There is growing evidence that successful infection of hosts by pathogens requires a series of independent steps. However, how multistep infection processes affect host–pathogen coevolution is unclear. We present a coevolutionary model, inspired by empirical observations from a range of host–pathogen systems, where the infection process consists of the following two steps: the first is for the pathogen to recognize and locate a suitable host, and the second is to exploit the host while evading immunity. Importantly, these two steps conform to different models of infection genetics: inverse-gene-for-gene (IGFG) and gene-for-gene (GFG), respectively. We show that coevolution under this scenario can lead to coupled gene frequency changes across these two systems. In particular, selection often favors pathogens that are infective at the first, IGFG, step and hosts that are resistant at the second, GFG, step. Hence, there may be signals of positive selection between functionally independent systems whenever there are multistep processes determining resistance and infectivity. Such multistep infection processes are a fundamental, but overlooked feature of many host–pathogen interactions, and have important consequences for our understanding of host–pathogen coevolution.