Inside the learning curve

We explore the specificity of volume-based learning in an outsourced setting. When producing a unit of output, the content of the knowledge gained can vary dramatically from one unit to the next. This suggests that while aggregate experience in learning-by-doing is generally valuable, not all prior experience has an equal impact on performance. To examine these differences we introduce a framework to unpack the multiple dimensions of experience that exist within one unit of work. We then empirically examine the customer-, domain-, and technology-specificity of learning. Our empirical setting is the context of outsourced radiological services where individual doctors at an outsourcing firm complete radiological reads for hospital customers. We find that customer-, domain-, and technology-specific experience-as compared to other experience-leads to improved productivity. We discuss the implications of our results for the study of learning and experience, as well as for outsourcers and the firms that use their services.