Information Architecture: Sharing the Shareable Resource Copyright 1991 CAUSE From _CAUSE/EFFECT_ Volume 14, Number 2, Summer 1991. Permission to copy or disseminate all or part of this material is granted provided that the copies are not made or distributed for commercial advantage, the CAUSE copyright and its dateappear, and notice is given that copying is by permission of CAUSE, the association for managing and using information resources in higher education. To disseminate otherwise, or to republish, requires written permission. For further information, contact CAUSE, 4840 Pearl East Circle, Suite 302E, Boulder, CO 80301, 303-449-4430, e-mail info@CAUSE.colorado.edu INFORMATION ARCHITECTURE: SHARING THE SHAREABLE RESOURCE by Douglas R. Vogel and James C. Wetherbe ************************************************************************ Douglas R. Vogel, Assistant Professor of MIS at the University of Arizona, has been involved with computer systems in various capacities for over twenty years. His current research interests bridge the business and academic communities in addressing questions of the impact of MIS on interpersonal communication, group problem solving, and organizational productivity. He has been particularly active in the development, facilitation, and evaluation of University of Arizona GroupSystems, an integrated set of software tools to assist groups in maximizing meeting productivity. Dr. Vogel received his Ph.D. in MIS from the University of Minnesota. James C. Wetherbe is Professor and Director of the Management Information Systems Research Center at the Carlson School of Management, University of Minnesota. Author of twelve books, including Executives' Guide to Computer-Based Information Systems, The Management of Information Systems, and Systems Analysis and Design, he is publisher of MIS Quarterly and editor of Data Base. Internationally known as a dynamic speaker and leading authority on the use of computers and information systems to improve organizational performance and competitiveness, he presented a general session address at CAUSE90 on "Competing with Computing." ************************************************************************ ABSTRACT: Developing management information systems to effectively accommodate and integrate the complex information needs of a college or university is increasingly relevant to successful planning, administration, and operation. This article describes a methodology for developing a long- range information architecture. Illustrations based on applying the methodology at the University of Minnesota are used throughout the article. Computer-based information systems have been in use at colleges and universities since the 1950s. As in most organizations, over the years information systems have proliferated to respond to diverse and changing priorities. Though these systems solve specific problems, they often fail to systematically consider overall organizational information requirements. A problem with this approach to developing information systems is that it tends to generate fragmented, piecemeal systems. Such systems are incompatible with each other and are difficult to integrate to support information requirements for decision-making. Data from computer printouts are processed painstakingly by hand, often by high-level staff members, to make them usable for decision support. The practice is widespread; the resulting level of frustration is immense. To resolve this situation, an organization must make the investment of concurrently studying the overall information requirements and developing an information profile or architecture to provide a "road map" for developing the various information systems that must be tied together to coordinate the management of different organizational efforts. Integrated systems do not in themselves guarantee information to support management decision-making. This can only be accomplished if a careful analysis is made of the decisions managers must make and of the information required to make those decisions. The key to success is to start by focusing on the decisions to be made and on the information needed to make those decisions effective and efficient, rather than to start from the decision-maker's frustrations of the moment (the source of many projects in information systems improvement). The purpose of this article is to describe a methodology and its application for developing a strategic, long-range organizational information architecture. This methodology calls for the development of a plan to ensure the best possible use of information resources through identifying key categories of information that can best enhance decision-making and operational productivity. To convey the manner in which the plan is formulated, illustrations based on applying the methodology at the University of Minnesota are used throughout the article. The scope of this article is limited to providing a conceptual framework for strategically directing information systems. It provides the information necessary to select categories of information that, if improved, can enhance decision-making and operations in a college or university environment. The article does not provide detailed specifications for specific projects, nor does it outline specific computer hardware or software requirements necessary to improve a category of information. However, it does define the courses of action necessary to achieve such project definitions. The first section provides the analytical framework used to conduct the information requirements analysis and develop information requirements plans, while the second section describes the stages of the planning model used to establish the information architecture. Benefits of the methodology are highlighted in a brief summary. ANALYTICAL FRAMEWORK Planning for computing and information systems has been characterized as one of the most difficult planning tasks. Computing and information systems have been integrated into virtually every dimension of modern organizations in both explicit and obscure ways. Indeed, it has been increasingly difficult to define exactly where information systems stop and other organizational activities begin. Managers on the average spend 50 percent of their time looking for information they need to do their jobs.[1] Accordingly, there is great potential to improve the productivity of managers and administrators by providing them with more effective and efficient information. However, improving information for managers cannot be accomplished by focusing on one manager. Managerial responsibilities and organizational structures change and, with these changes, information requirements change. Strategy for Information Requirements Planning An information requirements planning methodology must focus on the organization as a total entity in a way that is independent of organizational structure and personnel.