Effective management of threatened and exploited species requires an understanding of both the genetic connectivity among populations and local adaptation. The Olympia oyster (
Ostrea lurida
), patchily distributed from Baja California to the central coast of Canada, has a long history of population declines due to anthropogenic stressors. For such coastal marine species, population structure could follow a continuous isolation‐by‐distance model, contain regional blocks of genetic similarity separated by barriers to gene flow, or be consistent with a null model of no population structure. To distinguish between these hypotheses in
O. lurida
, 13,424 single nucleotide polymorphisms (
SNP
s) were used to characterize rangewide population structure, genetic connectivity, and adaptive divergence. Samples were collected across the species range on the west coast of North America, from southern California to Vancouver Island. A conservative approach for detecting putative loci under selection identified 235
SNP
s across 129
GBS
loci, which were functionally annotated and analyzed separately from the remaining neutral loci. While strong population structure was observed on a regional scale in both neutral and outlier markers, neutral markers had greater power to detect fine‐scale structure. Geographic regions of reduced gene flow aligned with known marine biogeographic barriers, such as Cape Mendocino, Monterey Bay, and the currents around Cape Flattery. The outlier loci identified as under putative selection included genes involved in developmental regulation, sensory information processing, energy metabolism, immune response, and muscle contraction. These loci are excellent candidates for future research and may provide targets for genetic monitoring programs. Beyond specific applications for restoration and management of the Olympia oyster, this study lends to the growing body of evidence for both population structure and adaptive differentiation across a range of marine species exhibiting the potential for panmixia. Computational notebooks are available to facilitate reproducibility and future open‐sourced research on the population structure of
O. lurida
.