How Fashion AI Transformed Two Online Retailers
YesPlz.AI, February 2021
Kolon Mall is one of the top 3 luxury fashion retailers in Korea, with thousands of high-end products available from both their in-house brand and outside, licensed brands. Shoppers can experience well-curated fashion collections on the website, such as highlighted new brand collections and weekly picks.
Kolon Mall is known for being an online retailer for fashion-forward customers. Because of the cut-throat Korean market for high-end fashion, Kolon Mall needed an innovative, integrated solution to revamp customer search and experience and stand out from competitors.
Kolon Mall partnered with YesPlz to transform their product recommendation system on their website.
Based out of Korea, Lately is one of the fastest growing online fashion retailers in the B2C space. Customers can find clothing, beauty items, shoes, and accessories from domestic and international merchants. Mom and pop stores sell their products through Lately’s online platform.
Customers love the wide range of items available on the website. Lately offers over 200k clothing items, and has over 5,000 daily updates to these products. With an ocean of products available, Lately needed a way to keep up with their constantly changing inventory and provide customers with a search solution to easily find products.
Lately and YesPlz partnered together to re-imagine the customer search experience.
Customers can’t find what they’re looking for using text search, leading to missed revenue opportunities for brands.
It takes hours to tag new product updates on a daily basis--and the information provided by retailers to Lately and Kolon Mall is often inaccurate, leading to even more time spent updating product information.
With other search solutions costing thousands of dollars per month, retailers are overpaying for search solutions that are inaccurate and don’t solve the problems above.
YesPlz implemented a two-step solution for Lately and Kolon Mall, targeting the problems of non-intuitive search, product tagging, and lack of new product discovery.
Both Lately and Kolon Mall decided to implement the Style Filter, a visual search tool where customers can visually demonstrate the specific product attributes they’re looking for, and a fashion-trained AI algorithm will show hundreds of relevant results.
The Style Filter filled the gaps where text search falls short. Instead of trying multiple keyword combinations to receive an intended search result, customers simply needed to show their intended style on the model, and fashion-trained AI finds relevant products.
Search Results for “Black Wide Cropped Pants” Using the Style Filter
Lately and Kolon Mall’s text-based filters returned sweaters, t-shirts, and vests for the search term “black wide pants.” Customers searching for a specific product type and fit and receiving inaccurate search results are unlikely to repeat their search, and instead, bounce from the retailer’s website due to a poor customer experience.
With the YesPlz Style Filter, customers can demonstrate the length, fit, and color of the pants using the model on the left side of the page, and receive results that are catered exactly to their search intentions. Because the search journey is simplified, customers are more likely to continue to purchase products and move seamlessly throughout their purchase experience.
YesPlz AI Visual Style Filter--Personalize the Search Experience
The Style Filter completely eliminates the need for a text-based search filter, also alleviating the problem of individually tagging each new SKU. Lately and Kolon Mall can manage their products more easily and efficiently.
Kolon Mall added YesPlz AI-powered product recommendations to their website to show customers more relevant products.
YesPlz AI product recommendations identified key product attributes to find relevant products to suggest to customers. As a result, customers are discovering new brands that previously weren’t coming up, solving the problem of brand insulation (customers repeatedly choosing the same brands). In addition, lesser sold products are shown in AI recommendations. However, these products are highly relevant, so as a result customers are leaving with larger overall basket sizes and retailers are selling older inventory--a win for everyone.
YesPlz AI Similar Product Recommendations
Whether customers are fashionistas or non-fashion savvy, it’s nearly impossible to guess which text keywords will lead to accurate search results, as product names differ across retailers.
Customers from Lately and Kolon Mall tend to return to brands with which they’re familiar to avoid a complicated and time-consuming search experience. But, customers aren’t discovering new brands, leading to lost revenue opportunities and smaller overall basket sizes.
By implementing the YesPlz Style Filter and similar product recommendations, both retailers and customers can have their needs fulfilled. The YesPlz Style Filter streamlined the search experience for customers, making it easier for customers to explain exactly what they’re looking for, without a long-string of keywords. When customers find the products they’re interested in, they continue to search for other products, return to retailer websites, and check out with larger overall baskets. From the point of view of the retailers, Lately and Kolon Mall were able to facilitate new brand and product discovery, leading to more purchases and more satisfied customers.
Fashion trends move fast in Korea, which means that product tags need to constantly be updated to keep up. Retailers looking to stay on top of their product tagging are left with limited options: either manually update product tags/keywords or pay for an expensive, uncustomizable search solution.
Both Kolon Mall and Lately are tasked with updating new products on a daily basis, and often receive incomplete or inaccurate information to accompany the product. Lately works with small retailers that provide large daily updates, which can take hours to tag. Kolon Mall faced an additional problem: the need for micro-customization in its tags, to include details such as the exact fabric of a jacket (quilted, fleece, wool).
Combine the task of individually tagging product attributes with thousands of SKUs, and the problem is not only overwhelming--it’s nearly impossible for retailers to keep up with.
The YesPlz Style Filter overcomes this problem by altogether eliminating the need to tag products. Because the YesPlz Style Filter uses artificial intelligence, those product attributes are automatically recognized and catalogued, making the lives of retailers that much easier. Retailers with large SKUs can rest easy knowing that these processes are now automated to their specifications--and can be customized to a level of micro-customization, depending on specific needs (such as Kolon Mall). As a result, both Kolon Mall and Lately have saved valuable time wasted on product tagging--and have a more organized system for managing the details of product attributes.
When looking for a dedicated search solution, Lately found that current search solutions were based on the number of searches per month, ironically penalizing retailers for having high search volume.
Even at a high price point, current search solutions didn’t solve the problem of constantly needing to update product information in the search solution based on the ongoing influx of new SKUs. Building their own search solution from scratch was expensive, and still didn’t provide a way to keep up with changing SKUs.
By utilizing the Style Filter and YesPlz Product Recommendations, Lately and Kolon Mall were able to create the best-possible online shopping experience for customers. Through an AI solution that is specifically trained to understand key customer attributes in fashion, both Lately and Kolon Mall provided customers with a unique shopping experience, filled with relevant product recommendations and easy-to-understand search.
Request a free 30-min consultation here
To learn more about YesPlz, check out our Resource Center
Too many retailers are stuck with static, expensive product recommendation systems that aren’t delivering accurate, fashion-forward recommendations to customers. That’s why we decided to create the YesPlz Product Recommendation Engine.
It's hard to know what customers are really looking for, and the fashion taxonomy can complicate the search process even further. In this piece, we demystify the fashion taxonomy and explain how understanding customer search intentions can shape your eCommerce strategy.