The emergence of social media influencers in the past decade has attracted considerable interest from scholars and marketing professionals. Scholars aim to elucidate the influence of these figures on consumer behavior, while marketers employ influencers as a crucial element of their strategies to enhance brand engagement. Electronic Word of Mouth (eWOM) has become an indispensable tool in digital marketing, playing a vital role in fostering consumer engagement. Sustainable fashion is increasingly recognized as a pivotal area of focus due to the fashion industry’s significant contribution to environmental concerns, including waste production, excessive water usage, and carbon emissions. Despite extensive theoretical research, empirical studies on influencer marketing in sustainable fashion remain limited. This study addresses this gap by analyzing Instagram© posts and engagement metrics and developing a data-driven framework for influencer impact assessment. The proposed model integrates machine learning and Principal Component Analysis (PCA) to compute and predict influencer scores, providing a structured method for influence quantification. Moreover, descriptive and correlation analyses offer deeper insights into the relationship between influencer activity and engagement patterns. Findings indicate that hashtag strategy plays a crucial role in engagement, with #sustainablefashion and #slowfashion generating the highest interaction, while #zerowastefashion and #upcycledfashion see lower engagement levels. The study also reveals that follower count alone is an unreliable measure of influence, as engagement is driven by content quality, language, and strategic hashtag use. The proposed model effectively quantifies influencer impact, achieving high predictive accuracy (R \(^2\) = 0.95, MSE = 0.0028). These insights provide a scalable, data-driven approach for influencer selection and campaign optimization, offering practical applications across various digital marketing domains.