Several smartwatch vendors have over the past years entered the market with a large range of devices that differ in the design qualities they offer. In this paper, we focus on design qualities that are either perceived, descriptive or physical, and we investigate which of them may lead people to prefer particular smartwatches over others. To do so we conducted a laboratory study with 93 potential smartwatch users and a trained panel study with 8 participants. Results were analysed through a multivariate statistical technique called Preference Mapping. The advantage of this technique is that it can relate a large number of design qualities to users’ preferences. Our findings show that participants can be divided into four groups with homogenous preferences, each emphasising a unique combination of design qualities. For example, some groups emphasised device shape, while others prioritised expensiveness for their preferred smartwatch. We conclude with implications of our work on practice and research.