Metrics & MoreArticle Recommendations CONSPECTUS: Single-atom catalysis (SAC) is a fascinating and rapidly growing field in heterogeneous catalysis. In less than 20 years, this has become one of the most widely investigated subjects by the catalytic community for various good reasons: the ability to synthesize active catalysts using a minimum amount of precious metals, the expected higher selectivity of SACs compared to assemblies of nanoparticles of variable sizes, and the fact that SACs represent a bridge between homogeneous and heterogeneous catalysis. The relative simplicity of SAC structures compared to classical heterogeneous catalysts based on supported metal nanoparticles has stimulated an intense simulation activity aimed at predicting new potential catalysts from first principles, often based on machine learning algorithms. This is a very ambitious objective and ultimately represents the final goal of every modeling activity: the possibility to provide realistic predictions of new material properties and catalytic reactivity. However, one of the main reasons that theory is useful is and remains the interpretation and analysis of experimental results, with the no less crucial goal of understanding the basic principles that determine a certain functionality or reactivity. The combination of advanced characterization techniques and theoretical calculations can provide a general conceptual framework to better understand structure−function relationships. In this Account, we will address this aspect in trying to provide answers to some fundamental questions related to the structure, stability, and activity of SACs. We will start by addressing a question that arises every time a new SAC is synthesized: where are the metal atoms? What is their coordination mode with the support? Once we have shown how to address this point, we will move on to the next fundamental question: do the single atoms stay put? How does the chemical environment of a SAC depend on the preparation or reaction conditions? Next, we will analyze the importance of a full characterization of the reaction mechanism to predict reactivity. SACs, due to their analogy to coordination compounds, can form intermediates that do not exist on the surface of metal electrodes. The formation of these intermediates can influence the kinetics of the process and must be considered in the simulation. Finally, we will briefly address a more general question: how does the catalytic activity of SACs compare with that of the corresponding supported metal nanoparticles on the one hand or homogeneous molecular complexes on the other? The final message is that to answer these questions and to take full advantage of quantum chemical modeling, the results of the calculations should be continuously verified with experimental data in a cross-fertilization that is beneficial for both sides.