Studying population genetic connectivity (i.e., identifying gene flow among populations and understanding their impacts on the genetic structure and diversity of populations) is first a matter of knowing what we work on, that is, accurately delimiting evolutionary units. Here, we focused on Pocillopora damicornis sensu stricto (or Pocillopora PSH04 sensu Gélin et al. in Mol Phylogenet Evol 109:430–446. http://dx.doi.org/10.1016/j.ympev.2017.01.018PANIST, 2017). From 458 colonies sampled within the tropical southwestern Pacific [Chesterfield Islands and New Caledonia (Grande Terre and Loyalty Islands)], Bayesian assignments and network analyses were conducted with 11-microsatellite loci to first evaluate the genetic partitioning of the colonies in distinct Secondary Species Hypotheses (SSHs), then in distinct clusters. Population genetic connectivity was then assessed for each cluster separately. Pocillopora PSH04 was partitioned into two highly differentiated SSHs (SSH04a and SSH04b), regularly found in sympatry. Furthermore, SSH04a was subdivided into two clusters (SSH04a-1 and SSH04a-2). This pattern of genetic structuring seems not related to clonality, but rather to the establishment of reproductive barriers. Nevertheless, considering each cluster separately, the populations appeared highly differentiated, suggesting relatively weak gene flow. This low connectivity among populations, coupled with the existence of cryptic species, brings new insights to the connectivity pattern of this understudied Pacific region.