Composite ceramic capillaries coated with thin palladium membranes are developed for the production of CO‐free hydrogen for PEM fuel cells, via alcohol steam reforming. The composite membranes are tested for pure H2 and N2, as well as for synthetic reformate gas. The aim is to develop a heat‐integrated compact membrane reformer for decentralized hydrogen production. In this context, a deep knowledge of the performance, behavior, and necessary treatment of the composite palladium membranes plays a decisive role in process design. The current contribution focuses on the main hurdles met while attempting to exploit the potential of ceramic supported capillary palladium membranes.