Drag reduction in turbulent channel flows has significant practical relevance for energy savings. Various methods have been proposed to reduce turbulent skin friction, including microscale surface modifications such as riblets or superhydrophobic surfaces. More recently, macroscale surface modifications in the form of longitudinal grooves have been shown to reduce drag in laminar channel flows. The purpose of this study is to show that these grooves also reduce drag in turbulent channel flows and to quantify the drag reduction as a function of the groove parameters. Results are obtained using computational fluid dynamics (CFD) simulations with turbulence modeled by the k–ω shear-stress transport (SST) model, which is first validated with direct numerical simulations (DNS). Based on the CFD results, a reduced geometry model is proposed which shows that the approximate drag reduction can be quantified by evaluating the drag reduction of the geometry given by the first Fourier mode of an arbitrary groove geometry. Results are presented to show the drag reducing potential of grooves as a function of Reynolds number as well as groove wave number, amplitude, and shape. The mechanism of drag reduction is discussed, which is found to be due to a rearrangement of the bulk fluid motion into high-velocity streamtubes in the widest portion of the channel opening, resulting in a change in the wall shear stress profile.