Blue Yonder Uses AI to Optimize Inventory, Pricing
Table of contents
What do high-energy physics and retail supply chains have in common? The heck if we know. But a German company called Blue Yonder, founded by a scientist who worked at Europe’s giant particle laboratory CERN, is using big data and artificial intelligence to help retailers optimize inventory and product pricing in impressive ways.
About Blue Yonder
Founded in 2008, Blue Yonder has taken in $75 million in private equity from Warburg Pincus. The technological brains behind the operation is Dr. Michael Feindt, who developed the NeuroBayes algorithm to record high-energy collisions in the Large Hadron Collider. The machine learning platform has been adapted to help retailers—both brick-and-mortar and e-commerce—to avoid costly explosions in inventory and find the sweet spot for pricing products to move off the shelf.
It shouldn’t come as any surprise that retailers are increasingly turning to AI. Mega-retailer Amazon is certainly leading the charge, employing about 1,000 employees whose jobs revolve around developing more sophisticated algorithms to sell us stuff we didn’t know we even wanted or needed. It’s not just about scrutinizing a customer’s behavior to custom fit a shopping experience or using AI to develop product reviews. Amazon, for example, is improving natural language processing to answer more complex customer questions and to detect fake product reviews. Dr. Freindt himself recently wrote about the AI trend in retail:
Retailers who are still discussing the benefits of machine learning and AI tech are already behind the curve—it is here, changing the industry today. It is a technology that has the potential to significantly change how retailers and consumers interact and purchase goods.
Blue Yonder Goes Beyond Historical Data
So what makes Blue Yonder’s AI platform smarter than the average inventory software? Its machine-learning algorithm makes forecasts that go well beyond historical sales data. It factors in everything from advertising to local weather forecasts to public holidays. It claims its solutions can reduce out-of-stock rates for its customers by as much as 80 percent and increase profits by more than 5 percent.
For example, for online retailer OTTO, Blue Yonder’s Replenishment Optimization platform began by analyzing three billion transactions from past sales, prices and stock levels. It then considers about 200 factors that could influence sales to make its predictions. OTTO has applied the technology for the third-party products it carries. It measures success by selling out of that inventory in 30 days or less. Blue Yonder’s algorithms get that done 90 percent of the time. By turning the job over to AI, OTTO also saw a drop in delivery times from as many as seven days down to one or two days.
In another case study, grocer Morrisons implemented Blue Yonder’s Replenishment Optimization platform for nearly 500 stores. That meant automating about 13 million ordering decisions, which reduced shelf gaps by 30 percent.
Blue Yonder’s platform isn’t just about keeping inventory fresh. It also applies machine learning to optimize prices for each product, particularly for fickle seasonal items such as women’s fashionwear. Earlier this year it also rolled out a new solution that extended its inventory optimization to distribution centers.
Beyond Blue Yonder
One of the best parts about shopping online, aside from the fact you can do it in your underwear while drinking beer, is that if you don’t like whatever plastic Made-in-China tchotchke that you bought, you can mail it right back—usually free of charge. That’s great for the customer but cuts into the e-commerce retailer’s bottom line. Just for the record, Americans reportedly return more than $260 billion in stuff each year. If only there was a way for retailers to forestall the inevitable return.
That’s the mission behind SupplyAI. The small San Francisco startup has a modest $350,000 in funding but only came onto the scene in 2015. It claims that its proprietary algorithm called ReturnSense can detect patterns in purchase behavior and predict the likelihood of a return on each purchase. It alerts its clients and sends a message to customers that makes them validate a purchase before anything is shipped. Having studied the customer’s behavior, the platform also presents the customer with additional options that may be a better fit (literally and figuratively). Clients reportedly include Macys and Kohls, which we were surprised to learn still existed.
One can be forgiven for thinking it’s mostly e-commerce retailers that are embracing the AI revolution. After all, you’re not going to replace a kindly, grandfatherly store greeter who doesn’t know where anything in the store is located with a robot unless you have a heart blacker than Voldemort’s. We found the next best thing—a smart tablet that helps customers navigate stores while feeding real-time inventory data to retailers.
That innovation comes from California-based Focal Systems, which has raised $2.62 million from a couple of Seed rounds, the most recent coming from last May. The Focal Tablet reputedly uses the same sort of machine learning and computer vision technology employed by Amazon Go and Google’s autonomous vehicles. Attached to the front of a shopping cart, the Tablet displays deals available wherever shoppers are currently in the store. It also targets digital ads based on the items already in the customer’s cart. At the same time, the Tablet detects gaps in inventory, which the company claims reduces stocking time by 50 percent. The startup is working on developing the capability for the Tablet to tally the contents of the cart and complete the checkout on the spot. The Tablet reminds us a bit of a piece of retail analytics tech developed by a startup called Nomi that we covered a couple of years ago before AI became trendy.
Update 01/03/2022: Focal Systems has raised $25.8 million in Series B funding to continue its rapid pace of growth and innovation with their AI-powered retail automation solutions, as well as expand its global sales and technical teams. This brings the company’s total funding to $67.7 million to date.
Retailers outside of businesses like Amazon, which itself is becoming as much a technology company as it is a retail store, have been a bit slower to embrace technology than other sectors of the economy such as financial services. Startups like Blue Yonder, SupplyAI, and Focal Systems seem to be among the first wave to employ machine learning and other forms of AI technology to an industry where the self-checkout counter still seems cutting-edge (and still buggy).