Incorporating emotions into evaluation and choice models: Application to Kmart Australia

This paper addresses the repositioning of Kmart Australia in 2011. It shows how by calibrating emotional as well as cognitive reactions and estimating their impact on purchase intentions, Kmart was able to focus its communications, improving market share. We measure nine key emotions, ranging from surprise to anger. Including these emotions significantly improves our model for likelihood to choose a store. Measuring emotions enabled Kmart s advertising agency to create a television commercial that tapped into the specific emotions that most strongly predict the likelihood to choose a store; that is, the model drove the development of the advertising creative. The resulting television commercial tested well and was effective when launched. At the individual level, cognitions and emotions changed dramatically. At the aggregate level, an econometric model showed that store visits were significantly enhanced. Kmart s EBIT (earnings before interest and tax) increased by 30 , whereas Kmart s main rival had almost no EBIT growth, despite vigorous attempts to counter Kmart s campaign. One of our key contributions is to incorporate emotions into marketing science models of evaluation and purchase intentions. We also provide a new methodology to measure emotions. The approach enables marketing science to participate in the design of marketing stimuli, rather than just testing preexisting ones.