The primary goal of this study was to examine the applicability of preference-based segmentation for a broad array of meals in the context of teenagers. A representative sample of 1,168 Norwegian schoolchildren provided an evaluation of 20 common dinner meals in terms of preference (liking). Cluster analysis was used to establish four distinct preference-based food segments. These four segments were termed Food Lovers, Fish Haters, Fish Lovers and Dislikers. The relationship between these four preference-based segments and demographic, consumption, attitude and lifestyle variables was finally modelled using a multinomial logit analysis. The results highlight the fact that social and family-related attitude and lifestyle variables have a greater ability to profile the segments and explain segment membership than demographic variables. The findings indicate that there are distinct and interesting differences between these segments. The findings have some implications for marketers within the industry in terms of effectively targeting the different market segments.