The performance of Multistage Manufacturing Processes (MMPs) can be measured by quality, productivity and cost, which are inversely related to the variation of key product characteristics (KPCs). Therefore, it is crucial to reduce KPCs' variations by not only detecting the changes of process parameters, but also identifying the variation sources and eliminating them with corrective actions. Recent developments in the Stream-of-Variation (SoV) and Statistical-Process-Control (SPC) methodologies significantly improve the variation reduction for MMPs. This paper provides a review on the reported methodologies by comparing these two categories of methodologies in terms of their variation propagation modeling, process monitoring and diagnostic capability. With an illustrative case study, it is concluded that the recent advancements of SoV and SPC methodologies significantly improve the effectiveness of variation reduction. The discussion on the drawbacks of both methodologies also suggests the future research directions. Copyright © 2010 John Wiley & Sons, Ltd.