<p>As the energy sector is the main source of anthropogenic greenhouse gas emissions and energy demand is constantly rising, expanding renewable energy sources is imperative. Offshore wind energy is key to this expansion. In the future, more powerful offshore wind farms will be installed in deeper waters. However, this presents new technical and logistical challenges with regard to large-scale substructures. Modular jacket substructures are one possible solution for scalable offshore infrastructure. These require connection systems that are both structurally and technically efficient. Slip joints fulfil these requirements particularly by transferring loads through controlled friction at contacting surfaces rather than through permanent mechanical connectors. This study aims to develop an innovative modular jacket substructure incorporating steel–steel and steel–concrete slip joints. The article focuses on characterising interface behaviour, friction mechanisms, and load-bearing performance across four slip joint material combinations: rolled steel–rolled steel, ground steel–ground steel, rolled steel–concrete, and ground steel–concrete. Surface roughness was quantified using both tactile profilometry and high-resolution 3D scanning, while joint fitting accuracy was assessed exclusively through 3D scan-based geometric analysis. Experimental push-out tests were conducted to determine friction coefficients and load-bearing capacities for each configuration. Additionally, a finite element model was developed to simulate the push-out test and predict the load-bearing capacity. The results revealed that the rolled steel surface exhibited pronounced height fluctuations, irregularity, and distinct macro-waviness. These features promoted mechanical interlocking at the rolled steel–rolled steel interfaces, which in turn led to increasing load resistance and friction coefficients with progressing displacement. In contrast, the other surface combinations exhibited the typical transition from static to kinetic friction. However, the statistical evaluation did not identify significant differences in load-bearing capacity between the material combinations under identical normal stress, and only the rolled steel–rolled steel interface showed a friction coefficient dependent on the applied normal stress. The numerical model captured the observed shear-lag behaviour and reproduced the experimental response with good accuracy. The overall predictive performance of the FE model was demonstrated by a coefficient of determination (R²) of 0.87, a mean absolute percentage error (MAPE) of 16.57%, and a root mean square error (RMSE) of 17.02. Overall, the ground steel–concrete interface proved the most favourable configuration for slip joint applications, as it provided a balanced combination of frictional resistance, stable sliding behaviour, moderate installation forces, and uniform wear characteristics. These features may facilitate assembly and disassembly while maintaining reliable load transfer, making this configuration a promising option for offshore slip joint connections.</p>