The interaction between residential preferences and dwellings is a complex system whose function thus far remains insufficiently explored. In this paper, we investigate housing functions as orchestrators of households’ residential mobility in the context of Swiss rental housing. We propose a theoretical multi-step model and use survey data from 878 Swiss tenants to inspect the model’s linkages. From the statistical analysis, we firstly observe that tenants’ residential satisfaction is more likely to increase when the gap between ideal housing functions and those actually fulfilled by the current dwelling decreases. Secondly, results show that the effectiveness of an event (e.g. a job opportunity) in triggering the move is significantly related to both residential satisfaction and the functions the dwelling fulfils prior to the trigger. Thirdly, findings show that these trigger events can be grouped into three types: radical change, problem-solving and opportunity. With a medium effect size, a radical change was found to bring about the strongest change in housing functions between past and current dwellings. Lastly, in line with the hypothesis that residential preferences vary over the life course, socio-demographic characteristics and tenancy types are found to be significant explanatory variables for households’ ideal housing functions. By disentangling the complexity of the housing system, the proposed multi-step model can be used to integrate households’ preferences with supply-side constraints in agent-based model simulations, thereby contributing to fostering the provision of quality housing, i.e. dwellings able to meet the needs of current and future occupants.