Designing agroecological cropping systems, which have enhanced biodiversity and that improve agroecosystem services, is recognized as the most likely method of improving the environmental sustainability of agriculture. However, tools and methods for designing such systems are lacking. To help to fill this gap, we propose a revised trait-based response/effect framework as applied to agroecosystems, which takes into account farmers' decision rules. The framework consists of a “Biophysical module”, which describes the biophysical functioning of the agroecosystem on a response/effect traits basis and a “Decision module”, which encompasses the farmer's choices that follow decision rules, to account for the high degree of human control of filters and community structure operating in cultivated systems. The introduction of the Decision module and its interactions with the Biophysical module opens new research priorities related to trade-offs between services, to species choice and to the relationships between the community composition, functional structure and the functions. Synthesis and applications. We proposed a revised trait-based response/effect framework as applied to agroecosystems, which incorporates farmers' decisions. This framework has great potential to address questions related to the strategic choices associated with multispecies cropping system design, from plant (species choices) to community (optimization of community composition) scales. It also contributes to improving the rationale to manage multifunctional agroecosystems, which extend beyond yield alone, by enabling the exploration of trade-offs between ecosystem services.