In this study, we demonstrate how materials science can be combined with the established methods of organic chemistry to find mechanistic bottlenecks and redesign heterogeneous catalysts for improved performance. By using solid-state NMR, infrared spectroscopy, surface and kinetic analysis, we prove the existence of a substrate inhibition in the aldol condensation catalyzed by heterogeneous amines. We show that modifying the structure of the supported amines according to the proposed mechanism dramatically enhances the activity of the heterogeneous catalyst. We also provide evidence that the reaction benefits significantly from the surface chemistry of the silica support, which plays the role of a co-catalyst, giving activities up to two orders of magnitude larger than those of homogeneous amines. This study confirms that the optimization of a heterogeneous catalyst depends as much on obtaining organic mechanistic information as it does on controlling the structure of the support. (C) 2012 Elsevier Inc. All rights reserved.