Advances in sequencing technologies have led to a rapid increase in available bacterial and archaeal genome information, but for much of this diversity, little ecological information is available. Reverse ecology provides a potential path forward by using genomic information to gain insight into the ecological associations and niche spaces of organisms. A crucial first step is to predict population structure, which provides the basis for analyzing genomes for evidence of ecological differentiation. Although delineation of bacterial and archaeal populations remains difficult, we outline how gene flow information can be used to identify populations as genetic units, which also are ecological units because adaptations can spread through them in a specific manner. This approach is particularly powerful when closely related populations are analyzed for signatures of differential selection that indicate recent ecological differentiation. Genome-wide association studies can also help identify mutations and genes underlying ecologically relevant traits. Albeit still in their infancy, reverse ecology approaches have the potential to order microbial diversity into genetically and ecologically cohesive units and hence provide the opportunity to test hypotheses about the evolutionary mechanisms creating and maintaining diversity within and between populations.