In the contemporary digital economy, intensified market competition has rendered conventional marketing paradigms inadequate for addressing consumers’ evolving heterogeneous demands. This context necessitates the development of computer-assisted methodologies for precision-targeted and personalized marketing interventions. The present study proposes an algorithm-driven user profiling system, implementing personalized marketing strategies through rigorous data preprocessing workflows that address missing values, outliers, and normalization requirements. Cross-industry marketing datasets were analyzed during the research process to validate the proposed framework’s efficacy. Experimental evaluations demonstrate robust classification performance, achieving 92% and 88% accuracy in identifying strategic customers and core customers respectively, while maintaining above 85% accuracy thresholds for potential customers and general consumer segments. These findings substantiate the practical value of integrating machine learning-driven consumer analytics with adaptive marketing strategies in modern commercial ecosystems. This is mainly due to the deep learning algorithm’s effective extraction of core features such as transportation frequency and freight revenue, which accurately captures the customer’s value and behavior patterns.