THE FUTURE OF FASHION RETAIL WITH AI.

Credit by: Mohit Agrawal



Artificial Intelligence (AI) is having a profound impact on all the industries and the way they operate. Fashion industry is no different! Opportunities to improve the customer experience & retail operations are immense and have a direct impact on top line and bottom line of the companies.

Fashion industry has myriad stakeholders and influencers right from designers to fashion houses to bloggers to celebrities and finally the customer herself. The trends and shopper preferences complicate the entire supply chain and retail operations. The complicated relationship between shopper preference and supply chain management has made the retail operations more of an art rather than science. Artificial intelligence can integrate this diverse set of data into engaging and actionable information.

Many industry players like Nordstrom, Shoe Dazzle, Sephora, etc. are investing in AI to suggest styles to their shoppers personalizing the shopping experience in the process. Start-ups like Tuzo are helping the retailers in taking the guessing out of the game. The insights from the shopper preferences can be used for better demand planning and product sourcing.

The fashion retail market is roughly half a trillion US dollars in Asia Pacific (excluding China) and half of this market is addressable market for AI personalization which makes it a very lucrative market.

Fashion Industry & AI: A few Use Cases


Automated Tagging: creating product catalogs is a time consuming manual process. The content writers are hired to tag the products and write descriptions hoping that the shoppers would be able to find the right product and like the way they have described the product. However, the content writers are not creating exhaustive list of tags and do make mistakes leading to poor SEO and visibility on 3rd party marketplaces. With computer vision, the process can be automated where the computer not only provides an exhaustive list of error free tags, but AI also writes the description in a way that it is likely to be more appealing to shoppers.



Personalization: who does not want shopping fast and efficient? Nobody wants to sift through large pile of irrelevant clothes before getting something of interest. There is a lot of data that is available about the shopper like the past purchase, browsing and click stream data that can be used to sort the storefront in the order of shopper preference. The shoppers may be asked to rank or choose a few images to get further insights about her preferences to strengthen the recommendations. Recommendations can be further enhanced by considering the body shape, height, etc. to recommend what would look good on the shopper. Various implementations of personalized recommendations have shown the conversion increase of up to 40% and increase in Average Order Value (AOV) by up to 50%.


Trend Tracking: There are many blog sites, fashion magazines, celebrity pictures and influencers on Pinterest and Instagram. Programs can be written to look at all the available sources of information and recognize patterns in terms of fashion trends.

The global, regional and local fashion trends and choices can be juxtaposed with the shopper preferences to give amazing style recommendations inspiring the shopper to buy other categories to complete the look, e.g. if a shopper buys a top then bottom wear, foot wear and accessories can be recommended in line with shopper preference and fashion trends


Virtual Mirror: Shoppers want to try on the clothes or make-up or the accessories before they buy. Even in the physical stores, it is not possible for the shoppers to try everything. Cosmetics giant, L’Oréal helps customers to try on lipstick, blush on, and eyelashes virtually on a magic mirror. Similarly, a few apparel retailers have started to put a virtual try-on mirror in the stores. These try-on mirrors have been made possible due to artificial intelligence


Visual Search: One of the challenges of the retailers is to help shoppers find the products that they want. If a shopper has seen an influencer picture on Instagram, how can she ask for the similar bag from the retailer? With visual search, it is now possible for the shopper to upload the picture and get the