The use of solar energy for photoelectrochemically splitting water into H2 and O2 has been widely investigated for producing sustainable H2 fuel. However, no commercialisation of this technology has emerged. Currently the main obstacles to commercialisation are: low solar-to-hydrogen efficiency, expensive electrode materials, fast degradation of prototypes, and energy losses in separating H2 from O2 and water vapour in the output stream. The FotoH2 consortium has identified a new scientific direction for achieving cost-effective solar-driven H2 production, and it has the capability of large-scale prototyping and field testing the proposed technology. FotoH2 introduces anion-exchange polymer membrane and porous hydrophobic backing concepts in a tandem photoelectrochemical cell, and a novel way to stabilise the photoelectrodes based on a surface phase transformation. This approach allows the use of cost-effective metal oxide electrodes with optimal bandgaps and a simple flow-cell design without corrosive electrolytes.