Authors:
Brian Taylor, Manda Miller, Ted Clark, Kristen Chung, Anvita Godavarthi, John Pistone, Scott Schneider
Abstract:
An AI-based tool is proposed to provide an eCommerce shopper with a recommendation of an item that is less likely to be returned.
Background:
Returns are a big profit killer for Retailers mostly because people buy with the intent to return, and it is difficult for shoppers to understand how items will fit and look when shopping online. Current solutions include making returns more centralized, including Happy Returns. However, returns consolidation services do not solve the problem of preventing the return before it ships to the customer. Conversely, this solution provides an alert that intervenes if an item is likely to be returned and provides a recommended item thar is less likely to be returned based off of that consumers behavior and other sales history.
Description:
Consumer online purchasing data is stored and aggregated, along with shopper preferences, reviews, and return data. Using A.I. algorithms, an item placed into a shopper's cart can be identified as "likely to return". At the point the shopper adds the item to their cart, a nudge recommendation pops up, that based on the algorithm delivers a personal recommendation for a particular size or color based on a few reasons such as "We recommend Size X or Style A based on your past purchases that you liked". In addition a promotional nudge is given, incentivizing the shopper to add the alternative less likely to return item. Also, you can recommend and promote removal of duplicate items found in the cart as well.
TGCS Reference 3235
