Atomic force spectroscopy has become a widely used technique for investigating forces, energies, and dynamics of biomolecular interactions. These studies provide dissociation kinetic parameters by pulling apart proteins involved in a complex. Biological complexes are studied under near-physiological conditions, without labeling procedures, and are probed one at time, the latter allowing to one obtain results which are not averaged over the ensemble. However, to gain reliable information, some experimental aspects have to be carefully controlled. In particular, the immobilization of molecular partners to AFM tips and Supports, required to force the molecular dissociation, plays a crucial role in determining the Success of the experiments. To actually resolve single interactions, multiple simultaneous complex dissociations have to be avoided, and nonspecific adhesions, commonly found in these studies, have to be recognized and discarded. This article is aimed at offering a critical revisitation of the atomic force spectroscopy technique applied to the study of biomolecular interactions, highlighting the critical points, identifying strategies to be adopted for a more reliable data extraction and interpretation, and pointing out the experimental and theoretical aspects which still need to be refined. To this purpose, we take advantage of the vast landscape of literature and then proceed into the details of our works. In this respect, we describe the general principles of the technique, the procedures for protein immobilization, and how they can affect the results. We emphasize the use of computational docking to predict molecular complex configurations, when unknown, as a useful approach to select proper anchorage architectures. Additionally, we deal with data acquirement and analysis, with regard to the force curve selection, to the force histograms interpretation, and to the theoretical frameworks used to extract kinetic parameters. Through this, we outline that AFS can be successfully used both to investigate complexes having very different affinities and also to reveal competitive binding mechanisms, thus gaining deeper information about molecular interactions.