Discovering interesting and previously unknown information - also known as feature extraction - has been a frustrating exercise. Many methods of feature extraction exist, but none of them is perfect. Furthermore, none use a criterion that addresses the specific needs of direct marketing models. The purpose of this article is to present a new methodology that uses the high performance computing of evolutionary computation (EC) for feature extraction. Most significantly, the specific paradigm of EC employed is used to address the specific requirements of direct marketers. Two case studies are presented using response and profit direct marketing models.