This document discusses discrete choice experiments for designing marketing surveys. It describes how discrete choice experiments work, including using choice sets with profiles that combine different attribute levels. An example choice set for marketing a new laptop is provided. Statistical models like multinomial logit are used to analyze responses. The document discusses optimal design criteria like D-optimality and Bayesian optimal designs that account for unknown parameters. It also describes using quadrature methods like Mysovskikh quadrature to efficiently compute Bayesian D-optimal designs for non-linear models. The document concludes with noting the importance of design methods for discrete choice experiments and benefits of software for implementing such studies.