Outcomes of a project designed to develop the knowledge base necessary for creating an expert system that would help nonprofit organizations select fund-raising software are presented in this paper. When the system is completed, its components will ask the user for information that will assist in determining the organization's administrative needs and will then make recommendations for the software most appropriate for meeting that organization's fund-raising needs. The project involved two separate studies, called respectively, the "Laboratory Study" and the "After Study." The first phase of the laboratory study developed a model query system based on the developer's personal knowledge. The second phase involved a literature review focused on software selection and fundraising software. During the after study, personal interviews were conducted with a total of three representatives: one each from a small, medium, and large organization. Telephone interviews were conducted with nine representatives: three each from a small, medium, and large organization. The after study expert system that was designed following field research incorporated the information gathered in the previous stages but added several more software features for incorporation into the query component. In summary, the after study model differed from the laboratory model in the following ways: it has more software features; the size of the organization is not considered in the queries; references have been placed in the software database; and organizational objectives are determined before using the query system. Appendices contain the laboratory induction table and expert system; a list of United Way agencies and the associated annual budget; a list of questionnaire features; interview questions; a review of fund-raising software and their features; records of the field structure of the software features database and software description database; and the expert system code and dialog. A list giving definitions of terms is also included. (Contains 13 references.) (LMI)