Hemiparasitic plants are considered as ecosystem engineers because they can modify the interactions between hosts and other organisms. Thereby, they may affect vegetation structure, community dynamics and facilitate coexistence as they are able to reduce interspecific competition by parasitizing selectively on competitive species and promote subordinate ones. In agri-environmental schemes, introducing the hemiparasite Rhinanthus has therefore been suggested as a low-cost method to increase grassland plant diversity, which is still subject to debate. 2 The majority of previous studies simply compared sites with and without hemiparasites. However, as hemiparasite effects are most likely density-dependent, we present a novel approach assessing the effect of Rhinanthus alectorolophus density on grassland plant diversity, yield and biomass quality. Moreover, we investigated whether functional plant composition and community mean traits are affected, which has been largely neglected in previous studies. The relationship between species richness and relative Rhinanthus biomass followed an optimum curve with highest values at 31% relative Rhinanthus biomass. At this Rhinanthusbiomass level, species richness was increased by 12% and yield decreased by 26% compared with plots without Rhinanthus. At relative Rhinanthus biomass > 60%, species richness was even lower than in plots without Rhinanthus. Overall, the biomass of grasses and the cumulative cover of legumes decreased linearly with increasing relative Rhinanthus biomass. Community mean trait analysis revealed that an increasing Rhinanthus density shifts the community composition towards smaller plant species. Biomass quality was not affected by increasing relative Rhinanthus biomass. In summary, our results of increased plant diversity − in line with a slightly lower yield but similar biomass quality − indicate that Rhinanthus is a suitable biological tool for grassland restoration.