Genetic Modeling In Direct Marketing

Data analysts in direct marketing seek to build models that maximize expected response and profit from solicitations, using various techniques in their toolkits. The standard techniques include the statistical methods of classical discriminant analysis, and logistic and ordinary regression. A recent addition in the toolkit is the artificial intelligence (AI) method of neural networks. The purpose of this article is to present the newest entry in the toolkit, the GenIQ Model©, a hybrid AI-statistics method.

I first provide a background on the concept of optimization, because optimization techniques provide the estimation of all models. I discuss the basics of genetic modelling, an AI optimization approach, which is the "engine" for the GenIQ Model. Then, I present the GenIQ Model as a technique that explicitly addresses the direct marketing objectives of maximizing expected response and profit from solicitations. Lastly, I discuss real case studies to demonstrate the potential of the GenIQ Model.