Docking calculations to predict ligands for G protein-coupled receptors (GPCRs) have become a widely used technique in recent years. Yet, for each new receptor system and receptor conformation the question remains how to best prepare a system to
enrich ligands over non-binders. Moreover, particularly in settings where (known) ligands should be docked individually and with high precision in order to enable further development, small details can determine success or failure. I will present cases where we have investigated the roles of conformation, presence of water molecules and protonation states of amino acids. Moreover, I will showcase docking campaigns that have led us to novel ligands for more exotic target sites, such as class F receptors or the dimerization interface of an orphan receptor.