The generation of catalyst design tools is important for developing both economically and environmentally friendly reactions. This thesis focuses on the use and development of such a tool, molecular volcano plots, which have the ability to estimate catalytic performance while also providing an ameliorated understanding of the intricate chemistry that influence reactions. The original work presented here is divided into three parts that cover: (1) exploration of volcano functions, (2) acceleration of catalyst-screening process and (3) extraction of the detailed chemistry for interesting chemical reactions from volcano results. The first examines additional functions of volcano plots in which, for the first time, substrate scopes can be screened using volcano plots as similar as catalysts. This type of volcanoes can estimate the reactivity of each substrate and also provide an enhanced understanding of substrate scope, an important facet of chemistry that is often overlooked by computation. The second section focuses on accelerating the catalyst-screening process through the use of machine-learning models. Specifically, we illustrate how machine-learning models can be used to predict the value of a descriptor variable that can be related to catalyst performance through volcano plots. This procedure allows us to expand the number of catalyst being screened to tens of thousands of species which, in turn, provides a much broader picture of catalyst behavior. The use of big-data techniques highlights the specific manner in which metals and ligands can be combined to identify tuned catalyst for a given chemical reaction. In the final section, we use molecular volcanoes to probe the catalytic hydrogenation of carbon dioxide to formate using transition metals paired with pincer ligands. This work ultimately identified a combination of group 9 metals with pi-acidic pincer-ligands as the best catalysts for this reaction. Using this same reaction, we also developed a molecular volcano variant that directly predicts an experimental observable, the turnover frequency (TOF), from the value of a descriptor variable, thus establishing a TOF volcanoes. Overall, this thesis demonstrates how molecular volcanoes can be used as a rational catalyst design tool in the field of homogeneous catalysis as well as route to uncovering chemical trends that provide greater fundamental understanding about why specific catalysts and/or substrates exhibit high functionality for a particular reaction.