Biodiversity-ecosystem functioning experiments have shown that plant species and functional group richness are important drivers of grassland productivity, but the impact that plant order of arrival (i.e. priority effects) has on grassland overyielding and its drivers (complementarity and dominance effects) has been overlooked so far. Using species-specific plant biomass data collected in mixture and monoculture plots of a grassland field experiment (Julich Priority Effect experiment) that manipulated the order of arrival of three plant functional groups (forbs, grasses and legumes), we quantified net biodiversity effects (overyielding) as well as complementarity and dominance effects in mixtures one and 2 years after sowing. In this experiment, priority effects were created by sowing one functional group 6 weeks before the two others. First, we tested whether plant order of arrival affected overyielding, complementarity and dominance effects. Second, we investigated whether the magnitude of net biodiversity, complementarity and dominance effects was dependent on the strength and direction of priority effects. We found that complementarity and dominance effects were affected by plant order of arrival during community assembly. In addition, we found that moving from negative to positive priority effects increased grassland overyielding, mainly via increased complementarity effects. These results highlight the need to combine biodiversity and assembly approaches in future ecosystem functioning research, as this will increase the predictive power of community ecology in conservation and ecological restoration. A free Plain Language Summary can be found within the Supporting Information of this article.