About

AI assistants promise self-service analytics, but most struggle to go beyond basic reporting and KPI retrieval. This session explores why today’s AI architectures limit advanced analysis and fail to provide lasting business value. Using real-world examples, Stephen shows how giving large language models richer business context — without exposing live or sensitive data — enables more accurate reasoning and more powerful analytics.

Attendees will learn how an AI-powered centralized data catalogue can act as a knowledge repository for their organization, helping both people and AI better understand data, build trust in insights, and reduce engineering effort. The session also outlines practical steps teams can take today to start treating data as a product and build a scalable foundation for true self-service analytics.

Takeaways 
• Why current AI assistant tools struggle with advanced analytics
• How a centralized data catalogue improves AI reasoning without using live data
• How to treat data as a product to build long-term competitive advantage
• Practical steps to start building a knowledge repository today