Using store level scanner data to improve category management decisions: Developing positioning maps

Óscar González‐Benito

doi:10.1016/j.ejor.2008.10.015

Using store level scanner data to improve category management decisions: Developing positioning maps

2009-10-01

6

PRE

AI

Ó

Óscar González‐Benito *

M

María Pilar Martínez‐Ruiz

A

Alejandro Mollá Descals

DOI：10.1016/j.ejor.2008.10.015

Original

Original request for help

Save

Abstract

En 中文

This paper provides evidence of the usefulness of aggregated point-of-sale scanner data to infer the positioning of competing brands, providing valuable information for category management and hence facilitating decision making. Specifically, the authors propose a methodology to study the internal market structure based on market share models with latent heterogeneity when only macro-level time series data (not individual choices) are available. The proposed approach assumes a multidimensional decomposition, latent in the preference structure that is implicit to these types of models. By empirically applying this approach, the authors (1) simultaneously identify both latent dimensions of competing brands and latent segments with different brand preferences, (2) explain the competitive positioning of brands Without using disaggregated consumer panel data, and (3) achieve greater predictive performance. The findings offer insights to academics and practitioners interested in improving the practice of category management. (C) 2008 Elsevier B.V. All rights reserved

Keywords:

Marketing

Management

Decision making

Market response models

Internal analysis of the market structure

Positioning maps

Using store level scanner data to improve category management decisions: Developing positioning maps

Abstract

AI Summary

Journal

Organization