Synergy between Customer Segmentation and Personalization

Abstract

Personalized pricing is on the rise in the retail industry, and it’s been used in many service platforms. Personalized pricing assumes that the platform has an idea of who the customer is, and allows the platform to incentivize the customers based on their characteristics and actions. In this work, we consider a personalized pricing problem with revisiting customers. We assume a customer’s utility depends on some feature information and some unobserved personal shock. For the case where customer utilities are unknown, we propose a heuristic that combines the idea of binary search and feature-based pricing, and show that it outperforms common benchmarks. We also demonstrate that if we have some information about either the customer utility or the dependence on the feature information, then we can learn the other much faster, and this is what we mean by the synergy between customer segmentation and personalization. This observation can help managers make better personalized decisions for customers that repeatedly interact with the platform.

Keywords

Pricing / algorithms / heuristics / personalization

References

[1] Shipra, Goyal Navin A. International Conference on Machine Learning, 2013

[2] S Nageeb, Lewis Greg, Vasserman Shoshanatitle A. Proceedings of the 21st ACM Conference on Economics and Computation, 2020

[3] Araman V F, Caldentey R. Dynamic pricing for nonperishable products with demand learning. Operations Research, 2009, 57(5): 1169-1188.

[4] Auer P, Cesa-Bianchi N, Freund Y, Schapire RE. The nonstochastic multiarmed bandit problem. SIAM Journal on Computing, 2002, 32(1): 48-77.

[5] Ashwinkumar, Langford John, Slivkins Aleksandrs B. Conference on Learning Theory, 2014

[6] Gah-Yi, Keskin N Bora B. Available at SSRN 2972985, 2019

[7] Hamsa, Harsha Pavithra, Perakis Georgia, Singhvi Divya B. Available at SSRN 3240970, 2018

[8] Besbes O, Zeevi A. Dynamic pricing without knowing the demand function: Risk bounds and near-optimal algorithms. Operations Research, 2009, 57(6): 1407-1420.

[9] Besbes O, Zeevi A. On the surprising suciency of linear models for dynamic pricing with demand learning. Management Science, 2015, 61(4): 723-739.

[10] den Boer A V, Zwart B. Simultaneously learning and optimizing using controlled variance pricing. Management Science, 2013, 60(3): 770-783.

[11] Broder J, Rusmevichientong P. Dynamic pricing under a general parametric choice model. Operations Research, 2012, 60(4): 965-980.

[12] Henry C C C, Bin C C, Wen C H. Pricing and promotion strategies of an online shop based on customer segmentation and multiple objective decision making. Expert Systems with Applications, 2011, 38(12): 14585-14591.

[13] Ningyuan, Elmachtoub Adam N, Hamilton Michael, Lei Xiao C. Available at SSRN 3430125, 2019

[14] Chen X, Wang Y, Zhou Y. Dynamic assortment optimization with changing contextual information. Journal of Machine Learning Research, 2020, 21(216): 1-44.

[15] Chen Z, Choe C, Matsushima N. Competitive personalized pricing. Management Science, 2020, 66(9): 4003-4023.

[16] Cheung W C, Simchi-Levi D, Wang H. Dynamic pricing and demand learning with limited price experimentation. Operations Research, 2017, 65(6): 1722-1731.

[17] M, Lobel I, Paes Leme R C. Available at SSRN 2737045, 2016

[18] Farias V F, Van Roy B. Dynamic pricing with a prior on market response. Operations Research, 2010, 58(1): 16-29.

[19] Elaheh, Ma Will, Simchi-Levi Davidtitle F. Available at SSRN 3443109, 2019

[20] Kris, Goh Joel F. Management Science, 2018

[21] Ferreira K, Simchi-Levi D, Wang H. Online network revenue management using thompson sampling. Operations Research, 2018, 66(6): 1586-1602.

[22] Joshua A, Liozu Stephan Mtitle G. Journal of Revenue and Pricing Management, 2020

[23] Michael H J, Bora K N, Assaf Z. Bayesian dynamic pricing policies: Learning and earning under a binary prior distribution. Management Science, 2012, 58(3): 570-586.

[24] Adel J, Hamid N. Dynamic pricing in high-dimensions (2019). The Journal of Machine Learning Research, 2019, 20(1): 315-363.

[25] Adel J, Hamid N, Simengtitle S. 2020 IEEE International Symposium on Information Theory (ISIT), 2020

[26] Bora K N, Assaf Z. Dynamic pricing with an unknown demand model: Asymptotically optimal semi-myopic policies. Operations Research, 2014, 62(5): 1142-1167.

[27] Andreas K, Mark F, Tom S. Are airline passengers ready for personalized dynamic pricing? A study of German consumers. Journal of Revenue and Pricing Management, 2018, 17(2): 115-120.

[28] Lihong, Chu Wei, Langford John, Schapire Robert E L. Proceedings of the 19th International Conference on World Wide Web, 2010

[29] Fanlin, Kazemtabrizi Behzad, Zeng Xiaojun, Dent Christitle M. IEEE Pes Innovative Smart Grid Technologies Conference Europe, 2018

[30] Mueller J M, Pommeranz B, Weisser J, Voigt K I. Digital social media and mobile marketing in industrial buying: Still in need of customer segmentation? Empirical evidence from Poland and Germany. Industrial Marketing Management, 2018, 73: 70-83. AUG

[31] Sheng, Bayati Mohsen Q. Available at SSRN 2765257, 2016