Weekly essays with actionable insights on what truly matters in data and AI and why they matter for your organization's long-term success. I write to clarify my own thinking, articulating ideas forces deeper understanding and broader perspective. My goal is for these reflections to offer practical value as you navigate your own data and AI journey.
Most governance frameworks fail by optimizing for coverage instead of business outcomes. Strategic governance requires three pillars: compliance requirements as your foundation, organizational needs addressing specific friction, and strategic enablers that unlock competitive advantage.
After twenty years working with data, I've seen the same pattern: companies discover data quality issues and respond with standardization initiatives, new policies, stricter controls. It never works. The problems just get hidden under bureaucracy while the dysfunction continues underneath.
Grocery retailers have a decisive advantage in AI shopping: their data. Generic AI agents struggle with product discovery, but retailers possess years of purchase history revealing individual tastes. The race is about combining targeted AI with proprietary data.
AI shopping agents face a fundamental problem: the less you see, the less you buy. By focusing on efficiency, they eliminate the product discovery that drives grocery sales. You can't ask for what you don't know exists, and impulse purchases represent over half of grocery revenue.
When planning groceries on Monday, you don't add the cookies you'll crave Friday. AI shopping agents face this planning gap: we shop for an idealized self, not the person who wants treats after a long day. Impulse purchases represent over half of grocery revenue.
78% of employees bring their own AI tools to work. The real challenge isn't Shadow AI itself, but distinguishing harmless productivity from genuine risk. Most violations happen because employees don't recognize when they've crossed the line with sensitive data.