Understanding the dynamics of socio‐ecological systems is crucial to the development of environmentally sustainable practices. Models of social or ecological sub‐systems have greatly enhanced such understanding, but at the risk of obscuring important feedbacks and emergent effects. Integrated modelling approaches have the potential to address this shortcoming by explicitly representing linked socio‐ecological dynamics. We developed a socio‐ecological system model by coupling an existing agent‐based model of land‐use dynamics and an individual‐based model of demography and dispersal. A hypothetical case‐study was established to simulate the interaction of crops and their pollinators in a changing agricultural landscape, initialised from a spatially random distribution of natural assets. The bi‐directional coupled model predicted larger changes in crop yield and pollinator populations than a unidirectional uncoupled version. The spatial properties of the system also differed, the coupled version revealing the emergence of spatial land‐use clusters that neither supported nor required pollinators. These findings suggest that important dynamics may be missed by uncoupled modelling approaches, but that these can be captured through the combination of currently‐available, compatible model frameworks. Such model integrations are required to further fundamental understanding of socio‐ecological dynamics and thus improve management of socio‐ecological systems.