Written by Cynthia Aadal, Senior Director of Retail & CPG
Retail and CPG aren’t failing at AI because they lack vision. They’re failing because they get stuck in the middle and stay there.
In my conversations with RCG leaders across retail and CPG, the pattern is consistent. Everyone has pilots. Everyone has models. Everyone has an AI roadmap. Very few can point to material enterprise value showing up in revenue, margin, or operating leverage.
The uncomfortable middle is where AI looks alive but it's close to death.
Stuck.
Models technically work, but no one fully trusts them. Insights are produced, but decisions don’t change. Data is centralized, yet accountability is fragmented. Teams are excited, but incentives, workflows, and ownership stay exactly the same. The business keeps running the old way, just with better slides.
In retail and CPG, this is deadly. Margins are thin. Complexity is relentless. Decisions compound fast. AI that doesn’t move demand, improve availability, optimize trade and media spend, or accelerate growth isn’t innovation, it’s cost.
What breaks AI strategies isn’t ambition.
It’s the refusal to operationalize.
Enterprise value only shows up when AI is tied directly to P&L levers, pricing, promotions, assortment, inventory, and customer lifetime value and embedded into how work actually gets done. Not as dashboards. Not as insights. As decisions that planners, merchants, marketers, and operators rely on every day.
Most organizations stop short. That’s the stall.
The opportunity isn’t to help clients do more AI. It’s to help them cross the uncomfortable middle, turning pilots into platforms, insights into execution, and data into decisions that scale the business. That means grounding AI in real retail and CPG workflows, forcing commercial accountability, and designing solutions executives trust because the results show up in the numbers.
AI strategies don’t fail because they’re too bold.
They fail because they never leave the middle.
That’s where value dies or scales.