
Sign up to save your podcasts
Or
This academic paper explores the design of fair and effective points-based rewards programs, which incentivize customer loyalty. The authors investigate two key challenges: individual fairness due to customer heterogeneity, as personalizing redemption thresholds can be perceived as unfair, and temporal fairness, as changes in thresholds, especially increases, can lead to customer dissatisfaction. They introduce a novel learning algorithm, Fair-Greedy, which minimizes changes in redemption thresholds and only decreases them, demonstrating that these fairness constraints do not significantly hinder revenue generation. The research employs a "Buy N, Get One Free" (BNGO) model and utilizes extensive numerical experiments to support its theoretical findings, providing insights into the practical implementation of such programs.
This academic paper explores the design of fair and effective points-based rewards programs, which incentivize customer loyalty. The authors investigate two key challenges: individual fairness due to customer heterogeneity, as personalizing redemption thresholds can be perceived as unfair, and temporal fairness, as changes in thresholds, especially increases, can lead to customer dissatisfaction. They introduce a novel learning algorithm, Fair-Greedy, which minimizes changes in redemption thresholds and only decreases them, demonstrating that these fairness constraints do not significantly hinder revenue generation. The research employs a "Buy N, Get One Free" (BNGO) model and utilizes extensive numerical experiments to support its theoretical findings, providing insights into the practical implementation of such programs.