PurposeThe purpose of this paper is to analyze consumers’ reaction to assortment composition.Design/methodology/approachThis study examines several scenarios: private label (PL)-only assortments; mixed assortments (PL and national brands (NBs)), and in the latter case both small and large assortments. Consumers’ reaction is measured through three dependent variables: store image, PL purchase intention and store-switching intentions. The authors ran a structural equation model (SEM) to analyze the influence of different explanatory aspects (product category involvement, attitude toward PL, value consciousness and assortment variety perception) on consumers’ reactions for each scenario. For this research, the authors carried out an online experiment with a sample of 1,400 individuals from a large panel of consumers.FindingsConsumers react differently to different assortment compositions, giving importance to the differences between the three assortment models analyzed. The results show that the composition of the assortment, either according to its size or its structure, influences consumers’ response in a significant way. The results demonstrate that store image exclusively affects PL purchase intention in PL-only assortments. Only in mixed assortments is there a relationship between the assortment variety perception and store image, product category involvement and PL purchase intention, and both store image and value consciousness are related to store-switching intentions. Store-switching intentions decrease when consumers intend to purchase PL, but strictly in PL-only and large mixed assortments. Finally, value consciousness and variety perception are positively related to PL purchase intentions only in large mixed assortments.Research limitations/implicationsOne limitation of this research is that it is restricted to the Spanish context. Second, the methodology is based on an online experiment, with the advantages and disadvantages that this entails. Third, the authors did not differentiate between high- and low-value PL which, if undertaken, could be of interest for observing how brand value affects the management of retail assortments. Finally, the authors did not differentiate regular buyers at these retail chains from those who are not.Practical implicationsThe comparison between different assortment compositions helps the authors to draw some very interesting conclusions. The estimation of different consumers’ responses is ideal for providing retailers with recommendations on how to frame their assortment strategies. Thus, the main recommendation of this study for retailers is to look for a “balance” between PL and NBs, i.e., to offer mixed assortments.Originality/valueAside from mixed assortments, this study estimates the consequences of assortments that are exclusively PL. The authors proposed and deployed a SEM, so this paper contributes to the retailing field by including multiple dependent variables – store image, PL purchase intention and store-switching intentions. The authors conducted an online experiment containing “real” brands, which is another contribution as it enables consumers’ response to be estimated in a “real” environment.