Data-Driven Fashion Trend Identification: Utilizing Text Mining for New Product Development Process

Historically, the fashion trend forecasting process has been deeply rooted in the expert judgment of industry professionals, who offer insights based on their profound understanding of cultural and societal dynamics. On the other hand, the rapid integration and incorporation of emerging trends into the New Product Development (NPD) process has been highlighted as critical for sustaining competitiveness in the evolving landscape of the fashion industry. Hence, this study evaluates the effectiveness of text mining in identifying fashion trends through online consumer discussions. Data from ‘The Fashion Spot’ forum were analyzed using text mining techniques to track consumer preferences for fashion attributes such as color, style, and material between 2019 and 2023. SHEIN was used as a comparison to understand how well consumer-driven discussions align with market offerings. Our findings reveal consistent preferences for classic colors and highlight the role of societal events in shaping trend shifts. The study contributes to the literature by demonstrating how data-driven approaches can complement traditional trend forecasting in NPD. Additionally, it provides practical recommendations for brands to integrate real-time data into product strategies, enhancing responsiveness to market dynamics.