Unstructured Data Analytics to Improve Digital Eligibility of E-Commerce Listings

With the businesses currently scaling up at a rapid rate, automation is a key component for sustainability. It ensures a quick turnaround time and fewer errors caused due to human interaction. This leads to an improved customer satisfaction. According to a retailer giant, only 33% of their active products are eligible to be purchased on their website. A lack of digital eligibility restricts them to sell their products only in their store. The generation of product description for the website is currently handled by vendors, and multiple products have descriptions which are either missing or not a good fit. This lowers the digital eligibility and in turn the number of products that could be listed on the website. We work on two different objectives using unstructured data, which help improve the digital eligibility of the products. One, we score the existing product descriptions using an algorithm which we developed. This informs us about the product descriptions which need to be updated. Two, we create a model to generate product description based on the product images. This automation will reduce the effort invested by vendors who manually write the product descriptions.