The restoration of shellfish reefs to soft‐sediment environments often relies on the translocation of donor stock, typically from aquaculture, to the seafloor. In New Zealand, subadult green‐lipped mussels (Perna canaliculus) grown on subtidal aquaculture long‐lines are being considered over adult conspecifics to increase restoration efficiency but are currently limited by predation and hydrodynamic dislodgment following transfer to the seabed. In this study, the survival of subadult mussels from five different sources representing a range of growth conditions (i.e. subtidal aquaculture long‐lines 1 and 2, subtidal shellfish aquaculture baskets 1, wild intertidal reefs 1 and 2) were compared across separate translocations (1 and 2) to see if careful consideration of stock source could improve subadult mussel survival following transfer to the seafloor. Mussel morphology (shell strength, attachment thread structure) and clumping behavior (perimeter:area ratios, clump densities, clump complexity) were compared among populations to explain relationships among prior stock growth conditions, mussel size, and survival. Following experimental translocations, high survival (>90%) was closely related to decreases in perimeter:area ratios of mussel clumps and increases in clump complexity. High survival groups of mussels were selected from stock with high shell compression strength (subtidal aquaculture long‐line 2, 95.5% survival), greater attachment thread number and thickness (aquaculture basket 1, 92.3% survival), or some combination of the two (wild intertidal reef 2, 99.3% survival). This study supports further consideration for incorporating subadult mussels into restoration provided they are selected from sources that produce individuals with the resistant characteristics outlined in this study.