Management of beef suckler cattle herds requires a difficult but vitally important balance between farm profits, animal health and welfare and sustainable food production. A dynamic programming (DP) model was implemented to investigate the consequences of replacement and management decisions on the interactions and possible trade-offs between animal welfare, fertility and profitability in breeding beef suckler cattle herds. The model maximized profit from the current cow and all successors by identifying the best keep/replace decision. The 150 states incorporated in the DP model were all combinations of: ten cow-parity, five calving periods including one barren state (five in total) as fertility indicators and three body condition scores at weaning as an animal welfare indicator reflecting feeding and nutritional conditions of animals. Statistical models were fitted to data from a breeding suckler cattle herd, consisting of performance records of 200 cattle over 5 years, to parameterize the DP model. Estimated parameters used in the DP model were: (i) probabilities of transitions between states and (ii) probability of involuntary culling. These estimates were used in the form of conditional probabilities of successful or failed (as a result of involuntary culling) transitions to the next state. In addition, statistical models were used to estimate probability of calving difficulty. There was strong evidence (P< 0·001) that parity affected calving difficulty and weak evidence (P = 0·067) that parity affected the incidence of involuntary culling. The DP model outcomes indicated that cows calving very early, i.e. those who conceived in the first 21 days after artificial insemination, showed reduced frequencies of calving difficulty as well as voluntary culling, and so gave better financial returns than late-calving cows and barren cows. As a result, fewer replacements were needed that reduced the frequency of calving difficulty, further implying a win–win scenario for both profit and welfare. In contrast, in late-calving animals, the frequency of calving difficulty increased and they were less profitable and more prone to be culled. Results of sensitivity analysis showed that the optimum voluntary culling rate was sensitive to commodity market prices. These findings suggest well-informed nutrition and reproduction management could deliver a win–win outcome for profit and animal welfare.