In each phase of semiconductor product manufacturing, there are different engineering focuses. Therefore, different statistical tools are introduced. In the pilot phase, it is critical to find out the best process condition and explore the process window. Design of experiments (DOE) is a handy tool to achieve the desired process conditions with limited resources and a shortened cycle time. In risk production, strict process control ensures process stability and early failure detection. Statistical Process Control (SPC) tools are widely involved to monitor process performance, ensure the process is under control and reduce quality variation. In the mass production phase, capacity extension is just as important as process stability, and full data matching is a key mindset to ensure quick extension without excursion. JMP can play a key statistical role throughout the whole life of a semiconductor product.