In Pursuit of Lost Subscribers: A Field Experiment on Effective Resubscription Strategies
39 Pages Posted: 27 Jun 2024
Date Written: June 22, 2024
Abstract
Many subscription-based platforms struggle with significant user churn rates and seek effective strategies to re-engage lost subscribers. While discounts have shown effectiveness in various contexts, they may not suffice for re-engaging lost subscribers in subscription-based platforms. This study, conducted with a leading online imageediting platform, reassesses this approach within the context of a subscription-based business model and introduces an innovative method to fulfill users' psychological needs: providing a feedback channel. Through a field experiment, we found that financial incentives are not effective in winning back customers, but providing a feedback channel can positively increase the re-subscription rate by a substantial 40.9% relative to the mean of the control group. To gain more insight, we conducted a vignette-based lab experiment. The results indicate that while a coupon is beneficial in providing financial incentives, it results in strong perceived unfairness among users, leading to a lower re-subscription rate. Conversely, the feedback treatment can increase users' satisfaction, likely because the feedback channel grants users a way to voice their concerns and suggestions. Furthermore, our heterogeneous treatment effects analyses indicate that providing a coupon to lost subscribers with a longer prior subscription experience is more detrimental to re-subscription. In contrast, the impact of providing a feedback channel is more effective for lost subscribers with a longer prior subscription experience or a shorter unsubscription time gap. Our findings challenge conventional marketing methods and underline the importance of adopting more appropriate strategies to address user churn problems on subscription-based platforms.
Keywords: User Churn Management, Subscription-based Business Model, User Re-subscription, Coupons, Feedback Channel
Suggested Citation: Suggested Citation