Successful identification of new chemical entities with drug-like properties in pharmaceutical and academic research groups involves an early screen and the use of a large number of public and proprietary chemical libraries. Before applying high-throughput experimental screening approaches, virtual screening strategies have been put in place in order to sort and filter this massive amount of compounds and data available at these very early stages. Chemoinformatic tools have a crucial role in this selection process and enable therapeutic chemists to focus very early on promising candidates. Virtual screening has conventionally been based either on models of the target or the ligand (molecule), but today these models include biopharmaceutical filters addressing right from the start of the project the ADMET (absorption, distribution, metabolism, excretion, and toxicity) properties of the molecules. Above all, chemoinformatic tools help chemists understand better the chemical diversity they can work with, especially when comparing chemical libraries. This paper will focus on exemples of the day-to-day use of chemoinformatics in screening programs. A large part will be dedicated to new tools (chemographic and pharmacographic approaches) being developed for the representation and analysis of chemical diversity, but also for combining chemical and biological information to expedite research programs.