PERSONALIZATION AND FORECASTING BASED ON BIG DATA IN E-COMMERCE

Abstract

The term “ecommerce personalization” relates to the set of practices in which an online store displays dynamic content based on customer data, such as demographics, intent, preferences, browsing history, previous purchases, and device usage - for instance, whether the customer is shopping on a mobile, tablet, or even a smartwatch. Catering to a customer’s needs is not just a present-state issue. E-commerce depends on stocking the correct inventory for the future. Big data can help companies prepare for emerging trends, slow or potentially booming parts of the year, or plan marketing campaigns around large events. E-commerce companies compile huge datasets. By evaluating data from previous years, e-retailers can plan inventory accordingly, stock up to anticipate peak periods, streamline overall business operations, and forecast demand. For instance, e-commerce sites can advertise large markdowns on social media during peak shopping times to get rid of excess product. To optimize pricing decisions, e-commerce sites can also give extremely limited-time discounts. Understanding when to offer discounts, how long discounts should last, and what discounted price to offer is much more accurate and precise with big data anlytics and machine learning.