A Bain & Company survey of 951 executives across nine sectors indicates that 40% of companies measuring AI-related cost reductions achieved 10% or less, falling short of internal projections used to justify spending. Primary data from the April survey highlights that 44% of firms funding subsequent AI initiatives rely on prior savings assumptions rather than verified returns, creating circular dependencies. Perspectives from financial services and manufacturing executives emphasize data accessibility barriers as the leading constraint, consistent with patterns in decade-long data modernization expenditures exceeding hundreds of billions globally. Counter perspectives from technology sector respondents note that early integration challenges may resolve with workflow adjustments, though the survey isolates this as secondary to structural data issues. Related primary analyses, including Goldman Sachs assessments from 2023 on tokenomics and ROI thresholds, align on the distinction between technological capability and value capture. The MIT Center for Collective Intelligence report from 2024 further documents 95% pilot failure rates tied to integration gaps, providing additional empirical context without attributing outcomes to single causes. Bain recommends initiating with available data feeds rather than awaiting full modernization, a procedural shift observed in select insurance and retail cases.