[2] Such a focus allows the information requirements plan to survive change in these areas. This can be accomplished by using a planning methodology that: * keys in on the goals and objectives of the organization; * addresses the needs of all levels of management of the organization; * provides consistency of information throughout the organization; * identifies and analyzes the information requirements of the underlying organizational decision-making process (e.g., the institution's budget process), rather than just analyzing the information requirements of departments (e.g., finance and accounting office); * defines the high potential payoff areas for improving information; and * allows for implementing short-range subsystems of information within a long-range information architecture. Figure 1 provides a basic model of such an information systems planning methodology.[3] The first six stages of the model focus on the processes necessary to determine a detailed information architecture, leading to identification of the information categories that need improvement, and will have high payoffs, for supporting decision-making and operations (including transaction processing). The last three stages of the planning model (stages seven to nine) relate to the definition, priority setting, and implementation of specific information systems projects. [FIGURE 1 NOT AVAILABLE IN ASCII TEXT VERSION] In contrast to this proposed methodology, the usual approach to information systems planning is to start with the sixth (or even seventh) stage each time a system generates a new frustration. The approach recommended here has been shown in research to produce a long- lasting and more useful information system and, in fact, was successfully employed at the University of Minnesota in 1985 by an information systems task force appointed by the University President to improve information integration and identifyand establish priorities for important applications. INFORMATION REQUIREMENTS PLANNING As depicted in Figure 1, the first stage of the planning model is to identify and define underlying organizational processes. An organizational process is a fundamental activity that is necessary for the operation of the organization. Certain processes are common to all organizations (e.g., financial processes) while others are unique to certain industries (e.g., checking accounts for banks, recruiting/admissions for colleges and universities). Organizational processes transcend functional boundaries. For example, even though recruiting/admissions is an organizational function as well as a process, when it is discussed as an organizational process, it transcends the admissions office, i.e., college deans and registrars are involved. The University of Minnesota is one of the largest public institutions of higher learning in the United States. It is a single comprehensive system with five separate campuses, 4,500 full-time faculty, and 58,000 students. The underlying organizational processes for the University of Minnesota were identified by the information systems task force as follows: * Instruction/Curriculum * Research/Creative Activity * Service (Community) * Strategic Planning/Institutional Research * Recruiting/Admissions * Registration/Advising/Monitoring * Placement/Career Development * Academic Appointments * Personnel Administration * External Relations * Financial Planning/Management Reporting * Facilities Management * Computing/Information Services * Auxiliary Services * Legal/Compliance/Law Enforcement Once the organizational processes are defined, the next stage is to relate specific managers and administrators to those processes. Relationships to organizational processes are defined as major involvement with decision-making responsibility. The relationship of specific administrators of the University of Minnesota with organizational processes was determined by analysis and task force review. The relationships are depicted in Table 1. Group Sessions Too often, information systems overload managers with information that does not help them make decisions. This happens because at some point the theory more information means better decisions begins to dominate information systems development. Unfortunately, as the amount of information generated increases, its usefulness for decision-making decreases.[4] Since information is used for decision-making, the systems development focus should be first on management decisions, and then on the information required to make those decisions. The third stage of the planning model provides such a focus by working toward defining the organization's information requirements in terms of management decisions. This is accomplished by holding a series of group sessions to which managers are invited based on their decision- making relationship to organizational processes. At the University of Minnesota, these sessions were coordinated by the information systems task force. The first step of the group session is to get the members to agree upon a statement of purpose for the organizational process under consideration. For example, the purpose for the recruiting/admissions process could be stated as "those activities related to the identification of prospective students, their recruitment, completion of the admission process, orientation, and initial registration." After the statement of purpose, the next step is to elicit information requirements using a structured approach based on responses to questions asked by the study team. The specific questions and the way they are asked is a key issue. Research at the University of Minnesota's MIS Research Center, in which several hundred managers in different organizations were interviewed, has found that the obvious question--what information do you need?--is the wrong question. It is the less obvious but properly asked indirect questions that do the job. For example, a good series of questions to ask is: 1. What are the major problems that this organizational process has in accomplishing its purpose? 2. How could they best be solved? 3. Can better information help? The first two questions set the stage for the third question, which reveals information requirements. Questions asked in the group sessions are derived from three different but complementary approaches to establishing information requirements: Business Systems Planning (BSP) developed at IBM,5 Critical Success Factors (CSF) developed at MIT,6 and Ends/Means Analysis.[7] Each approach provides a different view of information requirements.