Metal nanoparticles have in recent decades been the subject of intense research owing to their wide range of size and shape-dependent properties, which makes them interesting candidates for a variety of applications. Gold nanorods represent a particularly intriguing type of particle due to their tunable plasmonic properties. To benefit from these features, access to a large supply of high-quality, monodisperse nanorods is required, but has yet to be realized. The most facile route to obtaining nanorods is through seeded-mediated growth, a family of wet-chemical synthesis protocols. While such protocols have the potential of delivering high quantities of rods with finely tuned properties, progress is being hampered by a lack of theoretical understanding regarding the growth mechanism. The aim of this thesis is to lay the foundation for a comprehensive picture of the gold nanorod growth process. In particular, the role of the CTAB surfactant layer covering the nanorod surfaces is investigated. The discussion is based on first-principles density functional theory calculations, molecular dynamics simulations carried out in our research group, and a critical review of ideas found in literature. A framework for explaining experimental observations and predicting the outcome of synthesis processes as a function of their parameters is presented based on the surface phase diagram of CTAB. A key feature of the phase diagram, in addition to the conventional bilayer structure that is often assumed, is the existence of cylindral and spherical micelles. These micelles expose the underlying gold surface to incoming gold ions that fuel the growth. The importance of dispersive interactions in understanding transitions between the different phases is emphasized. Furthermore, different models for the anistropic growth are reviewed within the newly established framework, and are complemented by a discussion of the impact of the seed particle structure.