Mill initially focused on household solutions, but its co-founder and CEO Matt Rogers revealed that the food waste management startup had always aimed to expand into the commercial sector.

"Scaling to business clients was part of our plan from the A-round funding stage," Rogers said in an interview.

Now, Mill has finalized agreements with Amazon and its subsidiary Whole Foods Market, bringing its revenue model—profiting from processing food waste—into the public spotlight.

Starting in 2027, Whole Foods will deploy Mill’s commercial-grade food waste processors across all its stores. The equipment grinds and dehydrates waste from fresh produce sections, reducing costly landfill fees while converting residues into feed for Whole Foods’ poultry suppliers—cutting operational expenses.

Additionally, Mill’s processors collect data to help Whole Foods analyze waste patterns and optimize efficiency. "Our goal isn’t just streamlining waste disposal but reducing food waste at the source," Rogers emphasized.

Years ago, Mill first launched home-use food waste processors. The team, known for the Nest smart thermostat, ensured high design standards—devices so intuitive they’re "a joy to use," as one Silicon Valley adage goes.

"We deliberately entered the consumer market first—it helped us validate technology, gather data, and build brand loyalty," Rogers noted. Notably, many Whole Foods team members already used Mill’s home products before partnership talks began.

"That’s part of our B2B strategy," Rogers added. "We engage executives and suggest they try our home units first. It’s a proven way to spark interest."

Discussions with Whole Foods started a year ago. After months of piloting home units in select stores, Mill refined a commercial solution tailored to supermarkets.

The deal’s key driver? Mill’s AI-powered ability to predict and reduce waste preemptively. Its system analyzes whether discarded items could still be sold, addressing retail "shrink" (losses from spoilage or theft)—a critical edge in a competitive market.

Rogers credited large language model (LLM) advancements for this capability. At Nest, training cameras to recognize people and packages required Google’s massive engineering resources. With next-gen LLMs, Mill’s small team developed a superior AI system faster. "AI is our core engine," he said.

This tech accelerated Mill’s commercial rollout and diversified its revenue streams.

"Relying on one product or customer is risky," Rogers reflected, citing Apple’s iPod-dominated era (70% of revenue) as motivation for the iPhone. "We must build multiple pillars."

Mill continues expanding—next targeting municipal food waste solutions.

"We’ll keep strengthening our business with diversified, resilient models," Rogers concluded.