The truth is out there! How external validity can lead to better marketing decisions

Greg Rogers
Procter & Gamble

Didier Soopramanien
Lancaster University

INTRODUCTION

Academics and practitioners use conjoint experiments in various sectors such as travel demand, planning and new product development. Conjoint analysis is the process of collecting and analysing information on decisions made by consumers when they have to trade off different value propositions of goods and services. For a more comprehensive review of the various methods and applications of conjoint analysis, the interested reader is directed to the following references: Wittink et al. (1994); Green et al. (2001); Wittink and Bergestuen (2001). Conjoint experiments include paired comparison approaches using a card-sorting method, adaptive conjoint analysis (ACA) using techniques such as genetic algorithms, and choice-based conjoint (CBC). In CBC experiments, as opposed to other types of conjoint experiment, consumers choose a product proposition that consists of a bundle of attributes, rather than ranking attributes of a product. CBC has grown in popularity as it can be used as the analytic backbone to virtual shopping experiments (e.g. Campo et al. 1999). The market leader in choice modelling software provision, Sawtooth Software Inc., reports that CBC now makes up approximately 75% of all conjoint studies using its software – up from 50% just three years ago (Sawtooth Software 2007).