[8] (See the sidebar on the next page for examples of specific questions asked during the group session interviews.) A separate group session is conducted for each process, and each group session's interviews result in the definition of a variety of information requirements for the subsystems that support the process. Input is also captured on the value of having certain information items and how available those items are. Information Analysis Stages The study teams use information derived from the group sessions to accomplish the next three stages of the planning model--define major categories of information needed by organizational processes, relate those categories to organizational subsystems, and define potential payoff areas for information improvement. By coalescing the information items identified by individual study teams into generic information categories, an overall profile of information categories needed is developed. The categories are organized under group headings to enhance assimilation and aid in development of the overall architecture. At the University of Minnesota, for example, the task force aggregated the defined information categories under the following group headings: students, policies, university employees, external communications, programs, budgets, plans, facilities, and schedules. Mapping the information categories against organizational processes results in the representation of an overall information profile, to which the study team is then able to add values based on the information collected about importance and availability. Thus potential payoff areas for improving information are identified. During the group sessions, managers assess the value of having a category of information as high, medium, or low, with these assessments quantified as 3, 2, and 1, respectively; the highest score indicated for a category of information while studying a process is used as the final score. Managers also evaluate how available the category of information is, with 1 = high availability, 2 = medium availability, and 3 = low availability. By tabulating the scores for each information category across the matrix of organizational processes, a composite value score can be computed for each category of information. The composite value of a category of information is computed by totaling scores for each category of information by organizational process (with these scores derived by multiplying the availability rating by the importance rating). The group composite scores are computed by totaling composite values of information categories under each group heading. The results of the scoring process for the University of Minnesota example are presented in Table 2. The composite scores show the areas of highest payoff for improving information in the University. The group heading students, with information categories prospective, entering, returning, graduating, and alumni, constituted the highest payoff for information systems development of a cohesive group based upon a combination of value and lack of availability within organizational processes, and generalized utility across those processes. Thus, it represented a logical starting point for more detailed information systems analysis and subsequent development. Elicitation of Detailed Requirements The method used to derive the information categories for organizational processes can be used as a starting point to elicit more detailed data requirements for functional subsystems. In this next step, the flavor of the structured techniques used in eliciting the high-level information requirements provides a process through which the next lower layer of information categories can be derived through the use of personnel directly associated with the operation of organizational process functions. Statements of purpose can then be developed within functional areas for designated information categories. The matrix of information categories by organizational processes developed in the fifth stage, with associated attributes identified in the group sessions, serves to identify the specific information categories and functional areas pertinent to the development of more detailed, quantitatively oriented information items. This process can proceed heuristically until information categories are derived with quantitative characteristics capable of being represented in more traditional data models for subsequent computer-based implementation, in preparation for the last three stages of the information requirements planning model. A layering effect is achieved that not only involves the appropriate personnel at the proper level of abstraction of information, but further serves to bond subsystems meeting the needs of individual organizational units into a cohesive whole. Thus, the information architecture is useful as a framework within which more detailed designs will fit and integrate to form a comprehensive system that attains efficiency while preserving effectiveness. SUMMARY According to Dr. Edward Foster, Vice President for Academic Affairs at the University of Minnesota, the information architecture effort conducted at Minnesota provided the University with a useful framework to identify, select, and prioritize information systems projects, while ensuring their integration with other systems when appropriate and possible. The architecture has been used for these purposes for the past five years; in particular, student and financial systems have been designed and implemented using the architecture. The methodology for developing a long-range information architecture described and illustrated in this article has many benefits. * High-payoff information categories are identified to help in setting priorities for subsequent information systems development. * Problems of building separate, redundant information systems for different organizational processes are avoided. * Enhanced insight and organizational learning is facilitated as the group process reveals previously unrecognized information relationships. Overall, the approach described is increasingly relevant to the development of management information systems that effectively accommodate and integrate complex information needs on college and university campuses. ************************************************************************ BSP, CSF, and Ends/Means Analysis Business Systems Planning (Problems and Decisions) 1. What are the major problems encountered in accomplishing the purposes of this process/subsystem? a. What are good solutions to those problems? b. How can information play a role in any of those solutions? 2. What are the major decisions in managing this process/subsystem? a. What information is needed to make these decisions? Critical Success Factors 1. What are the critical success factors of this process/subsystem? (Note: most executives have four to eight of these.) 2. What information is needed to ensure critical success factors are under control? Ends/Means Analysis 1. What makes goods or services provided by this process/subsystem effective to users? a. What information is needed to ensure that it is being effective at pro- viding those tools or services? 2. How do you define efficiency in providing goods or services by this process/subsystem? a. What information is needed to evaluate its efficiency? ======================================================================== Footnotes 1 H. Mintzberg, The Nature of Managerial Work (Englewood Cliffs, N.J.: Prentice-Hall, 1973). 2 G. Dickson and J. Wetherbe, The Management of Information Systems (New York: McGraw-Hill, 1984). 3 J. Wetherbe and G. Davis, "Developing a Long-Range Information Architecture," a paper presented at the National Computer Conference, 1983; and J. Wetherbe, Systems Analysis and Design, Third Edition (St. Paul, Minn.: West Publishing, 1988). 4 R. Ackoff, "Management Misinformation Systems," Management Science 14:4 (December 1967): 147-156. ========================================================================