Identifying different types of coevolutionary dynamics is important for understanding biodiversity and infectious disease. Past work has often focused on pairs of interacting species, but observations of extant communities suggest that coevolution in nature occurs in networks of antagonism and mutualism. We discuss challenges for measuring coevolutionary dynamics in species-rich communities, and we suggest ways that established approaches used for two-species interactions can be applied. We propose ways that such data can be complemented by genomic information and linked back to extant communities via network structure, and we suggest avenues for new theoretical work to strengthen these connections. Quantifying coevolution in species-rich communities has several potential benefits, such as identifying coevolutionary units within networks and uncovering coevolutionary interactions among pathogens of humans, livestock, and crops.