Using product and retail choice attributes for cross‐national segmentation

Purpose – This study aims to use product attributes and retail display information to develop cross‐national segments. Design/methodology/approach – The study uses consumers' choice in wine stores to develop segments of consumers, based on the relative importance of 13 factors consumers use for the retail purchase of wine. Data are collected from 11 countries, using seven languages on five continents. Best worst scaling (BWS) and latent class analysis were used to develop the clusters. Findings – A three‐cluster structure provides a straightforward and robust segmentation across the 11 countries. This model fits better than one based on 11 single country segments. The three segments reveal different ways in which consumers choose wines: cognitive‐based, assurance‐based, and in‐store promotion‐based. Research limitations/implications – The samples are convenience‐based and do not represent the population of wine drinkers in each country. Choice criteria, including retail communications, can be used to develop useful and robust segments cross‐nationally. Practical implications – The three segments found in this study provide clear guidelines for wine marketers depending on whether they work for small or large wine companies. The use of choice attributes and BWS show the utility of this method in cross‐national research. Originality/value – This research demonstrates that product attributes and retail communication devices like labels and displays can be used for cross‐national segmentation. Applying BWS and Latent Class Clustering to choice criteria leads to clear, usable, and robust segmentation across a wide range of cultures and product use histories.