Strategic Planning for Information Systems

Strategic Planning for Information Systems: Requirements and Information Engineering Methods JAHANGIR KARIMI

JAHANGIR KARIMI received M.S. and Ph.D. degrees in management information systems from the University of Arizona, Tucson, in 1978 and 1983, respectively. Since 1983, he has been with the Department oflnformation Systems, University of Cincinnati, for a year, and the University of Colorado at Denver, where he is currently an Assistant Professor. His research interests include computer aids in the systems development process, software engineering, user interface design, information systems modeling techniques, and strategic planning. He has published in IEEE Transactions on Software Engineering, the Journal of Management Information Systems, and a number of conference proceedings. Dr. Karimi is a member ofthe Association of Computing Machinery, the Computing Society, and the Society for Information Management. ABSTRACT: The purpose of strategic planning for information systems (SPIS) is to provide a systematic process for developing a long-range plan for information systems on the basis of the organization's overall strategic plan. For an organization to perform SPIS properly, it must consider both the organizational and the technical aspects of planning. The organizational aspect emphasizes the necessity of an integrated information systems plan and its close link to the organization's objectives. The technical aspect emphasizes the necessity of planning for information systems architecture (ISA). A framework is proposed for (I) showing how the planning efforts should be coordinated in the form of a specific sequence of events for planning, modeling, and designing the ISA, (2) identifying organizational entities that should initiate and coordinate planning, modeling, and designing the ISA, (3) identifying the required processes for promoting the interactions necessary between the organizational entities for effective SPIS directions and actions, (4) distinguishing and then classifying the current SPIS methods that are useful for SPIS activities based on their characteristics as methodologies, techniques, and tools, and (5) identifying avenues for further research in terms of responding to the apparent lack that exists in the available methods. KEY WORDS AND PHRASES: Organizational strategic planning, information systems strategic planning, information engineering, information systems architecture.

Journal of Management Information Systems/Spring 1988, Vol. 4, No.4

CODvriaht © 2001. All Rights Reserved.

6

JAHANGIR KARIMI

1. Introduction THE PURPOSE OF strategic plannning for information systems (SPIS) is to provide a process for developing a long-range planning strategy for information systems within an organization on the basis of overall strategic plan [35]. The practical applications of SPIS have grown [45] to the point where there is a perceived need by practitioners to evaluate SPIS methodologies. A number of industries are successfully using information technology to obtain a competitive advantage [5]. Porter [56] suggests that SPIS can help an organization find ways to achieve a competitive advantage by using information as a strategic weapon. Resources can be allocated to the most important (current or future) information systems development projects. SPIS can help a firm to (1) become the low-cost producer for a given product or service, (2) define and service a specially defined market niche, and (3) differentiate its product offering from that of its competitors [12, 29]. In a recent study on key issues in information systems management by Brancheau and Wetherbe [10], improving SPIS was ranked first in a list of the ten most critical uses facing information systems (IS) executives. The study indicates the fact that it has become imperative for the IS executives to align SPIS with their company's strategic business plan. Using information systems for competitive advantage was ranked second in importance. Although more attention has been given in literature to specific issues that should be considered while conducting SPIS activities, relatively limited attention has been paid to the complex issues of the SPIS processes and how they can be executed [9]. This represents a major shortfall, both in practice and for the researchers. This paper offers insights on ways (1) to conduct SPIS activities and (2) to build an effective link between an organization's strategic plan and SPIS.

2. An Overview of Previous Research Findings RESEARCH ON management information systems (MIS) planning in the past fifteen years has provided a number of explanations about different aspects of SPIS. The findings can be summarized as falling into four major categories: (1) research focusing on issues that require attention for effective SPIS. A number of organizational factors have been identified that affect the success or failure of SPIS activities [3, 4, 42, 58, 61, 62]. These factors are summarized in Table 1; (2) research focusing on the importance of top management involvement in developing effective SPIS and ways for such an involvement to take place [17,18,64]; (3) research focusing on organizational behavior models [19, 47,59,60] which suggests different planning processes deserving further investigation; (4) research providing models and frameworks for the process of formulating SPIS based on the organizational strategic plan [8, 34, 20]. Choosing proper planning methodology based on the strategic role of the IS [37] and the multiplicity of the levels in the IS planning process [66] is also categorized at this level.

