Gartner: $1.5 Trillion AI Spend Faces Data Quality Barriers
Enterprise data leaders identify data quality as the top barrier to AI success, with 73% citing it above other factors, amid $1.5 trillion in AI investments for 2025.
Companies invested $1.5 trillion in artificial intelligence in 2025, according to Gartner, yet 73% of enterprise data leaders identify data quality as the primary barrier to AI success, ranking it above model accuracy, compute costs, and talent. Additionally, 60% of companies report little to no value from their AI investments, indicating that the issue lies not with AI technology itself but with underlying data problems.
The Complexity of Enterprise Marketing Stacks
Enterprise marketing teams manage data across multiple systems, with leads entering from sources like paid campaigns, content syndication, webinars, online forms, tradeshows, and telemarketing, all feeding into a marketing automation platform (MAP) that connects to multiple CRM instances, a unified data warehouse, analytics platforms, consent management systems, and AI models. This setup means bad data propagates through the stack, affecting segmentation in the MAP, routing in the CRM, storage in the data warehouse, reporting in analytics layers, and recommendations from AI models. B2B contact data decays at roughly 30% per year, and one study of over 1,200 business contacts found that 70% experienced at least one change within 12 months, such as job title updates or email changes.
The Propagation and Real Costs of Bad Data
When a bad record enters the system, it spreads across a 10-to-15 system stack, distorting lead segments, skewing AI scores, inflating pipeline forecasts, and poisoning training data for future models, according to the analysis in Demand Gen Report. The average enterprise CRM has a 25% critical error rate on contact records, and 94% of organizations suspect their customer and prospect data is inaccurate. The Sirius Decisions “1-10-100 rule” quantifies this as $1 to verify a record at entry, $10 to clean it later, and $100 if ignored, with costs multiplying in modern stacks where data syncs in real time across systems.
Impacts on AI Adoption and Organizational Performance
Bad data costs the average organization $12.9 million annually, per Gartner, and MIT Sloan estimates a revenue impact of 15–25%, effects that worsen with AI involvement as corrupted records feed models without human oversight. Forrester stated that data quality is the primary factor limiting B2B GenAI adoption, and Gartner predicts that through 2026, organizations will abandon 60% of AI projects lacking AI-ready data. A Sales Hacker survey of 250 Sales Operations Managers revealed that 41% of predictive lead scoring initiatives failed due to CRM data issues, while US B2B marketing data spending grows at just 0.5% according to eMarketer, contrasting with 36% year-over-year growth in AI tool spending. The BCG 10-20-70 framework emphasizes allocating 70% of AI resources to people and processes, including data governance, to address these challenges, as detailed in Demand Gen Report.
Wider Context in AI Development
As a widely-known issue in AI, poor data quality has long been a concern in technology adoption, framed here by the specific B2B statistics from the report.
