You may not have realized it, but eBay, a multinational online store, has been using artificial intelligence (AI) and machine learning technologies for over a decade specifically to match buyers with the right sellers. But that’s not the only way they’re using AI technology to improve buying and selling on their site. Devin Wenig, eBay’s president and CEO, wrote in 2016, “I believe that commerce in particular will be the focus of some of the most immediate and exciting applications of AI.” If recent developments are any indication, eBay is aiming to be a leader in implementing AI in online e- commerce.
With an estimated 179 million active buyers, 1.2 billion active listings, and tens of millions of sellers in 190 markets, eBay is where the world buys. Founded in 1995, the company has played an important role in creating an “inclusive and accessible business platform” ever since. Thanks to numerous acquisitions and technological advances over the years, eBay is one of the world’s top 10 e-commerce companies, and in 2018 it hired Jan Pedersen, formerly of Twitter and Microsoft, as vice president and chief scientist for artificial intelligence. The company has used two decades of data statistics and customer behavior to train AI algorithms. These algorithms are getting smarter every time buyers and sellers interact on the platform.
How eBay uses artificial intelligence and machine learning
eBay uses artificial intelligence in personalization, search, insights, discovery and its recommendation systems, along with computer vision, translation, natural language processing and more. They take a four-pronged approach with AI: “enhance the customer experience, build and deploy the right infrastructure for eBay’s unique needs, apply core AI and domain AI to everything we do, and ensure everything is done within the service. customer experience.”
Here are some concrete examples of how they use artificial intelligence.
- Improve the buyer experience: AI-based tools predict buyer needs, recommend products and inspire shoppers. With fine-tuned personalization options, the site strives to tailor the shopping experience to each shopper’s needs. This included the introduction of Interests, an option for shoppers to customize their shopping experience based on their passions and styles.
- Improve your search: eBay uses artificial intelligence to understand the search context and then provide a full spectrum of relevant products. With computer vision, the company was able to add image search to its capabilities and created an e-commerce device for camera phones that allowed buyers to use images as a search query. Artificial intelligence within the app will help shoppers find items similar to what they were looking at.
- Platform as a Service: eBay provides sellers with cutting-edge AI-powered tools so they don’t have to and to level the playing field for sellers who would never be able to compete with others who had the technology. Data, metrics and analytics are managed on the platform and can be used by the retailer.
- Price and Sales Optimization: Machine learning can help sellers identify the best prices, when to list a particular product, and when to promote to better attract buyers.
- Machine translation: Languages are no barrier between buyers and sellers on eBay thanks to the company’s machine translation capabilities. It’s so good that the order processing time for international clients has been significantly reduced.
- Advertising Engine: eBay Promoted Listings and off-site ads are powered by AI algorithms that improve the shopping journey and help buyers discover more eBay inventory.
- Take advantage of deep learning: Even though eBay has been using AI for the past decade, there were still complex problems it couldn’t solve. With advances in deep learning, the company has been able to scale to larger and more complex data sets, ultimately allowing them to process billions of data points and improve their services.
- Augmented Reality to Simplify Shipping: Finding the optimal box to ship your item just got easier thanks to eBay’s Augmented Reality technology in their app. This feature allows you to place a real-world item into a virtual shipping box.
AI in e-commerce
The platform also boasts an image cleaning option for sellers. This feature, which is still in beta, will take a picture and try to automatically separate the selected object from the background visual clutter.
“Search engines these days require a white, clean background on photos in many cases,” said Harry Temkin, vice president of dealer experience.
Sellers, he continued, “often take pictures in very interesting places,” such as on the stairs, in the kitchen or in the garage.
The beta feature will automatically crop the featured item from the photo. Now, manual input is still required in many cases, with users having to swipe around the edges of an object. However, the feature is getting sharper, Temkin said.
It’s a software that’s constantly learning,” he said. The more photos it goes through, the better it will perform.
In addition to its image features, eBay provides native automatic translation that allows buyers and sellers in different countries to see listings in their own languages.
The translation happens behind the scenes, Hamilton said, without users necessarily realizing it’s happening at all.
According to Hamilton, eBay’s model is 5% or 6% more accurate than off-the-shelf products.
“Being a global platform … not everyone speaks English,” Temkin said. “It’s useful to be able to use machine translation to convert an English list to German or Spanish or French.”
