Purchasing Pattern Exploration in Fashion Retail using Transaction Data

Abstract

Large volumes of fashion retail transaction data offer valuable insights into consumer purchasing behavior, yet extracting actionable knowledge remains challenging in a highly competitive market. This study analyzes an open-access fashion retail dataset containing product details, transaction values, review ratings, payment methods, and purchase dates to support management and marketing decisions. Consumer purchasing behavior is examined using predictive analytics and clustering techniques. The results show a high level of predictability in transactions, indicating that consumer purchases follow consistent patterns rather than random behavior. Clustering analysis further reveals the presence of distinct consumer segments, although the separation between these groups is relatively weak. Despite this overlap, the identified patterns provide meaningful insights into customer behavior. Overall, the findings demonstrate that transaction-level data can be effectively leveraged to support inventory planning, product assortment decisions, and segmentation-based marketing strategies. This research highlights the strategic value of fashion retail transaction data in enhancing data-driven decision-making and improving managerial effectiveness in the fashion retail sector.