Shellfish reefs are being restored in southern Australia to recover ecosystem services lost when reefs were destroyed. There is a focus on enhancing fish productivity, but the capacity of restored reefs to enhance fish production remains unknown, hampering estimates of return on investment. We quantified the density differences of newly-recruited juvenile fish on restored subtidal reefs, relative to those on unrestored, unstructured habitat, by surveying three paired reef-unstructured locations using unbaited stereo video deployments over 12 months. We used FishID software to automatically identify, size, and count fish in videos and then modelled productivity enhancement. Fourteen of the sixteen species occurring as new recruits were more common on reefs than unstructured habitat. Enhancement of fish production from restored reefs from a single year’s cohort is, on average, 6,186 kg.ha-1.yr-1 (SD 1,802) after enough time has elapsed for all species to have matured. Species harvested commercially or recreationally contributed 98% of production. Enhancement varied among locations, from 12,738 kg.ha-1.yr-1, the highest yet recorded anywhere, to 1.4 kg.ha-1.yr-1. The combination of videos and automated data extraction proved a reliable, cost-effective method. Our estimates of enhanced fish productivity underpin calculations of ecological, social, and financial benefits, supporting the business case for scaling-up restoration.