Abstract In today's complex environment, organizations must possess the best possible management talent available. Such precision is less crucial in times of economic prosperity, but obviously still desired. The traditional approaches in the selection of individuals for promotion produce variable results. Successful decision-making is positively correlated with increased position level. Thus the major difficulties have occurred in the first-line to middle range of management. In order to facilitate the selection process, one of the methods has been the implementation of assessment centres. The present article reviews the process, indicating the strengths and weaknesses. The most successful models are those that are in-house programs specifically tailored to the organization's needs, that have a dual role of assessment and development, that are based upon a thorough analysis of relevant job dimensions, and have a major reliance on situational techniques of evaluation. Introduction The escalating growth in industry coupled with the technological imperative has created major problems in manpower selection, availability, and development. As a result organizations have and are continuing to develop procedures which permit early identification of management talent. The first industrial centre was developed in 1956 at AT&T (Bray, Campbell & Grant, 1974), but the concept was first applied during World War I for rapid and effective selection of large numbers of officers and agents. The original AT&T model was a longitudinal research study of college hires, (The Management Progress Study), but quickly expanded into affiliated companies for identification of first line foremen and assessment of higher levels of management. Adoption by such companies as Standard Oil of Ohio, IBM, Sears, General Electric and J.C. Penney followed. Consulting firms began to offer assessment services as well as assistance in the establishment of "in-house " centres. The four salient aspects present in the initial AT&T program are still applicable today, namely: assessment in groups, assessment by groups, use of multiple measurement techniques, and a special orientation to management. Norton (1977) states that the literature on traditional methods for predicting managerial success which reflect maximum validity correspond to merely average validity of assessment centres. Obviously such conclusions will add to the proliferation in usage, but despite such positivism, there must be evaluation of the components in order to design the most effective program for the particular organization. Even more critical is the need for a total commitment to organizational development. Assessment is only one activity, and can be implemented through various perspectives. It can reflect a participatory approach or an authoritative model. Assessed individuals should be and often will prove to be psychologically receptive and thus open to development (Golembiewski, 1979). If the centre encompasses both evaluation and training in an organizational development framework, predictive validity should be higher. The major areas of review will be purpose, generic aspects, techniques, approaches, and research validation. Each of these dimensions will be presented individually, followed by an integrative approach. The objective is to identify the major components associated with assessment centres, and also to present the validity of the criteria within various contexts as they predict future managerial success.