An Integrative Model for Complex Conjoint Analysis

Abstract

Product complexity can challenge the use of conjoint analysis when the number of product features under study becomes large. A complex product exists whenever a marketplace offering is made up of simpler, component products. Tables and chairs make up dining room sets, and hotels offer rooms (size, layout, location) with beds, bathrooms, restaurants, exercise rooms and parking options. The challenge in modeling demand for complex products lies in simultaneously studying the major drivers of a purchase decision and also the components that comprise the offering. Demands on data increases as more product features are included in an analysis, and at some point, it becomes necessary to study some component product features separately. This paper proposes an integrative model for studying product features for a complex product utilizing multiple conjoint exercises within a single structure. We show that the integration of a focused analysis with an aggregate analysis of marketplace offerings requires adjustment beyond that expected from simple changes due to the scale of random utility error or estimates of price sensitivity. Our model is illustrated with data from the automobile industry where various option packages are available.