To expand the economic operating range of offshore wind farms, less common but innovative solutions regarding support structures should be considered. To increase the popularity of gravity-based foundations (GBF) within the offshore wind industry, a better understanding of the hydrodynamics on GBFs, as well as an improved insight in the optimal design process for these structures is essential. Therefore, this thesis proposes a universal design methodology for the preliminary design of concrete GBFs for offshore wind turbines. The hydrodynamic GBF load model is based on a linear potential flow solution and was validated using Computational Fluid Dynamics (CFD) data. It provides a sufficiently accurate and highly cost-efficient approximation of the wave loads during the conceptual design phase. For the detailed design phase, CFD is still required since it represents the state-of-the-art in wave load modelling. A simplified structural model was created to perform a natural frequency analysis to ensure structural integrity. Due to the high stiffness of GBFs, it is reasonably easy to design the first natural frequency of the structure to be within the desired soft-stiff region, and thus resonance is only a minor challenge. The main concerns for GBFs rather depend on the weight and cost-efficiency of manufacturing, transportation, and installation. The initial design is based on the environmental conditions, wind turbine data, and design guidelines. The complete model is verified using design checks based on industry standards, which ensures foundation stability for the design limit states. The requirements for foundation stability are often driven by the size and weight of the structure. Therefore, sensitivity analyses were performed with various design parameters to identify the main factors influencing the design. As designs often have multiple objectives, the optimal design depends on the objective considered and thereby numerous optimal designs may exist. For a GBF, these objectives are often related to weight, costs, or even both. Generally, a combination of design objectives with varying priorities should be considered. In addition, implementing optimization algorithms is recommended as manual preliminary design optimization is labor-intensive and has difficulties considering all the various design objectives. For future research, it is recommended to test the universality of this methodology. This can be done by modelling GBFs of varying shapes over a broad range of water depths, wave conditions, and design load conditions.