Achieving high selectivity is arguably the main upcoming challenge in heterogeneous catalysis for the 21st century. In complex reaction with competing parallel pathways, small changes in the relative values of the different activation energies are sufficient to switch the selectivity of those processes from one product to another. We in our laboratory have been carrying out mechanistic studies on the conversion of hydrocarbons on model metal surfaces to try to identify the key factors that control such selectivity. In this talk we will present a couple of examples of increasing subtlety from that work. In the first, we will discuss a reaction involving a C=C double bond isomerization, in particular cis-trans isomerization reactions. The second example will be focused on issues of bestowing enantioselectivity on solid surfaces by using cinchona alkaloid modifiers. In both case,s it will be shown how the mechanistic information obtained from surface-science studies can be extended to design more selective real catalysts.