Traditional hit-to-lead optimization assumes that upon elaboration of chemical structure, the ligand retains its binding mode relative to the receptor. Here, we build a large-scale collection of related ligand pairs solved in complex with the same protein partner: we find that for 41 of 297 pairs (14%), the binding mode changes upon elaboration of the smaller ligand. While certain ligand physiochemical properties predispose changes in binding mode, particularly those properties that define fragments, simple structure-based modeling proves far more effective for identifying substitutions that alter the binding mode. Some ligand pairs change binding mode because the added substituent would irreconcilably conflict with the receptor in the original pose, whereas others change because the added substituent enables new, stronger interactions that are available only in a different pose. Scaffolds that can engage their target using alternate poses may enable productive structure-based optimization along multiple divergent pathways.