A knowledge-based system for advertising design

In recent years, artificial intelligence research has provided new tools and techniques for marketing model builders. These tools, when combined with problem-solving knowledge from a specific domain, can be used to create expert systems. This methodology is most applicable in semistructured problem domains where the key relationships are logical rather than numerical and problem-solving knowledge is incomplete. One such problem in marketing is advertising design. In this paper, the authors describe the application of expert system techniques to the development of ADCAD, a system designed to assist advertisers of consumer products with the formulation of advertising objectives, copy strategy, and the selection of communication approaches. The paper highlights the assembly of ADCAD’s knowledge base of advertising design heuristics from multiple sources of published research and practitioner expertise and describes its structure and content. The authors discuss the procedures and obstacles associated with building implementing, and validating such an expert system. Their experiences with ADCAD suggest that knowledge-based systems have significant potential for the consolidation and integration of marketing knowledge as interactive input for decision makers. The authors conclude by noting several avenues for the future development of advertising expert systems.