Obesity and smoking remain major public health problems. Because of their widespread prevalence, large scale community-wide programs are needed to have an impact on public health. The purpose of this study was to determine the magnitude of response to four variations in recruitment and demographic characteristics of people enrolling in self-help programs for weight control and smoking cessation. A total of 76,900 households from three communities were offered correspondence-based programs in weight control and smoking cessation. Rates of enrollment ranged from 0.1 to 5% for smoking and from 0.3 to 12% for weight after adjusting for estimated prevalence of smoking and obesity. Recruitment rates were positively related to amount of money spent on recruitment per household. Sending sign-up packages only to those returning interest cards was less costly than sending sign-up packages to everyone. Those who received announcements of the program prior to receiving sign-up packages enrolled at a higher rate than those not receiving announcements. Mass media campaigns did not clearly increase enrollment. A flat fee/materials only option was about 10 times more popular than an incentive option which required a deposit and was refunded proportional to success. Enrollees were somewhat older, better educated and less overweight than those enrolling in traditional programs. Average recruitment costs per enrollee ranged from +14 to +27 in the three communities. Total enrollment was 1990 for weight and 298 for smoking across the three communities in what may be the largest direct mail recruitment effort and intervention program of its kind yet attempted.