We’re pleased to announce IBM watsonx Orders, an AI-powered voice agent for drive-thrus of quick-service restaurants. Powered by the latest technology from IBM Research, watsonx Orders is designed to help restaurant owners solve persistent labor challenges by handling almost all orders and interactions without the help of human cashiers, while delighting restaurant guests with quick service and accurate orders. 

Watsonx Orders joins IBM’s watsonx family of AI Assistants that help you easily deploy and scale conversational AI to maximize revenue and lower costs. It’s finely tuned to understand fast food menu items, customization options, limited time offers and promotions. Integrated into the corporate and restaurant-level points of sale, watsonx Orders can tell customers what’s available—and what’s not—at that moment. And it routes final orders to the cash register and screens and printers in the kitchen.

Across the industry, more than 70 percent of the revenue for quick-service restaurants comes through drive-thrus. Operators consider the drive-thru the most important sales channel by far, even more so than sales inside the restaurant. What’s more, the order taker position in the drive-thru is one of the most difficult and stressful among the crew—which is why it’s also one of the most difficult positions to fill for every shift. So when restaurant owners can’t find someone to fill the order-taker position, either because of unfilled job opening or absenteeism, they face disastrous revenue drops for shifts or whole days.

That’s why we designed watsonx Orders to be considered a valuable crew member: accurate, polite, speedy and reliable 24/7.

How watsonx Orders works

Watsonx Orders is a combination of hardware and software installed in the restaurant. It understands the full menu of the restaurant chains and knows which items are available at that specific restaurant by season and time of day. It knows the nicknames of every item and how customers can customize them. And we continuously update the software to be aware of limited time offers nationwide or locally.

Upon guests pulling up to the outdoor menu board, watsonx Orders greets them. Customers can start the conversation by saying, “I’m picking up a mobile order,” or “I’m a rewards member.” Or then can start ordering: “I want a cheeseburger, chocolate shake, large fries, no onions.” Thanks to watsonx’s natural language processing capabilities, the agent asks the customer to specify the size for the shake, and knows that the “no onions” is a customization option for the cheeseburger. 

Unlike many AI voice agents, watsonx Orders doesn’t convert each spoken word into text for analysis. Instead, it converts audio into phonemes—the individual sounds that make up words, such as the “sh” in “shake”—then into orders, vastly speeding up speech recognition. We’ve also used generative AI techniques to think of the hundreds of millions of ways that humans can express the thought “I’d like to order a hamburger.” We use the latest in neural net technology to block noise from freeways, airports, railroads, stereos and cars in the other lane.

On the backend, we connect watsonx Orders to the restaurant’s technology stack. Once we’ve gathered the order ticket, we convert it into commands that the point of sales (cash register) understands, and we send the order to the kitchen staff through displays and printers.

IBM makes and installs all the hardware and software itself for a smooth delivery to the restaurant owner.

The benefits of watsonx Orders

In the field, watsonx Orders is as accurate as a human order taker. But it can work every shift, every day, and every holiday. As such, the restaurant owner gains the equivalent of more than eight hours of labor every day, freeing up the crew to focus on delivering even better customer service. 

Just as importantly, watsonx Orders takes orders quickly—it speaks in a cheerful, polite voice, knows how to keep the conversation short, and helps the customer toward a decision without being pushy. Restaurant owners have told us speed of service and car counts per hour are among the most important measures of operations. So our team is constantly experimenting to shave split-seconds off each order without sacrificing accuracy. 

The system also knows how to upsell or sell additional items to customers for additional revenue with each order.

Should something go wrong, or if the guest insists on speaking with a human, the AI agent can gracefully hand the conversation over without any need for customers to repeat themselves. 

Get started with watsonx Orders

IBM continues to drive innovation for watsonx Orders to achieve the highest level of accuracy and fast service at every restaurant. We are interested in working with the corporate offices of quick-service restaurant brands to understand their pain points and customize the product to meet their specific needs.

Improve your drive-thru operations with watsonx Orders
Was this article helpful?
YesNo

More from Artificial intelligence

Responsible AI is a competitive advantage

3 min read - In the era of generative AI, the promise of the technology grows daily as organizations unlock its new possibilities. However, the true measure of AI’s advancement goes beyond its technical capabilities. It’s about how technology is harnessed to reflect collective values and create a world where innovation benefits everyone, not just a privileged few. Prioritizing trust and safety while scaling artificial intelligence (AI) with governance is paramount to realizing the full benefits of this technology. It is becoming clear that…

Taming the Wild West of AI-generated search results

4 min read - Companies are racing to integrate generative AI into their search engines, hoping to revolutionize the way users access information. However, this uncharted territory comes with a significant challenge: ensuring the accuracy and reliability of AI-generated search results. As AI models grapple with "hallucinations"—producing content that fills in gaps with inaccurate information—the industry faces a critical question: How can we harness the potential of AI while minimizing the spread of misinformation? Google's new generative AI search tool recently surprised users by…

Are bigger language models always better?

4 min read - In the race to dominate AI, bigger is usually better. More data and more parameters create larger AI systems, that are not only more powerful but also more efficient and faster, and generally create fewer errors than smaller systems. The tech companies seizing the news headlines reinforce this trend. “The system that we have just deployed is, scale-wise, about as big as a whale,” said Microsoft CTO Kevin Scott about the supercomputer that powers Chat GPT-5. Scott was discussing the…

IBM Newsletters

Get our newsletters and topic updates that deliver the latest thought leadership and insights on emerging trends.
Subscribe now More newsletters