This article presents a combined experimental/theoretical approach to accurately predict electrochemical redox potentials based on potential energy calculations. The approach works for experimental setups using different solvents and different reference electrodes and compensates for shortcomings in the prediction of redox potentials originating from the choice of DFT functional, basis set, and solvation model. The methodology is applied to two different sets of iron containing complexes which are used as redox catalysts in Chemically Regenerative Redox Fuel Cells (CRRFCs). For both sets of iron complexes with different 5 N donor ligands, an average deviation to the experimental values of < 0.02V is obtained. Expectedly, the deviation is slightly larger with changes being made in the first coordination shell, but is still within the predictive limit. The scheme is then applied to obtain ligands with both improved properties and lowest production cost.