Even though shape optimization is a powerful tool for designing aerospace vehicles, it can be time-consuming when high-fidelity models are employed. Thus, lower-fidelity simulations covering a wider design space can be a solution for shape optimization in the early design phases. With this in mind, the present work aims to develop a low-fidelity and fast method to conduct nozzle shape optimization. This method consists in using the free-form deformation (FFD) parameterization technique to control the nozzle shape by means of an optimization algorithm to maximize the coefficient of thrust determined by a two-dimensional method of characteristics (MoC). To verify the reliability of the proposed method, a similar optimization process is carried out, recurring to high-fidelity simulations, namely using an Euler solver, in the open-source framework text {SU}^2. This latter optimization process is established as a surrogate-based optimization (SBO) not only to mitigate the text {SU}^2 framework limitations in performing shape optimization on nozzles, but also as a way to reduce the computational power. A good agreement between the results from both methods is achieved, displaying solely a small offset concerning the optimal contour width and the coefficient of thrust. Hence, this proves the usefulness of the developed shape optimization strategy based on the MoC for the preliminary design of nozzles.