---------------"--'--_._-,------, Copyright © 2001 , All Rights Reserved

STRATEGIC PLANNING FOR INFORMAnON SYSTEMS

7

From the four major research areas on IS planning, the issue of how to operationalize an effective planning process, number (4) above, needs more attention. This view is also shared by a recent study on directions in information technology planning for practice and research in the 1990s by Boynton and Zmud [9]. Although the research in the last category has provided practical guidance for (I) recognizing the nature of the IS planning problems [66] and (2) selecting the appropriate stage of planning [8, 34], it has failed to provide specific guidelines on how planning efforts should be coordinated in the form of a specific sequence of events to plan, model, design, and implement an information systems architecture (ISA). Development of ISA was ranked eighth in the list of the ten most critical issues facing top IS executives. It has become critical as systems development decentralizes to business divisions and departments and as we approach what has been described as the "information economy" in the 1990s by Boynton and Zmud [9]. In the remainder of this paper, section 3 describes ISA and the necessity for its development from a technical and a managerial perspective. Section 4 presents an overall analytical framework for describing the nature and the sequence of activities for formulating SPIS and for building ISA. Section 5 describes (l) four organizational entities and their "information engineering roles, " which should initiate the different SPIS strategies, and (2) the nature of "align" and "exploit" flows between these roles which are critical for proper SPIS. Section 6 describes the linkage between the oganizational strategic plan and SPIS. The role of an "information engineering workbench" and the alternative "information engineering methods" useful for planning and development activities are described next. Section 7 presents conclusions and suggestions for future research.

3. Information Systems Architecture (ISA) TRADITIONAL MANAGEMENT STRATEGY for the automation process has been a bottom-up (local) approach in which various functional areas were automated on an application-by-application basis without consideration for integration and optimization at the organizational level. As a result, organizations discovered that these application systems were becoming increasingly interdependent, incompatible, redundant, and, in many cases, incomprehensible. Senior executives need to regain effective control of automation strategies in the face of rapidly growing information technology [4, 10,22,68]. They seek a proper top-down control of different automation strategies. Such a control requires a global (organizational) approach to design and management of data processing systems, since there are many important needs that must be met globally across the organization. As is shown in Table I, success in SPIS activities largely depends on, among other factors, the degree of integration in IS planning. Integration in IS planning is only achieved by planning based on a business-based modeling approach to systems


8

JAHANGIR KARIMI

Table 1 Organizational Factors Affecting the Success or Failure of Strategic Planning for Information Systems (SPIS) Activities • Degree of integration in information systems (IS) planning • Is planning horizon • Stage of the IS maturity • Status of the IS manager • Physical proximity of IS manager • Organization's resistance to planning • Organization's planning budget • Economic environment of organization • Volatility of organization's business system • Organization's overall complexity • Risk tolerance of organization

development [28]. To build a business model of an organization and to translate that model into integrated systems requires a change in systems development approach. At the heart of business-based systems is a model of the information requirement of the business (information systems architecture) composed of business entities, data, and processes. The information systems architecture (ISA) relates the organizational processes that must be performed to data classes that are required by those processes. ISA represents the information flow requirements of the entire organization. The logical modeling of the ISA, at minimum, is to (1) include the needs of all users of data processing services and (2) be done with minimum redundancy in data and process modeling across the organization. Several benefits are expected from the development of a global ISA as opposed to the traditional bottom-up, application-by-application systems development process. First, global ISA provides a framework that allows one development effort to build on another. Second, the ISA layout provides a blueprint for development strategy, thereby making a complex development effort more manageable by building a piece at a time and protecting the investment. Third, the full benefits of the global ISA can be realized without having to wait for ISA to be fully implemented. In fact, once the blueprint for ISA is completed, it can support rapid systems development, formulization of global perspectives that represent the business of the organization as a whole, and end-user automony [27]. In the following an analytical framework is proposed for SPIS. The proposed framework incorporates the strategy set transformation model proposed by King [34], the three-stage MIS planning model by Bowman, Davis, and Wetherbe [8], and the IS strategic postures model by Kotteman and Konsynski [37]. It also extends the existing models for SPIS by (1) showing how the SPIS life cycle relates to and extends the traditional systems development life cycle, (2) showing how planning efforts should be coordinated in the form of a specific sequence of events for planning, modeling, and designing ISA, (3) identifying organizational entities that should initiate and coordinate the planning, modeling, and designing of ISA, (4) identifying the required processes for promoting the interaction necessary between organiza-

Copyright © 2001. All Rights Reserved


9

tional entities for effective SPIS directions and actions, (5) distinguishing and then classifying the current SPIS methods that are useful for SPIS activities based on their characteristics as methodologies, techniques, and tools, and (6) identifying avenues for further research in terms of responding to the apparent lack that exists in the available methods.

