Due to its extensive network of around 4,700 Walmart stores and 600 Sam’s Clubs in the United States, which collectively employ about 1.6 million individuals, and its vast data sets, Walmart has gained a distinctive advantage in utilizing the potential of artificial intelligence (AI) to provide improved customer shopping experiences.
According to Anshu Bhardwaj, senior vice president of tech strategy and commercialization at Walmart, the company’s AI strategy centers around utilizing the technology and machine learning to enhance the customer and employee experience within its vast business environment.
Sam’s Club, for one, leverages its membership model to gather massive amounts of data on customers’ search queries, purchase history, and preferences which allows the retailer to improve its shopping experience, as customers are constantly leaving “breadcrumbs” that reveal their likes and needs, according to Bhardwaj.
To ensure that the desired products are available for purchase, Sam’s Club monitors roughly 6,000 items stored on shelves in warehouse-style stores averaging 136,000 square feet. To do that, Sam’s Club has devised an unusual solution: floor scrubbers.
As these machines move around the store, cleaning floors and keeping them tidy, they also capture real-time images of every item present. Each store has one of these scrubbers, which are equipped with inventory intelligence towers capable of taking over 20 million photos of all the products on the shelves daily.
“What this means is that I can distinguish Kellogg’s Froot Loops from Kellogg’s Frosted Flakes and the depth in which they’re stocked on the shelves,” said Bhardwaj. “And if you think about a shelf, it doesn’t always have items in the front. They could be staked in the back, and then there are shadows as well.”
That said, with some training, the algorithms are able to accurately recognize different brands and their stock placements while taking into account factors like light intensity and shelf depth, achieving an impressive accuracy rate of over 95%. Once a product reaches a certain threshold, the system sends an automatic alert to the stock room to ensure that the item remains in stock.
The AI-powered system can also determine the best course of action when a product is out of stock in the stock room but delivery is scheduled to include the missing item on that day. In these situations, the algorithm instructs the associate to take the product directly to the sales floor instead of the stock room. According to Bhardwaj, this approach has helped boost employee productivity by 15% since its implementation last year.
The Walmart shopping app also utilizes AI to enhance the customer experience. An example of this is the system’s ability to recognize when a customer has ordered Pampers before and to determine if the previously ordered size is still appropriate.
Bhardwaj noted that while Walmart’s AI journey spans several years, the company’s ultimate objective remains: to identify more effective ways of determining what customers want to purchase and how to deliver it to them.
That is a goal AI is helping achieve with simplification of the decision-making process, resulting in less friction between customer preferences and their ultimate purchase.