Social referral programs, in which individuals recommend products or services within their networks in return for rewards, have been widely adopted across digital platforms. This study explores the impact of incorporating uncertainty into the rewards of such programs, focusing on how senders and recipients perceive and react to uncertain rewards. We run a randomized experiment involving more than 160,000 users of a telecommunications operator in China and examine the effectiveness of different referral reward schemes. We find that referral programs are most successful when senders are incentivized with uncertain rewards and recipients are guaranteed certain rewards. Specifically, introducing uncertainty in the sender’s reward leads to a 20.9% increase in total referrals with recipients of these invitations more likely to engage in subsequent referrals and profitable in-app activities. In contrast, uncertainty in the recipient’s reward results in a 12.3% decrease in total referrals with invited recipients showing a lower propensity to make further referrals and reduced postreferral engagement. Additional online experiments identify distinct mechanisms driving these asymmetric effects: For senders, the uncertainty alleviates feelings of guilt, enhancing referral sharing and, thus, increasing the total number of referrals. For recipients, the adverse effects of uncertainty stem primarily from diminished perceptions of fairness and social pressure, and these deter engagement in the referral process. Our study sheds light on the complex dynamics of reward uncertainty in referral programs, offering novel insights into how it can be optimized to foster more engaged referral networks. This paper was accepted by Anindya Ghose, information systems. Funding: A.T. Li acknowledges support from the National Natural Science Foundation of China Grants 72301269 and 72332007 and the Fundamental Research Funds for the Central Universities Grant WK2040000085. R. Belo acknowledges funding by Fundação para a Ciência e a Tecnologia (UID/00124/2025, UID/PRR/124/2025, Nova School of Business and Economics, DOI: https://doi.org/10.54499/UID/00124/2025 ) and LISBOA2030 (DataLab2030 - LISBOA2030-FEDER-01314200). T. Li acknowledges support from the Erasmus Research Institute of Management (ERIM) Convergence AI, Data & Digitalisation (Convergence AI Lab – Immersive Tech) Gravitation Program Public Values in the Algorithmic Society (AlgoSoc). Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2024.05685 .
Li et al. (Thu,) studied this question.