Understanding and controlling the structure and composition of nanoparticles in supported metal catalysts are crucial to improve chemical processes. For this, atom probe tomography (APT) is a unique tool, as it allows for spatially resolved three-dimensional chemical imaging of materials with sub-nanometer resolution. However, thus far APT has not been applied for mesoporous oxide-supported metal catalyst materials, due to the size and number of pores resulting in sample fracture during experiments. To overcome these issues, we developed a high-pressure resin impregnation strategy and showcased the applicability to high-porous supported Pd-Ni-based catalyst materials, which are active in CO2 hydrogenation. Within the reconstructed volume of 3 × 105 nm3, we identified over 400 Pd-Ni clusters, with compositions ranging from 0 to 16 atom % Pd and a size distribution of 2.6 ± 1.6 nm. These results illustrate that APT is capable of quantitatively assessing the size, composition, and metal distribution for a large number of nanoparticles at the sub-nm scale in industrial catalysts. Furthermore, we showcase that metal segregation occurred predominately between nanoparticles, shedding light on the mechanism of metal segregation. We envision that the presented methodology expands the capabilities of APT to investigate porous functional nanomaterials, including but not limited to solid catalysts.