This document discusses market mix models and some of the challenges involved in their use. It begins by explaining how marketers turned to more scientific approaches like market mix models to address demands for accountability and ROI assessment. However, several challenges were overlooked, including misunderstanding what models can and cannot do, data integration issues, and the limitations of only using observational data. The document then provides an overview of how market mix models work today and some of the technical issues that must be addressed, such as collinearity of factors, data quality problems, and endogeneity. It concludes by noting the need to accommodate different effects like carryover, diminishing returns, and cross-sectional differences across segments.