Your Domino's Order Is Now Being Taken by an AI

You pick up the phone to call Domino’s, and a friendly voice answers. But suddenly you realize: it's not a real person. The system works not because the AI is perfect, but because Domino’s figured out something important: if the voice doesn’t feel right, people will hang up, even if the tech is fine. It’s not just about having cool technology; it’s about who gets the order when the phone rings. The real challenge is more about connecting with people than just the tech itself. 

How the AI Order Comes Together

Domino’s didn’t just create a fancy chatbot; they matched the voice customers hear with the assistant handling their orders. According to Business Insider, t his setup is now used for about 80% of phone orders in North America. The key to their success was making the voice sound like someone who could realistically work at that store.

This is possible by Rime Labs, which focuses on creating incredibly realistic text-to-speech technology. They hone in on regional accents, natural pacing, and even specific pronunciations for menu items (so yes, “MeatZZa” is pronounced just right). This cultural touch makes all the difference, it's what keeps a caller engaged instead of just hitting “operator.” In fact, earlier versions with a generic call-center voice led to a lot of hang-ups, but using a localized voice turned that around completely.

And to manage recognition and dialogue, Domino’s relies on ConverseNow. This system sorts through messy requests, confirms details, and keeps the conversation flowing. A key feature is the human handoff: if the system's confidence drops, the call can be transferred to a human, ensuring the sale goes through. Phones are still important on a busy Friday night, and Domino’s has automated the tricky parts while keeping a backup option available.

Some on Reddit Customers say the calls feel almost human, with natural pacing, no interruptions, and brief “typing” sounds that signal progress. Some people notice the signs (like the typing loop repeating), but they still complete the call because the flow is smooth and the information read-back is spot on. That’s the clever trick: a combination of credibility and momentum always wins over novelty.

Why Domino's Succeeded Where McDonald's Stumbled

Domino’s focused on a specific area and natural delivery. Pizza orders are repetitive and well-defined, making them ideal for specialized models. However, the real breakthrough was cultural: they made the bot sound like a genuine crew member. In contrast, the results in the industry are mixed. For example, McDonald’s ended its IBM voice drive-thru pilot in June 2024, while Wendy’s is planning to roll out its Google-powered assistant to 500–600 locations in 2025. They’re in the same space but achieving different results, because it’s not just about the voice; it’s about having convincing, local conversations.

The Risks involved

When you place AI in a financial context, you also take on the associated risks. Domino’s and their vendor ConverseNow faced a class-action lawsuit in California in 2024, claiming that recordings were used to train AI without obtaining proper consent. This serves as a crucial reminder that disclosure, storage, and opt-out options are not just paperwork; they are integral parts of the product. It's essential to create the user experience and secure permission simultaneously, or it’s best not to launch at all.

The Takeaway

Domino’s didn’t succeed by simply emphasizing “AI”; it won by ensuring that the first moments of a phone call felt local, skilled, and seamless. When the greeting sounds genuine and the information read-back is accurate, the order is completed. In a landscape crowded with chat demos, that’s the only metric that truly counts.