 154 | 87 | $155.91 | $242.91 |
X | 130 | 26 | 5 | 5 | 1 | 1 | 136 | 32 | $57.35 | $89.35 |
Y | 10 | 50 | 0 | 0 | 0 | 0 | 10 | 50 | $89.61 | $139.61 |
Z | 25 | 75 | 0 | 0 | 0 | 0 | 25 | 75 | $134.41 | $209.41 |
Total | 915 | 4236 | 20 | 175 | 8 | 53 | 928 | 4464 | $8000 | $12,464.00 |
This cost model represents costs for a time period—for example, a week—and the product mix that occurred during that week is indicated in the Volume column. For some analyses, it may be sufficient to consider the average cost contribution of this service based on a typical product mix. For other types of analysis, such as pricing, it is important to understand the costs of the various options as well as the typical volumes, since marketing strategy should reflect profitability of different products and product options. A robust cost analysis model would support consideration of costs and pricing under simulated variations of product mix.
Note that the total cost of a particular product configuration in a time period (i.e., based on a specific mix) is computed here by adding the associated marginal costs of all the operations that contribute to that product. In the example, this would include the product base cost and the cost for any associated options. It must also include the cost of components produced by services that are not performed in direct response to a customer order, as where orders are filled from inventories. These may be included as cost of materials in the primary production process.
Consequently, billing rates for service units are approximations based on expected workload, product or service option mix, and use of support service units. If the workload goes down, the cost per unit goes up because there are fixed costs involved. Nevertheless, users of a service unit need to be able to plan for the costs they will incur as a basis for planning and decisions that may affect their operations as well as when and how this service unit is used.
Purchasing service units are affected by SOA in two ways: (1) purchasing services are part of the product value chain for acquisition of products and services that support product development and production capabilities and (2) purchasing must manage the acquisition of outsourced business services.
Purchasing is generally viewed as a support service, only indirectly involved in the delivery of customer value. In SOA, the impact of the cost, quality, and timeliness of purchasing service units has an impact on both product development and production operations. Delays in purchases can affect the timely introduction of new products and their success in the marketplace. The cost, quality, and timeliness of production components and services have a direct impact on production value chains. Purchasing is responsible for managing the performance of suppliers and therefore is responsible for supplier performance and its impact on customer value. Similarly, purchased products and services impact other value chains.
Requirements for outsourced business services should be developed by the Business Architecture service unit, but the purchasing service unit must obtain outsourcing bids, provide information to support the outsourcing and selection decisions, and establish a contract with appropriate service requirements and levels of service specifications. Purchasing has an ongoing responsibility for managing the contract relationship, but day-to-day monitoring and problem resolution may be managed by a branch of the organization equipped to assess performance, understand the context and technology of problems, and work with service users and providers both to resolve problems and respond to changing business needs.
In both cases, a supplier should have a single point of contact within the enterprise that assesses supplier performance and takes responsibility for its impact on associated enterprise value chains.
The fundamental role of the Human Resources (HR) services does not change, but to support an agile, service-oriented enterprise, it should focus particular attention on the impact of service unit autonomy, the challenges of continuous change, and the need for appropriate incentives. This may involve the services of industrial psychologists, particularly during periods of substantial transition, but also where employee dissatisfaction or turnover are high.
Service units expose interfaces but limit exposure of implementation. They serve multiple users, so they must not adapt to the unique requirements of particular users at the expense of others. These characteristics can isolate service unit employees from the rest of the enterprise. Where employees do have direct involvement with service unit users, they must limit their activities to the terms of the service unit agreement or they will inflate the time and cost to deliver services.
For example, an information systems application development service unit must provide a solution to a customer requirement, but as the solution unfolds, the customer realizes the need for additional features. The application developer must resist the temptation to continually improve the solution or the cost of the application will increase and the project cost and delivery schedule will not be met. Instead, the application developer must enforce a change control process that involves assessment of the impact of each new requirement and establishes an agreement regarding increased cost and delayed delivery. This can interfere with interpersonal relationships and may frustrate both the application developer and the customer until they accept the necessary user/provider relationship discipline.
HR must help employees adjust to these customer/supplier relationships. This is not a concern only for service units like application development; it affects the entire enterprise as it is transformed to an SOA.
In addition to changing relationships, employees will experience significant organizational transformations and changes in job responsibilities as the enterprise implements a service-oriented architecture and as it develops a culture of continuous change as an agile enterprise. This may affect both employee motivation and skill requirements.
To become comfortable with continuous change, employees must see change as an opportunity rather than a threat. They must see changes in their jobs and the need for new skills as opportunities for growth. They need to see the enterprise as supporting growth so that people advance into new roles rather than being replaced.
HR should work with Enterprise Transformation to support employee transitions and ensure that employees have the necessary skills and understanding to perform in their changing roles.
HR should work with Business Architecture to bring consideration of employee capabilities and attitudes into the design of the enterprise. HR should give particular attention to alignment of organizational goals and development of appropriate incentives to promote employee satisfaction and productivity.
For example, collaboration should be recognized as an opportunity for personal recognition along with career growth and greater self-esteem for the participants. However, unless there is clear benefit to the participating service unit, the manager may view this as competing for resources that are needed to maintain or improve the performance of the service unit he or she manages. At a minimum, the service unit should bill for the participant's services. This offsets the disincentive and also ensures that the cost of the collaboration activities is understood and properly authorized. Additional motivation for manager support should be considered.
HR must work with top management to define appropriate incentive programs. Some choices of incentives may be affected by government regulations. Incentives should address both the motivation of individuals and the motivation of managers and should recognize the difference between contributions that improve a service unit and those that achieve improvements on a broader scale. There may be a need for different types of incentives for employees of different service unit types, and some may be specific to individual service units.
In many cases, financial incentives cannot be based on measurable criteria but must be at the discretion of managers and project leaders. The funding for incentives should be based on identifiable improvements that directly or indirectly improve customer value or reduce identified risks to acceptable levels.
In the Agile Governance Framework of Figure 9.1, Value Network Services represents the organization structure responsible for the service units that contribute to product value chains.
Product value chains are composed of the service units that contribute directly to value delivered to customers. Elements of the enterprise value chain were introduced in Chapter 2 and discussed earlier in this chapter. The product value chain segment is depicted again in Figure 9.8. A product value chain comprehends the full life cycle of a product, from concept through production and customer support. A production value chain identifies the value contributed to individual units of production for delivery to customers.
An enterprise can have multiple product value chains that may be managed in groups as lines of business based on the nature of the product and market. Product value chains and lines of business are the primary focus for strategic planning because they generally are primary sources of expenses and revenue aes ssssnd they have direct impact on customer satisfaction and enterprise profit.
Value chains are a primary basis for top management analysis, planning, improvement, and control of the operation of the enterprise. Value chain modeling is essential for understanding the contributions of service units and for understanding the sources and impact of service unit problems and potential changes. In the next chapter we will see how value chain models, along with other forms of models, represent different viewpoints of a more comprehensive model of the enterprise that supports more effective planning, decision making, and governance.