In many cooperatively breeding species, females mate extra-group, the adaptive value of which remains poorly understood. One hypothesis posits that females employ extra-group mating to access mates whose genotypes are more dissimilar to their own than their social mates, so as to increase offspring heterozygosity. We test this hypothesis using life history and genetic data from 36 cooperatively breeding white-browed sparrow weaver (Plocepasser mahali) groups. Contrary to prediction, a dominant female's relatedness to her social mate did not drive extra-group mating decisions and, moreover, extra-group mating females were significantly more related to their extra-group sires than their social mates. Instead, dominant females were substantially more likely to mate extra-group when paired to a dominant male of low heterozygosity, and their extra-group mates (typically dominants themselves) were significantly more heterozygous than the males they cuckolded. The combined effects of mating with extra-group males of closer relatedness, but higher heterozygosity resulted in extra-group-sired offspring that were no more heterozygous than their within-group-sired half-siblings. Our findings are consistent with a role for male–male competition in driving extra-group mating and suggest that the local kin structure typical of cooperative breeders could counter potential benefits to females of mating extra-group by exposing them to a risk of inbreeding.