Constraining the mechanisms that control organic matter (OM) reactivity and, thus, degradation, preservation and burial in marine sediments across spatial and temporal scales is key to understanding carbon cycling in the past, present, and future. However, we still lack a quantitative understanding of what controls OM reactivity in marine sediments and, as a result, how to constrain it in global models. To fill this gap, we quantify apparent OM reactivity (i.e., model-derived estimates) by extracting reactive continuum model parameters (a and v) from observed benthic organic carbon and sulfate dynamics across 14 contrasting depositional settings distributed over five distinct benthic provinces. Our analysis shows that the large-scale range in apparent OM reactivity is largely driven by the wide variability in parameter a (10−3 < a < 107) with a high frequency of values in the range 100 < a < 104 years. In contrast, inversely determined v-values fall within a narrow range (0.1 < v < 0.2). Results also show that the variability in parameter a and, thus, in apparent OM reactivity is a function of the whole depositional environment, rather than the traditionally proposed, single environmental controls (e.g., water depth, sedimentation rate, OM fluxes). Thus, we caution against the simplifying use of a single environmental predictor for apparent OM reactivity beyond a specific local environmental context. In addition, diagenetic model results also indicate that, while OM fluxes exert a dominant control on depth-integrated OM degradation rates across most depositional environments, apparent OM reactivity becomes a dominant control in depositional environments that receive exceptionally reactive OM. Model results also show that apparent OM reactivity largely controls the relative significance of OM degradation pathways, and thus the redox zonation of the sediment, as well as depth of the sulfate-methane transition zone and rates of anaerobic oxidation of methane. Consequently, apparent OM reactivity also determines uptake and consumption of benthic terminal electron acceptors and nutrient recycling fluxes across a wide range of different depositional environments. In summary, our large-scale assessment not only further support the notion of apparent OM reactivity as a dynamic ecosystem property and highlights its crucial role for benthic biogeochemical cycling and exchange, but it also provides the first quantitative constraint on the most plausible range of reactivity parameters a and v. It thus represents an important advance for model parameterization as it largely alleviates the difficulty of determining OM reactivity in such models by constraining it to only one variable, i.e. the parameter a.