“Fashion is visual, not numbers or text”- Artificial & Visual Intelligence in fashion eCommerce
On the 22nd November, Luis Manent, CEO & Co-Founder at Wide Eyes Technologies, was part of the panel of tech-experts at EcommFashion, the unique event in Spain that connects fashion ecommerce managers with startups, to discuss the impact of AI in the fashion industry. A dynamic talk took place where our CEO Luis Manent analyzed the eCommerces and apps of the big industry players like H&M, Amazon, Asos and Privalia to show how artificial intelligence can help fashion retailers in providing smooth shopping experience to its users and maximize sales.
Mobile shopping, today
Mobile’s influence on shopping behavior has been on the rise. Now more than half of all internet traffic coming from a mobile device and an estimated 60 percent of all online purchases in Black Friday came from smartphone users. In Spain, one out of tow. That puts even more pressure on brands to improve their mobile and/or app experience.
Mobile Shopping is the future (and present) of online purchase experience. In this sense, the mobile search engines also have been evolutioning, adapting to the user’s necessities and the new tendencies. “The camera is the new keyboard”, stated Luis Manent.
Search by image experience
“Search by image is now a common feature on leading retailers”, he said. Players like H&M or Asos are offering the ability for shoppers to shop products by style by uploading an image. It is a step forward towards creating ease and comfort for the users to find products, taking shoppers directly from inspiration to satisfaction. In this sense, the CEO of Wide Eyes, stated that “the real revolution of visual search is on one-click discovery“. Retailers need to take visual search solution to the next level to delight capricious customers and shoppers needs with the real ‘shop the look’ service.
The one-click discovery revolution is presented by image recognition technology like Wide Eyes’ Shop-The-Look solution which detects all products on image (not only one item) automatically, without human interaction. Users really can shop their favourite influencer’s outfits (not only one item), recreate some of the most stylish looks and steal style trends from uploading images.
Smart Recommendations on PDPs
The way the products are presented to the customer is a key factor to increase online sales. The product recommendations (or suggestions) systems on PDPs help customers to access automatically a greater number of products, and also to navigate the product catalog in a more visual, intuitive and attractive way, explained Luis Manent. In fact, 35% of Amazon’s revenue is generated by its recommendation engine. So, are you showing the right recommendations to your products?
Is there a magic formula? Maybe the success lies in showing relevant products based on recently viewed, inspired on browsing history, bestsellers o suggest outfit recommendations…. Probably there is no magic formula. But after analyzing the masters of fashion like Zara, Asos and Privalia, we can intuit which solutions can work better in fashion and why. They all offer visually similar recommendation, concluded the co-founder of Wide Eyes Technologies.
The customization of recommendations based on visually similar products is very effective to provide the right content to areas like ‘you may also like’ and keep the customer journey going until they find what they’re looking for. Also, in hight-impact campaigns like Black Friday, the similar recommendation system is essential to solve out-to-stock automatically, offering users visually in-stock similar products as alternatives. Luis Manent stressed at EcommFashion that “we need recommendations that create meaningful paths”.
Only smart organizations are using Visual AI technology to enhance the SEO of your product catalog. Now the Fashion ecommerces offer different options of search: by image, by text, by voice, … But search engines always deliver product images. Hence the importance of assigning advanced and customer driven tags and keywords to all product images of catalogue to improve discoverability and search results quality is undoubtful, explained Luis.
In this sense, Wide Eyes auto-tagging API can automatically assign relevant tags to every trending item that appears in the image, including tags that make the difference based on visual characteristics; because details do matter in fashion.
Probably these are the most used (and demanded) artificial intelligence solutions for fashion retailers. In the end, “fashion is visual, not numbers or text” and “customers should spend less time on searching and more time on buying”, assured Luis Manent.