The spatial structure of biodiversity — i.e., how the numbers and kinds of species vary systematically across the globe, in response to factors such as resource availability and environment — is crucial for explaining the exceptional richness of life on Earth. Because of this spatial structure, diversity scales with area, causing the comparatively modest counts of species coexisting in local communities to scale up to spectacular numbers at regional to global levels. Understanding how the spatial structure of biodiversity changed through the Phanerozoic is therefore critical for understanding how biodiversity itself was assembled through deep time. To date, though, no study has systematically quantified how diversity–area relationships varied through the Phanerozoic, or which spatial scales were most important for driving overall biodiversity change. Here, I present genus-area relationships (GARs; both local–global, and within geographic regions) and patterns of hierarchical beta diversity across local to global spatial scales for Phanerozoic marine animals. Changes in the intercept of the GAR closely track previously-documented variation in diversity at local and regional spatial scales, including an abrupt increase across the Cretaceous–Paleogene boundary. The slope of the GAR, by contrast, varies within set bounds, but does not trend throughout the entire Phanerozoic. However, GAR slopes become flatter after mass extinctions. Steeper GAR slopes are associated with reef-supporting regions, suggesting that such environments may have hosted higher beta diversity. Decomposing total beta diversity across hierarchical spatial scales shows that smaller spatial scales have always contributed substantially more to overall beta than larger scales, but that beta at larger scales has been more variable through the Phanerozoic. These results demonstrate that while there has been variation in the spatial scaling of marine animal biodiversity through the Phanerozoic, driven by factors such as environmental variation, no directional or secular trend is apparent. Variation in the extent of colonised area may have also driven changes in global diversity. However, reliably inferring such changes is limited by the incomplete spatial coverage of the sampled fossil record.