The company is going through a reformation process that will move primary payment processing from PayPal to in-house managed payments.
And products offered on the marketplace are increasingly being catalogued through company-owned descriptions, images, and specifications to bring more commonality to the platform.
Some have called this process the “Amazonification” of eBay. However, in today’s eCommerce space it seems to make sense bring a shopping experience to consumers similar to other online retailers.
The Power of Image Search
But cataloguing every product is not workable. Long term there are millions of listings with unique designs that do not neatly fall into a specific catalog product.
For these products, eBay appears to lead most of the competition with its continued improvements on its Image Search technology.
When it launched the technology last year, the Image Search feature already did a superb job finding products based on images uploaded by shoppers on their phone.
But eBay hasn’t stood still and continued to improve the technology to stay ahead of the curve.
In a blog post published today, eBay pointed out a few interesting facets of how it is improving image search and how this will benefit buyers.
“It’s important to note that one of the reasons eBay’s computer vision models are so powerful is because we are training the models on varying qualities of images—from professional or stock photos to amateur, dimly lit photos with complicated backdrops. Our dataset consists of 3,828,735 clothing products from 1,219 brands spanning across a wide range of clothing types. These are real-world eCommerce images from our catalog. For every product, we collect an image, title and a set of attributes from which we extract the brand information.”
Hadi Kiapour, Research Scientist for Computer Vision, eBay & Robinson Piramuthu, Chief Scientist for Computer Vision, eBay
The blog post is a real good read. It highlights to sellers that eBay grasps the issues it faces with this technology, which is primarily bad homegrown images.
Unlike other online retailers which mostly rely on “perfect” studio images, eBay’s listings contains millions of smartphone images created in non-studio-perfect environments.
While clean product images will always result in better search rankings, eBay knows many sellers do not have the knowledge, resources, or time to create such images.
By going the extra step with this technology, this seems to be an area eBay is working to level the playing field between smaller and high volume sellers.
READ MORE: eBay Upgrades Their Image Search Tool
If you like to comment on this story, please use the comments section below. Or head over to our Facebook Group for Small Business Sellers and interact with other small business owners.