Mathematical psychologists and statisticians have, over the last 50 years, refined methodologies for zeroing in on individuals' preferences using heterogeneous discrete choice tasks. Yet visual design elements, due to their high‐dimensional, holistic, and interactive nature, do not lend themselves well to measurement using extant methods for preference elicitation.

We incorporate real‐time, interactive, 3D‐rendered configurations into such measurement frameworks, using scalable machine learning algorithms to adaptively update visual designs. At the heart of the method is a parametric decomposition of an object's geometry, along with a novel, adaptive “bi‐level” query task that can estimate individuals’ preferences among visual designs.

We illustrate the method’s performance through simulation and a field test for the design of a mid‐priced sedan, using real‐time 3D rendering and an online panel. Bayesian part-worth estimation and online training via ranking SVM allow the proposed method to elicit trade-offs between design and more traditional elements generally, and also pinpoint which design details differentially drive (heterogeneous) preferences.