“Digital-native retailers remain ahead of incumbents’ technology investments. Retailers that fail to innovate with cutting-edge technology will cede share” . Traditional retailers are unquestionably under attack from e-commerce giants like Amazon. Amazon is able to leverage huge amounts of customer data and machine learning techniques to provide a seamless, targeted, and enjoyable experience to online shoppers. But traditional retailers aren’t out of the game yet; in particular, Walmart is utilizing machine learning to create its own competitive advantage in an effort to fight back. Walmart “is on the cutting edge when it comes to transforming retail operations and customer experience by using machine learning” , and “is increasingly becoming a technology company, seeing its transformation into an omni-channel retailer as critical to its continued growth” .
A retailers’ product offering is effectively a buying experience, and established brick and mortar retailers “insist that their physical operations give them an advantage e-tailers can’t match .” This advantage is called omni-channel retail, for which Walmart provides the definition: ”We want to ensure a seamless experience between what customers do online and what they do in our stores” . Walmart’s retail footprint generates huge amounts of data, and machine learning can be applied to this data to create engaging omni-channel products that e-tailers don’t have the option to offer. “A wealth of data…is now available; thanks to machine learning, retailers and category managers are now able to analyze not only structured sales history but also unstructured data” . Machine learning is an important mega-trend for Walmart because it allows for omni-channel product development that will allow them to compete against Amazon for the future of retail.
Walmart has recently rolled out numerous product offerings that required machine learning for development. For example, Walmart is encouraging shoppers to order online and pick up in-store, which typically results in add-on sales (a typical benefit of omni-channel presence.) They have used machine learning to offer targeted discounts on online-only merchandise when shoppers agree to pick up in-store . They have also installed towers in their retail locations where customers can scan their online order receipt to have a conveyor belt deliver their order within 45 seconds, the logistics of which are powered by machine learning algorithms . Walmart is improving the customer shopping experience with “Scan and Go Shopping.” Customers can us the Walmart App to complete some aspects of the checkout process before they reach the counter, resulting in a more streamlined, time saving process. Machine learning is employed in both computer vision and sensors for security to make this possible; future developments could bypass the in-store checkout process entirely . Walmart is also using machine learning to enhance their home delivery service product by optimizing the delivery routes of their associates .
Walmart is taking a long-term stance on the competitive advantages that machine learning brings. They are developing facial recognition technology (powered by machine learning algorithms) to identify unhappy or frustrated shoppers. “As the machines learn to identify different levels of frustration through the facial expressions of those in line, it could trigger additional associates to run the checkouts and eventually could analyze trends over time in a shoppers’ purchase behavior” . Walmart has made large investments in Silicon Valley: their “Store No. 8” venture launched recently, with a mission of “creating proprietary next-generation robotics, virtual & augmented reality, machine learning and artificial intelligence technology” .
I would recommend that Walmart take some additional steps to address the e-commerce competitive threat. Walmart could further capitalize on growing IoT technology by integrating tags (like RFID systems) on products to monitor use, stocking level, and location within a household. For instance, if a tag reader was installed in a refrigerator, Walmart could scan everything placed inside and let customers know not only when to restock but also when items are due to expire. RFID systems could monitor how often customers pick up products like laundry detergent, predict how much is left of their stock, and automatically place the item on a personal shopping list in the Walmart App . IoT leverage could help create personalized advertising and expand cross-selling activities. Additionally, partnering with a company that provides a voice assistant, like Google, would allow an expansion of Walmart customer service and product advertising. These devices, driven by machine learning to become “smarter” over time by ingesting large quantities of data, could be taught to make tailored recommendations of Walmart products that would fit individual customers’ needs .
In closing, I will posit the following open questions: first, can Walmart recruit the necessary technical talent to manipulate the machine learning required to offer these future products? And, secondly, does Walmart’s core customer demographic (lower-middle class families) actually want these omni-channel “bells and whistles,” or is Walmart’s real goal to bring in a wealthier shopping demographic?
 Bloomberg Intelligence analysts Anurag Rana, Poonam Goyal, Gili Naftalovich and Morgan Terrant, “Artificial intelligence speeds-up disruption in retail,” Bloomberg Professional Services (February 01, 2018).
 Bernard Marr, “How Walmart Is Using Machine Learning AI, IoT and Big Data To Boost Retail Performance,” Forbes (August 29, 2017).
 Mass Market Retailers, “Walmart’s E-Focus,” Volume 35 No. 2, (January 29, 2018).
 Jennifer Marks, “The Battle for Market Share,” Home and Textiles Today, (September 4, 2017).
 Bizcommunity, “The benefits of machine learning for retail,” SyndiGate Media Inc., (May 23, 2017).