4. Analytical Framework consists of two dimensions, breadth and depth. These dimensions form the axes of a graph (called the DB-space) which is then used for classifying current SPIS methods (defined in section 4.2).

THE ANALYTICAL FRAMEWORK

4.1. The Breadth Dimension The breadth dimension of the analytical framework is an extension of traditional frameworks for the system life cycle in which the overall mission and nature of the organization are included. It is composed of the following phases: -organizational analysis -strategy-to-requirement transformation -logical systems design -logical-to-physical transformation -systems implementation The traditional system development life cycle starts with looking at the feasibility (i.e., cost and benefits) of developing an application. However, the strategic implications of all current and future applications to the organization and to the global ISA are usually not considered. Hence, the breadth dimension deals first with deriving the information systems strategies from the organizational strategies and then deals with the implementation details of information systems projects. Figure I shows the five phases of the breadth dimension. The rectangles denote activities that are performed. The circles show products from the preceding activity that are used in the subsequent activity. The breadth dimension thus describes both what is being done and what results. The combination of activity and its product is referred to as a phase.

4.1.1. PHASE ONE-Organizational Analysis The purpose of phase one is to examine the mission and nature of the organization and its environment and to translate the organizational mission statement to a concise, accurate, and formal statement of organizational objectives and strategies. Organizational analysis is unfortunately the weakest phase in the breadth dimension [68]. As stated by Zachman [72], organizational analysis is in its formative stages; however, every business that continues to grow and evolve has to employ


ORGANIZATIONAL ANALYSIS

r--

o

STRATEGY TO REQUIREMENT

LOGICAL DESIGN

TRANSFORMATION

LOGICAL TO PHYSICAL TRANSFORMATION

o

o

~

cO·

;r

@ N

o

T

~

SYSTEMS IMPLEMENTATION

Figure 1. Phases of the Breadth Dimension Note:

ISA

= information systems architecture.

• • •

SYSTEMS IMPLEMENTA TION

• • •


11

organizational analysis in some form. Methods for organizational analysis ideally should meet two criteria [35]: 1. relate the organizational strategies to the information systems strategies and the ISA requirements so that a change in organizational strategy would be supported by the ISA; and 2. incorporate the notion of IS resources as a strategic resource (or "competitive weapon") in the organizational strategic plan in order to identify opportunities to use those resources. Therefore, organizational analysis needs to connect organizational strategies to information systems strategies. 4.1.2. PHASE TWO-Strategy-to-Requirement Transformation Organizational strategies are used to derive information systems strategies and information systems requirements for ISA modeling. As it is presented later in Figure 2, and supported by previous research [13, 57], organizational strategies could also be led by technology evolution and resources. Planning for ISA should specifically include the implications of the business objectives and the organization strategic plan on the strategic directions setting of information systems technology. It requires primarily business-oriented people who understand the information requirements of an organization (see section 5). Modeling for ISA consists of the following activities: -global entity relation modeling -conceptual data modeling -process modeling -data/process modeling integration

4.1.3. PHASE THREE-Logical Systems Design The purpose of phase three is to design data, application, and geographic architectures using the ISA requirements. The planning requirement at each level is described below. Data architecture represents a blueprint of the databases that should be designed from an organizational standpoint. Planning for data architecture includes implementation priorities, the amount of resources that must be invested, and the probable returns on investment for each database project. Application architecture defines the application areas necessary to support the ISA and the relationships between those application areas. It also reflects where applications are currently implemented. Other parts of the application architecture include interrelationships between applications, the way each application supports the organization's strategies, and the methods of data sharing between these applications. The geographic architecture, in turn, describes where applications will run, where databases will be located, and what communication links are needed between the locations.


