A Customer Rewards Program (CRP) or a loyalty program is characterized by customers being rewarded for repeated purchases. The goal of these programs is to increase sales and profit through offering purchase incentives in the form of delayed rewards, as a result influencing customer behavior and in the end increasing company profit. It is an open debate in the literature if such reward programs are truly profitable for the company running it. This thesis therefore proposes a new approach to designing a CRP based on optimization methods with a focus on maximizing the expected profitability of the program. The key finding of the thesis is that such an approach can be applied to CRPs in order to increase the profitability of the companies running such programs. The approach offers promising results towards applicability to specific business segments and as a result towards maximizing the expected profits resulting from customer rewards programs. Since a largescale CRP may require significant developments on existing infrastructure and can pose high costs through the rewards redeemed during the program, the perspective of being able to maximize profits through fine-tuning the program parameters could be of great help. It can even be crucial to a company considering large investments on necessary developments or aiming to increase the returns on an existing CRP infrastructure. As part of the thesis we have analyzed the – at the time of writing – running CRP programs of several major companies present on the Hungarian market based on publicly available data in order to evaluate the applicability of the proposed optimization approach to real-world CRP designs. The analysis done on these real-world CRP designs indicates that the proposed optimization approach is applicable to a class of CRP designs witnessed to be popular among companies operating in various business segments. We have also investigated Hungary’s largest pharmacy-chain to gain insight into real-world CRP design considerations. Based on guided interviews conducted with the head of marketing at the aforementioned pharmacy-chain we conclude that the relevant information needed for our proposed optimization approach can be available to a company from business data (e.g. through billing and accounting information) readily available in an integrated IT infrastructure. We also conclude that the proposed theoretical results on CRP design are based on valid mathematical assumptions and correlate to accounting considerations related to the introduction of a CRP program.