Blog

Nov 28, 2023

Home Depot Explains How It’s Improving On

The Home Depot is applying computer vision technology to “narrow the gap” between its in-stock inventory — when the retailer’s systems indicate a product is in-store — and its on-shelf inventory —

The Home Depot is applying computer vision technology to “narrow the gap” between its in-stock inventory — when the retailer’s systems indicate a product is in-store — and its on-shelf inventory — products on shelf and available for sale to the customer.

The home improvement retailer is “bullish” on the rollout of computer vision technology to help with this, noted Ann-Marie Campbell, EVP, U.S. stores and international operations, during the company’s Q2 2023 earnings call.

Earlier this year, it announced the rollout of a homegrown, new machine learning-powered app called Sidekick to help its associates better prioritize tasks. Part of the company’s suite of mobile devices dedicated toward improved associate and customer experiences (hdPhones), the app communicates the highest demand product, shelves that need restocking, and where excess product can be found on overhead shelves.

“It is critical that we have the right products in-stock in the right quantity and on the shelf available for purchase, particularly for unassisted sales,” noted Campbell. “That’s why you hear us talk about our focus on improving our on-shelf availability, or OSA, positions.”

Computer vision enables technology to provide specific locations of depalletized product that is stored in Home Depot’s overheads, she explained.

“To start, associates will take a picture of bays using their HD phones. These images then feed into our systems and provide a single real-time view of inventory that can then seamlessly integrate into applications like Sidekick. Powered by machine learning, Sidekick directs associates to key bays where OSA is low or out. This helps our teams prioritize the highest-value task inside their respective stores.

“The beauty of the machine learning model is that the algorithm is continuously learning as computer vision images are captured and Sidekick tasks are completed. So it will get better and better at directing our associates to the right bay at the right time. While it’s early days, as we have begun implementing this technology, we have seen meaningful improvements in OSA, increased associate engagement and productivity and higher customer service scores.”

Home Depot beat Wall Street’s revenue expectations Tuesday but posted a 2% year-over-year sales decline. “While there was strength in project-related categories like building materials, hardware, and plumbing, we continue to see pressure in certain big ticket discretionary categories,” noted CEO Ted Decker.

Earlier this year, the retailer announced it’s spending $1 billion to give frontline associates wages in every U.S. market at least $15 per hour. The new wages went into effect on February 6.

“I am happy to share that our approximately $1 billion of annualized compensation investment that we announced earlier this year is having the intended effects,” said Campbell.

“In order to provide the best customer experience in home improvement, we must focus on cultivating the best associate experience in retail,” said Campbell.“This means not only investing in competitive wages and benefits but also providing tools, training, and development opportunities that make working at The Home Depot an endurable and rewarding experience.I am happy to share that our approximately $1 billion of annualized compensation investment that we announced earlier this year is having the intended effects.”

During the quarter, Home Depot continued to see meaningful improvement in its attrition rates, particularly among its most tenured associates, as well as more consistent staffing levels resulting in improved customer service, productivity, and safety, she noted.

“These improvements are exactly what we set out to achieve with this wage investment.”