MnO x −Na 2 WO 4 /SiO 2 is one of the best-performing catalysts in the oxidative coupling of methane (OCM) to C 2 hydrocarbons (C 2 H 6 and C 2 H 4 ). The current mechanistic concepts related to the selectivity to the desired products are based on the involvement of crystalline Mn-containing phases, the molten Na 2 WO 4 phase, surface Na−WO x species, and the associated lattice oxygen. Using in situ X-ray diffraction, operando UV−vis spectroscopy, spatially resolved kinetic analysis of product formation in steady-state OCM tests, and temporal analysis of products with isotopic tracers, we show that these phases/species are not categorically required to ensure high selectivity to the desired products. M 2 WO 4 /SiO 2 (M = Na, K, Rb, Cs) materials were established to perform similarly to MnO x −Na 2 WO 4 /SiO 2 in terms of selectivity−conversion relationships. The unique role of the molten Na 2 WO 4 phase could not be confirmed in this regard. Our alternative concept is that the activity of M 2 WO 4 /SiO 2 and product selectivity are determined by the interplay between the lattice oxygen of M 2 WO 4 and adsorbed oxygen species formed from gas-phase O 2 . This lattice oxygen cannot convert CH 4 to C 2 H 6 but oxidizes CH 4 exclusively to CO and CO 2 . Adsorbed monoatomic oxygen species reveal significantly higher reactivity toward overall CH 4 conversion and efficiently generate CH 3 radicals from CH 4 . These reactive intermediates couple to C 2 H 6 in the gas phase and are oxidized, to a lesser extent, by the lattice oxygen of M 2 WO 4 to CO and CO 2 . Adsorbed diatomic oxygen is involved in the direct CH 4 oxidation to CO 2 . The electronegativity of alkali metal in M 2 WO 4 was established to affect the catalyst ability to generate adsorbed oxygen species from O 2 . This knowledge opens the possibility to influence product selectivity by controlling the coverage by adsorbed and lattice oxygen via reaction conditions or catalyst design.