The Sequential Agglomerative Sorting task, a new methodology for the sensory characterization of large sets of products

The use of the free sorting task is compromised when the number of products to be assessed is too important. Some solutions have been put in place to manage these situations: (a) the use of incomplete presentation designs and (b) the use of predefined prototypes. These are unfortunately not optimal: the first brings out sensory dimensions of variability that are approximated at the level of the panel of participants; those revealed by the second may be conditioned by the predefined prototypes. The aim of this article is to present a new protocol of data collection that allows the characterization of large sets of products. It has been named the Sequential Agglomerative Sorting task (SAS task). It is partially based on a combination of the two previous solutions. All products are presented to the participant in the form of subsets. The latter are agglomerated by means of prototypes defined by the participant herself/himself. Thus, it reveals sensory dimensions that are neither approximated at the panel level nor conditioned by predefined prototypes. In this article, we present all the steps of implementation of the SAS task. A first study designed to test the performance of the SAS task is also presented. Practical Applications The Sequential Agglomerative Sorting task (SAS task) allows to obtain the sensory characterization of large sets of products. This new protocol of data collection was successfully applied to a set composed of 40 fragrances (study in this article). It can be applied to any kind of products: food, cosmetics, sport products, and so forth.