School districts have begun examining the feasibility of, and in some cases are developing and implementing automated systems for, managing and evaluating instructional programs. This paper describes and analyzes the issues and problems that emerged over the course of three projects--a large suburban school in the West, a consortium of five small schools in the East, and an Army division in the Northwest--which were directed toward designing, developing, and implementing computer systems for managing and evaluating educational programs. Discussion is organized in four parts: (1) an overview; (2) a description of the approach and concepts; (3) insights into project successes and failures; and (4) recommendations for future directions. The overview covers the common objectives of systematizing and facilitating access to information for program management, supporting program improvement efforts, and providing evaluative information to assess program outcomes and effectiveness. A general systems model is used to describe the approach and concepts and to illustrate the organizational complexity of the projects. Potential conflicts between the goals of groups within the system and problems associated with the hierarchical structure of authority and responsibility are noted. Insights into the utility of a general systems approach are discussed in the context of synergism, system structure, optimization, stability, and equifinality. It is recommended that the development process be structured to (1) facilitate the functioning of relevant parties so that the overall goal is achieved; (2) contribute to communication and feedback among all parties; (3) consider the needs of all parties; and (4) gather information on system functioning for revision and improvement. (DJR)