In this study, we compare the performances of well injection and pond infiltration in controlling seawater intrusion in an unconfined coastal aquifer through two scenario groups: (1) a single injection well versus an elliptic infiltration pond and (2) an injection-extraction well pair system versus an elliptic infiltration pond-extraction well system. Comparison is based on quantitative indicators that include the interface toe location, saltwater volume, and maximum net extraction rate (for scenario 2). We introduce a method to determine the maximum net extraction rate for cases where the locations of stagnation points cannot be easily derived. Analytical analysis shows that the performances of injection and infiltration are the same, provided that the pond shape is circular. The examination of scenario group 1 suggests that the shape of the infiltration pond has a minor effect on the interface toe location as well as the reduction in the saltwater volume, given the same total recharge rate. The investigation of scenario group 2 indicates, by contrast, that the maximum net extraction rate increases significantly with the increasing ratio of b to a, where a and b are semiaxes of the ellipse parallel and perpendicular to the coastline, respectively. Specifically, for a typical aquifer assumed, an increase of 40% is obtained for the maximum net extraction when b/a increases from 1/200 to 200. Despite that the study is based on a simplified model, the results provide initial guidance for practitioners when planning to use an aquifer recharge strategy to restore a salinized unconfined coastal aquifer.