- Sex reversal technology, realized through androgenic gland (AG) manipulation, was recently introduced as a process for production of all-male producing broodstock. This technology exploits however, a relatively small number of sex-reversed broodstock. Thus, both genetic improvement via a breeding program and prevention of inbreeding are needed to ensure the sustainability of such technology. Three wild strains of prawns originating from geographically (though not necessarily genetically) isolated locations in India [Gujarat (G), Kerala (K) and West Bengal (WB)] were assessed for their suitability as breeders for all-male production. In addition, their potential for a selective breeding program was evaluated. A comparative evaluation of early sex segregation, sex reversal, growth performance, and population structure in the three selected strains was performed. Among the purebred strains, after eight months of grow out in earthen ponds, growth performance of the WB strain was the best (59.39 ± 1.08 g), while that of G was the poorest (26.50 ± 0.94 g). Strain-additive genetic effects for body weight at harvest were highest for the WB strain (+ 45.9%) and lowest for the G strain (− 28.3%). Body masses of WB × K and WB × G crosses were 14.2% and 8.8% above the mean mass of the purebred strains, respectively, while that of the K × G cross was 23% below this value. In most crosses, males reached heavier mean body weights than did females with higher frequencies of the large male morphotypes being seen in the WB purebred strain and its respective crosses. Reciprocal effects for body mass ranged from 4% to 14.9% below the mean of the purebred strains. These negative signs mean that in the two crosses involving the WB strain, growth performance is higher when this stain was used as the sire strain. Similarly, the growth performance of the K × G cross was higher when the former was used as the sire strain. Average heterosis effect for body weight was minor (− 0.51 ± 0.73) and did not differ significantly from zero. The high correlation between strain additive effects (the major source of variation in growth) and total performance for body weight ( r = 0.927) indicate the existence of valuable genetic variation that could be exploited in a selective breeding program. For all-male production, males from the three strains were segregated at early post-larval stages and microsurgical AG removal was performed. In all the strains, similar low levels of complete sex reversal into functional neo-females (genetic males) were realized (0.17% − 0.34%). These produced relatively small numbers of neo-female to be crossed with normal males to produce the desired all-male population, but raise the possibility that such a process could result in a genetic bottleneck. Thus, a genetic improvement scheme for each strain integrated with periodical crosses of the resulting neo-females from one strain with males from another strain is suggested to avoid inbreeding.