Abstract
With the rapid expansion of the global blind box market and its growing share in the online cultural and creative consumption sector, understanding the factors that shape user experience in blind box online purchases has become increasingly important. This study adopts a mixed-methods approach that integrates large-scale text mining with the Kano model and the entropy weight method to systematically identify and prioritize key experience attributes. Based on 18,981 valid consumer reviews collected from major Chinese e-commerce platforms (Tao Bao, Jing Dong, and Su Ning), four core dimensions—product quality, shopping service, price value, and emotional/social interaction—along with 19 specific factors were extracted. The Kano classification results indicate that image–product consistency, logistics speed, and gift or collection-related elements function as must-be attributes; cost-effectiveness, quality consistency, workmanship, authenticity, and unboxing expectations are identified as performance attributes; design safety, packaging robustness, and transportation security are categorized as attractive attributes; while brand labeling, after-sales service, and social sharing are classified as indifferent attributes. Further entropy weight analysis reveals substantial differences in the relative importance and evaluation dispersion of attributes within and across Kano categories. By integrating asymmetric satisfaction effects with information-based importance weighting, the proposed Kano–entropy framework advances a refined prioritization perspective for blind box experience attributes, with practical relevance for decision-making in product, logistics, and service design.