Industrial acceptance of organic solvent nanofiltration (OSN) is hampered by the slow process of membrane screening based on trial and error for each solute-solvent couple. The extensive experimental screening is still necessary due to limitations of predictive models. The basic physico-chemical processes are well understood, but the complexity and variety in the details of the competing interactions challenge the accuracy of predictions. Recently, data-driven techniques showed their potential in the field of OSN. We apply data-driven techniques for ceramic membranes, which have been one of VITO’s technological focus points in OSN activities for many years. The non-swelling property of these membranes reduces the complexity of the separation process and consequently that of the prediction models required. At the same time, the known sensitivities of data-driven techniques are no less present here: the amount of (experimental) data needed to represent the chemical space of interest, dimensionality of input data, quality and completeness of data, … In addition to these input data challenges, there is the demand for the explainablity of trained prediction models and the connection of the gained knowledge with the physico-chemical understanding of the separation process. In the end, this is what can lead to well understood designs and leverage the acceptance of the technology and its contribution to sustainable chemistry.