12

JAHANGIR KARIMI

Planning at this level includes details about accessibility of data and computing resources by each organizational unit; hardware and software alternatives; the different architecture design approaches; and the risks, benefits, and preliminary cost estimates for each alternative and for each architecture. Note the change in the nature of planning-from being mostly a managerial and business-oriented plan in the strategy-to-requirement transformation phase to a technically oriented one in this phase. There are three foundations for the design of the global ISA [26]. First, the global design of the ISA should be based on subject orientation of data. Second, there should be a clear separation between the primitive data (i.e., detailed, structured usage, non-redundant, used to run the organization) and the derived data (i.e., summarized, unstructured usage, redundant, used to manage the organization). Third, there should be a disciplined flow of data between the different modes of operation (i.e., production and decision support).

4.1.4. PHASE FOUR-Logical-to-Physical Transformation Phase four consists of taking the general design of the data, application, and geographic architectures, decomposing them into subsystems (or portfolios of applications), deciding on the detailed design of each subsystem, and setting up the priorities and making commitments to implementation. The result is the detailed systems design implementation plan that describes the steps to implement a specific subsystem or application. In general, the product of this phase includes the following: -schema and subschema specification of databases -software specification -specification of systems components

4.1.5. PHASE FIVE-System Implementation Phase five occurs many times, once for each system defined in the detailed systems design implementation plan. The result is an operational subsystem that supports a business function of the organization. This phase is similar to the traditional system life cycle that is initiated with a feasibility study of an application. The exception is that the system implementation starts with a notion of the ISA, guiding the implementation in its integration with other subsystems. A complete system implementation plan should include three major components: the project scope, the development strategy, and the organizational impact. The cost and time frame of the projects, as well as the expected benefits from a successfully completed project, need to be defined. Risk assessment needs to be done on the projects, and the projects' control systems need to be adjusted to minimize the risk that a particular project poses to the organization.

Copyright © 2001. All Rights Reserved

STRATEGIC PLANNING FOR INFORMA nON SYSTEMS

13

4.2. The Depth Dimension To build the business model of an organization and to translate that model into the ISA requires a set of information engineering tools, techniques, and methodologies. Information engineering (IE) is defined as the process of translating a corporate focus (a strategic plan, expressed as an organization mission statement) into the logical design of the ISA. The depth dimension consists of three levels: methodology, technique, and tool. These are defined below: The term methodology is defined as "the analysis of the principles ... of inquiry in a particular field" (see Webster's New World Dictionary, 3rd College edition). This definition emphasizes the conceptual basis for performing IE. That is, it highlights questions like the following: -What factors are important? -How do these factors relate to or impact on one another? -What management actions should be taken? -What are the desirable outcomes? The term technique is defined as "a procedure for accomplishing a desired outcome." This definition emphasizes the procedural details for performing IE. In particular, a technique would specify the steps used in performing IE, along with the necessary inputs and results from each step. The term tool is defined as "an instrument for performing a procedure." In particular, a tool is some tangible aid (e.g., analysis form or computer-assisted software program) used in performing some aspect of IE. For the purpose of this discussion, the term IE method is defined as a generic term that refers to individual or combinations of methodologies, techniques, and tools used for IE. A graph was constructed using the breadth dimension as the horizontal axis and the depth dimension as the vertical axis. This graph is shown in Figure 2. In section 5, the two dimensions of breadth and depth together form an analytical framework that can aid in identifying different organizational entities that should initiate and coordinate the planning, modeling, and design of the ISA. Later, in section 6.2, we show how the analytical framework can also aid in identification and classification of IE methods to support those information engineering roles.

5. Information Engineering Roles As SYSTEMS DEVELOPMENT decentralizes to business divisions and departments, the ISA becomes a critical issue because of its global scope, its unstructured nature, and the lack of expertise in planning for the ISA. Recent studies [9, 10] point to the importance of identifying the organizational entities that should initiate the different SPIS activities throughout an organization. However, they fail to provide any specific guidelines. The organizational entities and the nature of their interactions for effective SPIS are defined below. As it is shown in Figure 2, an ongoing process should be performed by knowl-


CONCEPTUAL PLANNER

METHODOLOGY

Orgonization Strategic Planning

"'