Despite some initial resistance, pharmacogenomics is now finding widespread use and application throughout all phases of clinical drug development in many pharmaceutical companies. Applications, feasibility, and deliverables of phar-macogenomic studies are largely dictated by sample sizes, availability of clinical (phenotypic) endpoints, and existence of working hypotheses, and therefore vary with the phase of clinical development. Variability in a given clinical endpoint is the main driver and prerequisite for pharmacogenomic investigations. Applications can therefore readily be classified into three broad categories of clinical endpoint: phar-macokinetics, efficacy, and safety, although there may be overlap among categories depending on underlying mechanisms. Applications range from mechanistic, regulatory, trial design (inclusion/exclusion), product differentiation, companion diagnostics, and portfolio decision-making. The promise of pharmacogenomics has been advertised for some time. We are gradually beginning to see the fruits of our labors in the context of the pharmaceutical industry, as apprehensions surrounding this technology fade in favor of recognition that a better understanding of our compounds is beneficial in the long run. In an industry that is hungry for innovation, alternative approaches, even if associated with some unknowns and some risks, are imperative. More developments are expected as experience with the application of pharmaco-genomics grows (including both successes and failures) and as the industry continues to work both competitively and collaboratively to realistically apply this technology toward therapeutic innovation and evolution of the basic science.