Individuals\' relatedness to their groups often changes with age, potentially favouring age-linked trends in social behaviours, and these kinship dynamics have been invoked to explain the evolution of life history traits such as early reproductive cessation and post-reproductive helping. While models showed that simple demographic parameters (e.g., the rates of male and female philopatry vs dispersal, and of local vs non-local mating) suffice to predict the patterns of kinship dynamics breeders experience, they make the unrealistic assumptions that survival and fecundity are age-independent, thus fail to capture the full complexity of kinship dynamics. In particular, they cannot consistently predict how individuals\' relatedness to their groups should change during their pre- and post-reproductive phases. Here, we extend existing models of kinship dynamics to allow any age-linked changes in mortality and fecundity of both sexes. We describe the new model, and demonstrate its improved predictive power, by comparing the observed kinship dynamics of the killer whales to the predictions of the previously available models (assuming constant mortality and fecundity) and of our new model (considering any observed age-specific survival and fecundity). We show the new model better predicts the data, successfully capturing the observed \'three-stage\' female kinship dynamics: an individual\'s relatedness to her group decreases initially while juvenile, increases across reproductive lifespan, and decreases again during post-reproductive life. These predictions demonstrate the power of our model in generating new insights into the theory of social life history evolution (e.g., explaining why post-reproductive lifespan evolved yet are constrained).