This meta-analysis was conducted as an effort to reply to the questions ‘did formulated feed perform as well as natural feed in terms of growth and survival of finfish which had weaned on it? And what accounted for the contradictory research findings on this issue?’ 42 articles were included from which data on final weight, FCR, HSI and survival rate were extracted. High heterogeneity was detected among assays of final weight, FCR as well as HSI, and thus, subgroup analysis and meta regression under random effect model were used to analyze the effect sizes; for survival rate, the heterogeneity was neglectable (I2 = 0.00), so the overall effect size estimate was accepted directly after adjustment by trim-and-fill method as non-neglectable publication bias was indicated. Leave-one-out analysis was conducted to test the impact of individual studies on the combined effect sizes. Results showed that fish fed on formulated feed had significantly lower final weight and significantly higher FCR than those fed on natural feed, while they survived as well as the counterparts. Although the combined effect size of HSI had no significant different from 0, liver alteration was detected in many species and further research are needed. Species of the subjects and their trophic levels, feeding ration, protein content of natural feed and the experiment duration partially accounted for the contradictions among the individual studies. In addition, the relatively poor performance of formulated feed was not due to protein or lipid deficiency; further studies are needed to explain the gap and then fill it.