Analysis of 34S/32S of sulfate in rainwater and soil solutions can be seen as a powerful tool for the study of the sulfur cycle. Therefore, it is considered as a useful means, e.g., for amelioration and calibration of ecological or biogeochemical models. Due to several analytical limitations, mainly caused by low sulfate concentration in rainwater, complex matrix of soil solutions, limited sample volume, and high number of samples in ecosystem studies, a straightforward analytical protocol is required to provide accurate S isotopic data on a large set of diverse samples. Therefore, sulfate separation by anion exchange membrane was combined with precise isotopic measurement by multicollector inductively coupled plasma mass spectrometry (MC ICP-MS). The separation method proved to be able to remove quantitatively sulfate from matrix cations (Ca, K, Na, or Li) which is a precondition in order to avoid a matrix-induced analytical bias in the mass spectrometer. Moreover, sulfate exchange on the resin is capable of preconcentrating sulfate from low concentrated solutions (to factor 3 in our protocol). No significant sulfur isotope fractionation was observed during separation and preconcentration. MC ICP-MS operated at edge mass resolution has enabled the direct 34S/32S analysis of sulfate eluted from the membrane, with an expanded uncertainty U (k = 2) down to 0.3 ‰ (a single measurement). The protocol was optimized and validated using different sulfate solutions and different matrix compositions. The optimized method was applied in a study on solute samples retrieved in a beech (Fagus sylvatica) forest in the Vienna Woods. Both rainwater (precipitation and tree throughfall) and soil solution δ 34SVCDT ranged between 4 and 6 ‰, the ratio in soil solution being slightly lower. The lower ratio indicates that a considerable portion of the atmospherically deposited sulfate is cycled through the organic S pool before being released to the soil solution. Nearly the same trends and variations were observed in soil solution and rainwater δ 34SVCDT values showing that sulfate adsorption/desorption are not important processes in the studied soil.