A small town on New York’s Long Island is the unlikely destination for one of the most forward-thinking experiments on the future of artificial intelligence and retail.
In April, Walmart unveiled its new Intelligent Retail Lab in Levittown, New York, a 40,000 square-foot trial of AI-enabled technology designed to improve the store experience. Using thousands of cameras, combined with other technology like sensors on shelves, Walmart will be able to recognize when specific products are running low and automatically issue alerts to restock shelves when employees are needed most.
It’s a fancy solution to one of retail’s biggest problems, recognizing that the ideal customer experience involves having the right product at the right time at the right place.
However, with all of these futuristic advances with respect to in-store technology, as well as automation that reduces the friction of getting goods to the consumer, it’s important to note the danger of deploying these tools on an unstable foundation.
Indeed, an in-store AI application like Walmart’s IRL only works if retailers can accurately predict demand and manage their supply chains appropriately. Without corresponding improvements to retailers’ pricing and promotions systems, which may be decades old, this added technology runs the risk of introducing added pressure to an aging supply chain.
Many retailers struggle to accurately forecast their inventories at the store-level, especially during heavy promotion campaigns, and are years away from implementing visual AI into their stores.
The more practical, scalable, shovel-ready application of AI for retailers is in the supply chain. Using AI to predict consumer demand, retailers can order only the amount of merchandise that they can sell and allocate their inventory more effectively. Doing so can reduce stockouts and overstock situations for significant revenue improvements.
A More Practical Use of AI for Retailers
Capturing more than 1.6 TB of data every second – with enough connecting cable to scale Mt. Everest five times – Walmart’s IRL store is a testament to the company’s technological prowess and vision for the future. It naturally draws comparisons to Amazon Go convenience stores, where a combination of AI-enabled cameras and shelf sensors work together to automatically checkout customers.
But in either scenario, processing that much data requires a huge amount of processing power. In Walmart’s case, the company had to build a massive onsite data center to support IRL’s data load with more than 100 onsite servers.
To update all of its nearly 5,000 stores nationwide with the same technology would be a massive investment. It’s unlikely the average shopper will see the IRL store’s technology at their local Walmart anytime soon given the upgrade cost, and it’s unclear how consumers will react to AI-enabled, in-store tracking over the long term.
That doesn’t mean that AI won’t be there working behind the scenes to improve their experience, though. Artificial intelligence and machine learning technology are already helping some retailers improve their supply chain management without the need for big investments in IRL technology.
For example, cloud-based AI solutions focused on retail’s most complex business processes around pricing, promotions, fulfillment, merchandising, loyalty and personalization can be deployed in months. This already-available technology is scalable across locations and can harness customer data from all ERP and order management systems.
By analyzing millions of data points, retailers can gain greater confidence in their demand forecasts down to the chain, store and SKU level while decreasing the frequency of stock-outs. Interestingly, in Walmart’s IRL case, more accurate demand forecasts would complement their in-store AI inventory management by ensuring employees have enough product on hand to fill empty shelves.
In addition, business leaders can use cloud-based AI tools to test the introduction of new promotions and products on existing inventory to determine overall value and avoid cannibalization.
Bottom line: Amazon Go and Walmart’s IRL store may offer a glimpse into the future of retail, but executives don’t need to spend millions upgrading their stores to take advantage of cutting-edge AI technology. The immediate value of AI is in the supply chain and operations processes.
It may not be as visible to consumers, but retailers will surely appreciate its impact on their bottom lines.
Kerry Liu is the Co-Founder & CEO at Rubikloud where he has led the company to become one of the world’s largest AI platforms for large retailers. Founded 5 years ago, Rubikloud has grown to over 100 people, opened offices in four countries, and raised venture funding of $45m. Kerry works to manage and maintain a thriving company culture that recruits the brightest talent in the industry, while also maintaining relationships with global retailers and investors. Kerry is also actively involved in the start-up ecosystem as an advisor and angel investor.