When to Target Customers? Retention Management using Dynamic Off-Policy Policy Learning

50 Pages Posted: 11 Dec 2022 Last revised: 2 May 2024

See all articles by Ryuya Ko

Ryuya Ko

University of Tokyo

Kosuke Uetake

Yale School of Management

Kohei Yata

University of Wisconsin - Madison - Department of Economics

Ryosuke Okada

ZOZO Research

Date Written: May 2, 2024

Abstract

We propose a method to learn personalized customer retention management strategies when
customers’ intentions to purchase evolve over time. Working with a Japanese online platform, we
first implement a large-scale randomized experiment, in which coupons are randomly sent to first-time buyers at different times. The experimental data allow us to estimate personalized dynamic
retention policies using off-policy policy learning methods. We extend the existing methods by allowing inter-temporal budget constraints and feasibility constraints. Our offline evaluation results
show that the optimal dynamic policy is more cost-effective than baseline policies. Finally, we test
the optimal policy online to confirm its performance.

Keywords: Retention management, Off Policy Learning

Suggested Citation

Ko, Ryuya and Uetake, Kosuke and Yata, Kohei and Okada, Ryosuke, When to Target Customers? Retention Management using Dynamic Off-Policy Policy Learning (May 2, 2024). Available at SSRN: https://ssrn.com/abstract=4293532 or http://dx.doi.org/10.2139/ssrn.4293532

Ryuya Ko

University of Tokyo

Kosuke Uetake (Contact Author)

Yale School of Management ( email )

135 Prospect Street
P.O. Box 208200
New Haven, CT 06520-8200
United States

Kohei Yata

University of Wisconsin - Madison - Department of Economics ( email )

1180 Observatory Drive
Madison, WI 53706
United States

HOME PAGE: http://www.koheiyata.com

Ryosuke Okada

ZOZO Research

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
390
Abstract Views
1,160
Rank
152,386
PlumX Metrics