The arsenate (AsO4) and phosphate (PO4) mobility in aerobic soil is affected by soil organic matter (OM). This study was set up to quantify the interaction between OM and AsO4 with an observational, experimental and computational approach. The adsorption of AsO4 was measured with the radiotracer 73AsO4 in samples taken from different horizons of two soil profiles. In four samples, the OM concentration was increased experimentally. The AsO4 adsorption data were analysed with the CD-MUSIC model using ferrihydrite, with OM as competitor and isotopically exchangeable phosphorus of the soil as the total PO4 bound on the reactive surface. The solid–liquid distribution coefficient (KD) of 73AsO4 increased by more than two orders of magnitude with a decrease in total organic carbon (OC) concentration. The addition of Suwannee River OM (∼ 1 g OC kg-1) to samples with small OC (∼ 2 g kg-1) decreased the KD values 15-fold, whereas the effect was less (two-fold) in samples with a large OC (∼ 30 g kg-1). Soluble AsO4 and PO4 could be described well and simultaneously by introducing surface reactive OM (RO−) as an adjustable parameter in the geochemical model. The fitted RO− increased with increasing OC concentration, with a slope of 1.3 ± 0.15 mmol RO− g-1 OC. The amount of RO− expressed per mol iron (Fe) and aluminum (Al) hydroxides was maximum at a molar ratio of ∼ 0.34 at > 10 g OC kg-1 soil, which corresponds to earlier published capacities of an organo-mineral association that might affect potential soil C sequestration. Our research showed that OM enhanced the mobility of AsO4 and PO4; however, the surface reactive OM fraction needs experimental quantification. Highlights: Quantification of organic matter–arsenate competition in soil. Include organic matter in geochemical models to predict arsenate and phosphate mobility in soil. Surface reactive organic matter is determined by total organic carbon and Fe and Al hydroxides. Arsenate adsorption is controlled naturally by organic matter concentration.