Large multinational companies have developed extremely stable routines to manage high product and process complexity. A significant challenge is the continuous evolution of these routines through the use of new methods, allowing them to adapt to new technologies and competitive conditions. Jens Kohler examines how new methods can be successfully implemented and how critical situations in knowledge transfer can be overcome. He presents four case studies from the automotive industry and numerous findings from expert interviews. The author contrasts the empirical results with common theories of the innovation process and knowledge transfer. From this, he derives a three-stage procedural model and demonstrates that the objective, rational, and declarative aspects of knowledge transfer are often overestimated.