We critically highlight some evidence for the importance of soil biodiversity to sustaining (agro-)ecosystem functioning and explore directions for future research. We first deal with resistance and resilience against abiotic disturbance and stress. There is evidence that soil biodiversity does confer stability to stress and disturbance, but the mechanism is not yet fully understood. It appears to depend on the kind of stress and disturbance and on the combination of stress and disturbance effects. Alternatively, community structure may play a role. Both possible explanations will guide further research. We then discuss biotic stress. There is evidence that soil microbial diversity confers protection against soil-borne disease, but crop and soil type and management also play a role. Their relative importance as well as the role of biodiversity in multitrophic interactions warrant further study. Henceforth, we focus on the effects of plant and soil biodiversity on nutrient and water use efficiencies as important ecological functions in agroecosystems. The available evidence suggests that mycorrhizal diversity positively contributes to nutrient and, possibly, water use efficiency. Soil fauna effects on nutrient and water use efficiencies are also apparent, but diversity effects may be indirect, through effects on soil structure. We present a conceptual diagram relating plant and soil biodiversity with soil structure and water and nutrient use efficiencies as a framework for future studies. We then consider how cropping systems design and management are interrelated and how management options might be interfaced with farmers’ knowledge in taking management decisions. Finally, we attempt to express some economic benefits of soil biodiversity to society as part of a wider strategy of conserving and using agrobiodiversity.