The discovery of small-molecule ligands against protein targets of pharmaceutical relevance represents a central problem in academic and industrial research laboratories. However, current ligand-identification technologies are highly demanding both in terms of time and resources. In contrast, DNA-encoded chemical libraries represent an emerging methodology, which allows a rapid and cost-effective idenfication of novel ligands. Such libraries consist of small organic molecules, which are individually coupled to DNA tags serving as PCR-amplifiable identification bar codes. DNA-encoding allows the in vitro selection of binding molecules from large compound repertoires by affinity capture experiments against an immobilized protein of interest, in analogy to established selection methodologies like antibody phage display. Different library construction strategies have been investigated in order to optimize library performance and to explore problem-specific applications. The advent of high-throughput sequencing technologies enabled the synthesis and use of DNA-encoded chemical libraries comprising up to millions of library members. This article reviews the development of library construction strategies, commenting on newly identified binders against pharmaceutically relevant proteins and on associated challenges and trends in the field.