DataShopping with images: Ebay’s visual search tool


Did Kim Kardashian’s latest outfit have you literally drooling, but don’t know where to get it? Or maybe you’re scrolling through Pinterest and spot some decorations that would be perfect in your own living room, but don’t know where to buy them? As they say, a picture is worth a thousand words. And that’s what eBay thought too. It’s estimated that by 2020, 50% of searches will be done through images. Over 75% of consumers already say that they use photos they’ve seen online as inspiration for their purchases.

The power of images: do they make shopping easier? 

EBay — who recently got a facelift with bright colors, modern elements and a font family all its own — launched a new feature last month: image search.

In a world dominated by social communication, retailers have started adopting visual solutions to adapt to the needs of their users, who are more and more accustomed to reacting in real time. Platforms that rely on visual search technology will find that they get greater customer loyalty, as well as quicker responses compared to the input they receive. The growing popularity of this technology among retailers is also thanks to better accuracy compared to traditional word-based searches.

How to use eBay’s visual search tool

There are 2 tools for image searches that are debuting on the pioneering web platform: Find it on eBay and Image Search. Both features, which will be integrated in eBay’s native app, were announced last July, and are so far living up to their high expectations.

Find It On eBay lets you start your search on any social platform — like Facebook — or while you’re browsing your favorite influencer’s blog. Just click on the share tool and the app will show a list of similar products related to the image that caught your eye. The Image Search feature will be within the eBay app: you’ll just need to click on the camera icon in the search bar to begin a search based on an image in your camera roll or one you’ve just taken.

These new features are still in the process of being rolled out. Image Search is already available for Android and iOS, while Find it on eBay is only on Android.

Deep learning and the convolutional neural network

Ebay’s new tools are building on the latest advancements in artificial intelligence: deep learning and computer vision. The company has reported that model they used is the convolutional neural network, which turns the image a user uploads into a representation that can then be compared with other images in eBay’s database.

The downside of this kind of technology however is that the key customers in this social era — millennials — are looking for a certain brand, not just a product that visually resembles another. It doesn’t matter that Olivia Palermo’s basic t-shirt is essentially the same as dozens of other ones picked out by artificial intelligence. Millennials aren’t just buying the esthetics of the product, they’re buying the brand.

Not just Ebay, it’s Asos, Google and Pinterest too

The most important competitor for any retailer today is undoubtedly Amazon. In their drive to make sales and gain a bit of a margin, many retailers are trying to combine the experience of in-store shopping with the convenience of doing it online. With this in mind, midway through last year, Asos had introduced the possibility to search for outfits using photos instead of keywords. Google and Pinterest have also started using visual search technology to facilitate use of their applications. Pinterest in particular has introduced several functionalities: there’s Pinterest Lens, Shop the Look, and the image recognition feature has been extended to the ads on their site.

EBay is at a technological advantage due to the rich set of user-generated images and item data we’ve accumulated through the years, says Mohan Patt, VP of Buyer Experience at eBay. “By applying machine learning technology, eBay can deliver a fast and reliable shopping experience backed by one of the world’s largest commerce data sets.”

Nadeesha Dilshani Uyangoda

Editor freelance, blogger, attivista, studentessa. Lettrice ossessiva e scrittrice compulsiva. Italiana per cultura, srilankese per eredità — sempre a metà tra due mondi. Millennial che preferisce la carta, ma si adegua al digitale per paura dell’estinzione.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.