A long-standing scientific challenge in the field of alcohol deoxygenation has been direct catalytic sp3 C–O defunctionalization with high selectivity and efficiency, in the presence of other functionalities, such as free hydroxyl groups and amines widely present in biological molecules. Previously, the selectivity issue had been only addressed by classic multistep deoxygenation strategies with stoichiometric reagents. Herein, we propose a catalytic late-transition-metal-catalyzed redox design, on the basis of dehydrogenation/Wolff–Kishner (WK) reduction, to simultaneously tackle the challenges regarding step economy and selectivity. The early development of our hypothesis focuses on an iridium-catalyzed process efficient mainly with activated alcohols, which dictates harsh reaction conditions and thus limits its synthetic utility. Later, a significant advancement has been made on aliphatic primary alcohol deoxygenation by employing a ruthenium complex, with good functional group tolerance and exclusive selectivity under practical reaction conditions. Its synthetic utility is further illustrated by excellent efficiency as well as complete chemo- and regio-selectivity in both simple and complex molecular settings. Mechanistic discussion is also included with experimental supports. Overall, our current method successfully addresses the aforementioned challenges in the pertinent field, providing a practical redox-based approach to the direct sp3 C–O defunctionalization of aliphatic primary alcohols.