A loyalty program offers a reward to bond customers to a company or its products or services. This study compared the medium effect with the consensus information effect in a loyalty program. In a typical loyalty program, consumers earn a number of points for every purchase they make and then they can exchange a specified number of points for a desired reward they want to choose. In other words, consumers usually have to exert effort in order to achieve a desired reward, however, the immediate payoff of their effort is not the reward they actually care about. Instead it is merely an instrument, a token, which has no value in itself but can be traded for the reward they care about. The instrument is merely a ‘medium’ between their effort and the desired reward as in the following: effort →medium→reward. Because the medium is inherently worthless consumers should skip it and base their decision solely on the effort→reward relationship. However, the effort→reward relationship typically not direct given, and consumers have to infer it from the relationship between effort→medium and medium→reward. In order to make a reasonable choice, consumers should pay equal attention to these two types of relationship. However they often fail to pay equal attention to these two types of relationship and maximize the not just the effort→ reward but also the effort→reward. When consumers are faced with options entailing different outcomes, the presence of a medium can alter what option they choose. That is, consumers choose the option with a bigger medium even though the economic value of the option is not changed by the medium. Hsee et al. (2003) called this phenomenon “medium maximization (or medium effect)”. Meanwhile, people are influenced by the consensus information. Consensus information is defined as information involving others’ opinions about or evaluations of an attitude object, for example, “80 percent of 300 customers in Korea have selected the ○○○”. This study applied these two effects to customer loyalty program and investigated which one has more powerful influence to customers’ choices. Our hypotheses are followed; H1. A medium would affect on consumers’ choices of the reward option, H2, An consensus information would influence consumers’ choices of the reward option, H3. The consensus information effect would be more powerful than the medium effect. To test all the hypotheses, 149 respondents were assigned to one of three groups (control group vs. the medium presented group vs. the consensus information presented group). All the respondents were asked to make choices between the two different rewards (e.g. white wine vs. red wine). For the diverse 11 products including hedonic and utilitarian goods, we measured and compared the medium effect with the consensus information effect. The comparisons showed that the consensus information effect is more influential than the medium effect on consumers' choices. In the comparisons, additionally, we observed the positive correlation between the two effects. In sum, this research implies that the medium and the consensus information can be used to maintain customers although the economic value of the options (rewards) are the same.