Conservation translocations may be a useful tool for the restoration of declining freshwater invertebrates, but they are poorly represented in the literature. We conducted a before-after/control-impact (BACI) experiment to test the efficacy of conservation translocation for re-establishing abundant populations of the amphipod Gammarus lacustris, a declining species and wildlife food resource in depressional wetlands in the upper Midwest of the United States of America. Each study site (n = 19) contained at least one treatment wetland receiving translocated G. lacustris from a local donor and one control wetland. We selected study wetlands based on a suite of wetland characteristics and randomly assigned recipient versus control treatment. Gammarus lacustris was detected post-translocation at only 2 of 22 recipient wetlands (1 of 19 sites). Overall, there was a statistical increase in G. lacustris density in recipient wetlands compared to controls; however, the results were of minimal biological significance due to being driven by a single site with low G. lacustris densities. Accordingly, our results suggest that future conservation translocations of amphipods might be successful if limited to recently restored wetlands or informed by a more complex habitat suitability model to differentiate dispersal limitations from habitat limitations. To develop such a model would involve identifying the fewest, most influential physical and biological factors (e.g. wetland size/structure, fish, aquatic vegetation, and water chemistry) from the numerous inter-related factors that correlate with the abundance of naturally occurring G. lacustris; candidate wetlands to receive amphipods would be those for which the model predicts abundant G. lacustris but in which they do not presently occur.