Résumé : Climate, landscape composition, management practice, and wild bee pollination are all variables thought to play significant roles in commercial apple production. However, how these variables affect production efficiency under field-realistic conditions has not been investigated at large geographical scales. We combined intensive standardized field surveys (using netting and pan traps) with structural equation models to explore the relative impact of biotic and abiotic variables on bee diversity, apple yield and fruit quality, and their ability to represent reliable proxies of apple production. Here we show that apple yields are mainly driven by management practice, without evidence for a significantly superior contribution by managed honey bees. Total wild bee diversity, while negatively correlated with honey bee dominance, promoted apple quality by enhancing seed set number. Our study demonstrates that even across a broad geographical range there is potential to harness wild bee diversity as nature-based solution and as a substitute to an exclusive reliance on honey bees in the context of commercial apple production.