Abstract Informed technology decision-making requires a structured understanding of cost evolution over time. A dynamic approach integrating learning curves and process-based cost modeling is introduced to examine learning in manufacturing. The approach is applied to the case of a hydroforming process, and quantifies the cost impacts of learning improvements in cycle time, downtime, and reject rates. A comparison with cases of automotive assembly and wire drawing illustrates that variation in learning is tied to the individual process cost structure. The results show aggregate cost evolution is strongly dependent on cost structure and that major cost elements may not align with major cost improvement-through-learning opportunities. The analyses can be used to focus intentional learning activities on primary learning operational drivers.