Intermediate cities have had a key role in urban dynamics and development in Latin America since the second half of the twentieth century, yet the theoretical and methodological approaches to research about these types of city have not been suited to their complexity. Most research on them reduces to demographic criteria, overlooking the fact that they are characterised by intense socio‐spatial interactions with their surrounding rural areas and other nearby urban centres. This is a serious shortcoming in terms of the potential use of such research for urban planning, and the resulting development processes of these cities. The present work proposes that while research based on the physical patterns of the movement of people (including commuting) are relevant, it is necessary to delve deeper into the process of mobility in order to improve the comprehension of intermediate cities in Latin America. First, the article illustrates the weak points of traditional approaches to intermediate cities and from this proposes that the theoretical and methodological contributions of the new mobility paradigm (Sheller and Urry 2006, Environment and Planning A 38 207–26) can contribute considerably to the study of such cities, as well as the phenomena of intermediation inherent to them. The article then introduces a new methodological proposal, based on mixed mobile methods, that contributes to empirical research and understanding of intermediate cities. An empirical application of the methodology is performed in Chile, specifically in the city of Villarrica in the La Araucanía Region. Based on this application, the article provides some important conclusions regarding the research experience and the theoretical and methodological approach necessary to attend to the socio‐spatial complexity of intermediate cities in Latin America.