How Home Depot is rebuilding retailing with AI

Amid a turbulent housing market, Home Depot has steadily invested in AI use cases for shoppers and workers.

Over the past few years, Home Depot has been rebuilding its business with more artificial intelligence intended to make shopping easier and workers more efficient. But the home-improvement retailer’s tech-focused C-suite team leading these efforts has also been recently refurbished.

Franziska “Fran” Bell became Home Depot’s chief technology officer in April, after most recently serving as chief data, AI, and analytics officer at automaker Ford Motor. Eleven months before her appointment, 27-year Home Depot veteran Angie Brown ascended to the role of chief information officer. And yet another key technology executive is Jordan Broggi, who became executive vice president of customer experience and the online channel in June 2024.

Some of the top AI applications these executives oversee include an AI assistant called Magic Apron and a customer service AI system built with Google Cloud, the latter recently tested in 50 stores and proving during the pilot program that the voice agents could understand what a customer was calling about in 10 seconds. Internally, Microsoft Copilot has been made available to office workers, Anthropic’s Claude coding system is helping speed up software development, and machine learning algorithms are guiding more efficient workflows for store associates.

Brown says that all the AI investments need to link to one of three core priorities: support merchandising within the physical stores, cultivate an interconnected retail ecosystem that involves digital channels, and grow business with contractors, builders, and other professionals who tend to spend a lot more at Home Depot than do-it-yourself (DIY) shoppers.

And while some technologists have recently aimed to focus their AI efforts on fewer, bigger use cases, Brown says she doesn’t approach her investments with such a restrictive mindset.

“Am I going to limit the number  of use cases that can leverage AI to solve a problem?” Brown rhetorically asks. “I don’t want to. If AI can help solve those problems that we have already identified from a business perspective, I’m not going to hold them back.”

Home Depot, ranked No. 25 on the Fortune 500, is among the retailers that have shown resilient sales even amid a muted economy and inflation fears from the war in Iran that have dampened consumer sentiment. Last month, the company reported that net sales grew 4.8% in the fiscal first-quarter from year-ago levels, though it acknowledged that homeowners were delaying larger projects due to worries about higher gas prices, layoffs, and other economic uncertainties. 

Home Depot and rival Lowe’s must also confront a weak housing market, which has been stung by stubbornly high interest rates and rising building material expenses. These headwinds are particularly inopportune for the spring market, traditionally the busiest for the housing sector.

The company’s top technologists divide up their work by giving Brown oversight of the company’s technology strategy, infrastructure, cybersecurity, and software development. Broggi oversees Home Depot’s $25 billion e-commerce business, merchandising, and the customer experience for digital channels, while Bell steers product management, data, and AI.

One of Broggi’s biggest projects has been Magic Apron, which can answer shopper questions and summarize product reviews and debuted in March 2025. Magic Apron’s generative AI capabilities are trained on Home Depot’s product data and contextualized, but Broggi said that when it launched, “the consumers loved it and the pros hated it.”

Home Depot learned that the web-based Magic Apron system was asking pros questions that were too simplistic. Home Depot pulled the pro version offline and is in the process of fine tuning the large language models for a better user experience for that group of shoppers.

Magic Apron can also field questions from the retailer’s employees. Brown is in the early stages of rolling out the functionality to their smartphones and another upgrade down the road will make the tool multilingual. 

“Generative AI is becoming more and more a part of everybody’s life,” says Brown, explaining why these AI assistant tools are being adopted by employees.

Another AI use case is internally known as “order intelligence,” which looks backward at millions of data points from Home Depot’s past deliveries and assesses a risk score that takes into account problems such as whether the property may require a gate code, or perhaps sits on a winding, narrow path where a 22-foot delivery truck is better than a 36-foot truck. The system can proactively reach out to customers with any potential problems and provide more accurate delivery times.

Broggi says customers don’t know or even care much that generative AI is working in the background. “They just want their stuff delivered on time, complete, undamaged, and with clear communication,” he added.

Another area of focus is developing a generative engine optimization strategy, known as GEO, as consumers spend more time shopping on AI chatbots like Google’s Gemini, OpenAI’s ChatGPT, and Claude. Home Depot allows shoppers to buy its goods on ChatGPT, while also supporting Google’s Universal Commerce Protocol, which advocates for a common language to support agentic commerce.

Broggi says that thus far, the retail strategy for the AI shopping platforms hasn’t been clearly defined. The AI companies developing the platforms have changed priorities a couple of times, he said. “They’ve got to try to figure out how they want to go to market.”

John Kell

Send thoughts or suggestions to CIO Intelligence here.

This story was originally featured on Fortune.com