Rapid advances in information technology have made it possible to generate, analyze, and distribute large data sets of material properties.  One of the great challenges in materials research is to effectively make use of these data sets to accelerate the design and development of new materials. To this end, I will discuss how data can be leveraged in three ways:  1)  I will demonstrate how machine learning can be used to efficiently develop cluster expansions for alloy catalysts that give unique insights into their structure, stability, and catalytic activity.  Examples will be drawn from work on platinum alloy surfaces and nanoparticles for the oxygen reduction reaction.  2)  I will demonstrate how genetic programming can be used to rapidly identify relevant descriptors for material properties, facilitating rational materials design.  3) I will demonstrate how informatics can be used to rapidly generate highly efficient k-point grids, addressing a longstanding problem in computational materials research.  Our group has developed a publicly accessible k-point grid server backed by a database of tens of thousands of k-point grids, and we estimate that for a broad range of computational methods the use of this server should nearly double the speed of electronic structure calculations on materials.