Successful reproduction in a single breeding event has consistently been shown to reduce condition, fecundity and survival to the following breeding season. Few studies have examined the cumulative costs of frequent reproduction on survival. Here we use a dataset of female red deer (Cervus elaphus) from the Isle of Rum, Scotland, to model survival probability within a mark–recapture framework. By including both recent reproduction and long-term cumulative reproductive effort in the models we tested whether knowledge of lifetime reproductive effort improves our estimates of survival probability. We found that the fit of the model was significantly improved with the inclusion of longer-term measures of reproductive history. Heterogeneity in the reproductive performance of individuals influenced the expected survival cost of reproduction, with high cumulative reproductive effort associated with high survival, except with individuals reproducing in their first year where reproduction was associated with a decrease in survival. This work emphasises the need to account for reproductive history when estimating the survival probabilities of animals.