This study addresses the challenges of low efficiency in coupon distribution and suboptimal redemption rates within the Online to Offline (O2O) model. Leveraging multi-source data encompassing users, merchants, and coupons, we systematically construct a multi-dimensional feature engineering framework. This framework incorporates fundamental attributes, temporal features, user behavior patterns, merchant characteristics, interaction features, and combinatorial features. The research comprehensively compares the performance of various machine learning models—including Logistic Regression, Random Forest, Gradient Boosting Trees, XGBoost, Support Vector Machines, and Neural Networks—on the coupon redemption prediction task. Experimental results demonstrate that the Logistic Regression model achieves the best performance, with an AUC value of 0.9997. Feature importance analysis identifies user historical usage rate, discount depth, and user-merchant distance as the core influencing factors. Based on these findings, we propose targeted distribution strategies involving tiered user operations, differentiated coupon design, and geographical adaptation. This research provides both theoretical foundations and practical guidance for intelligent coupon marketing in O2O scenarios.

Zihao Zhou, Ruilin Liu, Zheng Yang, Kaiyao Wang, Zhaorui Hou, Xiuyi Wang. TempoFuse: Coupon Redemption Prediction in O2O via Temporal Feature Mining and Multi-Algorithm Fusion. Academic Journal of Business & Management (2025), Vol. 7, Issue 8: 137-146. https://doi.org/10.25236/AJBM.2025.070816.

[1] Liu Yanqiu, Yue Huina. Pricing Model and Solution of Coupon Promotion [J]. Journal of Shenyang University of Technology, 2014, 36(02): 194-198.

[2] Shou Zhigang, Liao Ziling, Zhang Yi. Research on the Effect of Coupon Promotion: A Comprehensive Test Based on Meta-Analysis [J]. Luojia Management Review, 2022(01): 75-96.

[3] Wang Kangcheng. Research on Prediction of Coupon Usage Based on Marketing Response Model [C]. Central South University of Economics and Law, 2025-03-16—2025-04-15.

[4] Lu Ping, Chen Xiaotian. Prediction of Online Coupon Usage Based on Gradient Boosting Tree Model [J]. Science Technology and Engineering, 2019, 19(18): 234-238.

[5] Xu Ning, La Lei. Prediction of New Retail Coupon Usage Behavior Based on XGBoost [J]. Journal of Southwest China Normal University (Natural Science Edition), 2019, 44(03): 101-105.

[6] Huang Bo, Hu Xia, Zhai Jingzhi. Research on O2O Coupon Distribution Strategy Based on Machine Learning [C]. Proceedings of the 18th (2023) China Management Conference and the 10th Anniversary Symposium of the Belt and Road Initiative, 2023-07-28.

[7] Song Xia. Research and Application of O2O Coupon Usage Prediction Based on XGBoost [D]. Southwest Jiaotong University, 2021.DOI:10.27414/d.cnki.gxnju.2021.002056.

[8] Qu Guanqiao. Research on the Impact of O2O Food Delivery Price Promotion Strategies on Purchase Intention [D]. Dongbei University of Finance and Economics, 2020. DOI: 10.27006/d.cnki. gdbcu.2020.001385.

[9] Li Zonghui. Research on the Full-Channel Synergy Strategy of Retail Enterprises Based on Coupon Distribution [D]. Nanjing University of Science and Technology, 2021. DOI: 10.27241/d.cnki.gnjgu. 2021. 000049.

[10] Ren Yue. Research on Precise Distribution of Coupons in the New Retail Model [D]. Dalian Maritime University, 2022.