Hungry for Learning: Wendy’s Will Croushorn

2025-12-16 · Source: MIT Sloan Management Review · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Data Science & Analytics, Emerging Technologies & Innovation · Depth: Intermediate, extended

Summary

Wendy's product manager Will Crouchhorn discusses the Fresh AI initiative, which uses AI agents to manage drive-thru orders, aiming to enhance customer experience and accessibility. The system currently handles approximately 150,000 orders daily with 95% accuracy, supporting both English and Spanish, with plans for more languages. Crouchhorn highlights the complexity of drive-thru orders, noting 67 trillion possible ways to order the menu in English, and emphasizes AI's role in providing consistent, personalized service. The initiative, developed in collaboration with Google Cloud, focuses on removing communication barriers for customers with diverse language needs or atypical speech patterns, such as stutters. The development process is iterative, prioritizing continuous learning from customer feedback, including monitoring "sorry" metrics to identify and reduce friction points, and drawing inspiration from diverse fields like Disney Imagineering and the MIT museum.

Key takeaway

For product managers and executives considering AI integration, prioritize initiatives that address specific customer pain points and enhance accessibility, rather than simply adding AI for its own sake. Focus on iterative development, leveraging customer feedback (even from social media) to refine the system and build trust. Ensure your data infrastructure is designed to be consumable by AI agents to maximize value and enable scalable, consistent customer experiences.

Key insights

AI can transform customer experience by automating complex interactions and enhancing accessibility at scale.

Principles

• No order is "normal"; account for immense variability.
• AI development is iterative; learn continuously from feedback.
• Focus on what AI does well, empower humans for what they do best.

Method

Wendy's Fresh AI uses speech-to-text models, trained with Google Cloud, to process complex drive-thru orders in multiple languages, displaying real-time transcripts for confirmation and accessibility, and continuously refining performance based on customer interaction data.

In practice

• Track "friction metrics" like customer/agent "sorry" counts.
• Seek inspiration from diverse, non-competitor industries.
• Design data systems for consumption by both humans and AI.

Topics

• AI Agents
• Drive-thru Automation
• Customer Experience
• Responsible AI
• Data Management

Best for: Executive, Product Manager, AI Product Manager, Director of AI/ML, MLOps Engineer

Related on AIssential

• A New Kind of Marketplace — Building trust and a robust "harness" infrastructure are key to enabling disruptive, agent-driven marketplace experiences.
• Hungry for Learning: Wendy’s Will Croushorn — AI can transform complex customer interactions by enhancing accessibility and consistency at enterprise scale.
• A Peek Behind The CX Forum Curtain: Five Questions With Host Rick Parrish — Human foundations like creativity and trust are crucial for AI to create compelling customer experiences.
• What we lose when artificial intelligence does our shopping — Automated AI shopping agents offer efficiency but risk eroding consumer autonomy, psychological benefits, and social aspects of shopping.
• Red Lobster CEO Plans To Make The Seafood Chain AI-Forward - Black Enterprise — AI integration is crucial for restaurant operational efficiency and strategic recovery, as demonstrated by Red Lobster's turnaround plan.
• AI-Empowered Customer Service, From Hype to Scalable Operations - with Shri Nandan of Comcast — Successful AI in customer service hinges on solving defined customer problems and organizational readiness, not just technology.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by MIT Sloan Management Review.