Strategic Coupon Allocation for Increasing Providers' Sales Experiences in Two-sided Marketplaces

TL;DR

This paper tackles provider retention in two-sided marketplaces by redefining fairness as maximizing the number of providers who experience at least one sale. It introduces Sales Experience Rate (SER) as a provider-centric uplift metric and develops a coupon allocation framework that optimizes SER uplift via uplift modeling, transforming the nonlinear objective into an actionable integer linear program. Through real-world Mercari data, the SER approach with item quality assurance outperforms item-focused baselines in increasing the number of successful providers, while maintaining robustness across campaign iterations. The work advances CRM practice in marketplaces by prioritizing long-term provider diversity and sales experiences, with practical implications for sustaining network effects and provider loyalty.

Abstract

In a two-sided marketplace, network effects are crucial for competitiveness, and platforms need to retain users through advanced customer relationship management as much as possible. Maintaining numerous providers' stable and active presence on the platform is highly important to enhance the marketplace's scale and diversity. The strongest motivation for providers to continue using the platform is to realize actual profits through sales. Then, we propose a personalized promotion to increase the number of successful providers with sales experiences on the platform. The main contributions of our research are twofold. First, we introduce a new perspective in provider management with the distribution of successful sales experiences. Second, we propose a personalized promotion optimization method to maximize the number of providers' sales experiences. By utilizing this approach, we ensure equal opportunities for providers to experience sales without being monopolized by a few providers. Through experiments using actual data on coupon distribution, we confirm that our method enables the implementation of coupon allocation strategies that significantly increase the total number of providers having sales experiences.

Figures (6)

• Figure 1: Provider Sales Experience Rate (SER): SER is defined using the probability of no items being sold and its complementary event probability
• Figure 2: No items from Provider A can be sold without coupons, but one item from Provider B can be sold without coupons, so we do not need to attach coupons to make Provider B a successful provider. Coupons have resulted in an increase of successful providers through Provider A but not through Provider B.
• Figure 3: Uplift in the number of items sold: For different campaigns, the item unit greedy model (I-Greedy) achieves the best lift in the number of items sold while other models seek the lift in the number of successful providers, not in the number of items sold.
• Figure 4: Uplift in the number of successful providers: SER models with item quality assurance outperform other models focusing solely on item sale rates. Without quality assurance, the SER model's performance is close to other methods, underscoring the critical importance of this assurance preprocessing in practical applications.
• Figure 5: The number of treated providers: Provider unit Greedy (P-Greedy) model focuses on treating more providers with the same number of coupons. On the other hand, the SER model treats fewer providers and boldly allocates coupons to the same providers to maximize their conversion rates.