Design of a selective membrane reactor requires a kinetic model (or rate expression) for transport through the membrane and the effects of various co-adsorbates. In many cases the membrane area, the most expensive component in the process, is determined by the hydrogen transport rate (permeance) while its sensitivity to the catalytic reaction rate is marginal. The purpose of this opinion is to examine the initial steps made in using advanced computational tools (DFT) to predict permeance and adsorption competition from first principles. Several comparisons of predictions with experimental data have been conducted as well as Pd/M material composition optimization of such membranes. Making this comparison useful requires extending current knowledge to various crystal facets and obtaining better characterization of the membrane surface. In the long run one would expect to gain understanding on more complex phenomena like reaction on the membrane and membrane deactivation.