Single-nucleotide polymorphisms (SNPs), common variations among the DNA of individuals, are being uncovered and assembled into large SNP databases that promise to enable the dissection of the genetic basis of disease and drug response (i.e., pharmacogenomics). Although great strides have been made in understanding the diversity of the human genome, such as the frequency, distribution, and type of genetic variation that exists, the feasibility of applying this information to uncover useful pharmacogenomic markers is uncertain. The health care industry is clamoring for access to SNP databases for use in research in the hope of revolutionizing the drug development process. As the reality of using SNPs to uncover drug response markers is rarely addressed, this review discusses practical issues, such as patient sample size, SNP density and genome coverage, and data interpretation, that will be important for determining the applicability of pharmacogenomic information to medical practice.