A Comparison of Probabilistic Prize Promotion Schemes

Article excerpt

Based on the rank-dependent expected utility model (Quiggin, 1991), hypotheses are formed in this study regarding optimal prize promotion structure with reference to associated probabilistic aspects. Referencing both modeling works and related behavioral theories, we compared several design schemes. Using purchase scenarios of a low-value product (bread) and a high-value product (cell phone) we determined the optimal design among promotion schemes that differ by winning probability, prize amount, and number of prize value levels. We found that a promotion offering a combination of high-value prizes plus some low-value prizes was invariably preferred over a promotion offering only high- or low-value prizes. We also explored whether these high- or low-value prizes should be in a series of ascending value or prizes at each value level should be of the same value and whether or not some moderately valuable prizes should also be included.

Keywords: probabilistic prize promotion, rank-dependent expected utility, promotion structure.

(ProQuest: ... denotes formulae omitted.)

The term probabilistic prize promotion refers to the activity of providing chances for consumers to win various kinds of prizes such as cash, gifis, opportunities to travel, and so on, through activities such as competition, prize drawing, or games (Kotier & Keller, 2006). Compared with other forms of promotion, probabilistic prize schemes are more flexible because they can fit into different structures with different content. However, many factors are likely to affect their effectiveness, such as prize structure, the value of promoted products, monetary value of the prizes, and the probability of winning a prize (Erev & Haruvy, 2010; Howard & Barry, 1990). When choosing probabilistic prize promotion as a marketing tool, promotion organizers tend to be subjective and blinkered (Erev & Haruvy, 2010). Therefore, a scientific, systematic, theoretical, and empirical study of the factors involved in the process of probabilistic prize promotion is likely to be of great practical value to those responsible for designing marketing strategies for enterprises.

Design is an important aspect in probabilistic prize promotion. However, in previous studies only probability levels under extreme conditions, that is, high or low probability, have been examined (e.g., Chen & Jia, 2005). Comparisons among various kinds of probabilities, including moderate levels and the intricate structure of incentive/probability design, have seldom been addressed in a real-life setting. In this study our aim was to determine the optimal design of incentives, first through a process of modeling, verification, and development, and then by experimenting with the propositions we had developed in the modeling stage and that were also based on relevant theories of behavioral economics.

Literature Review

Probability /prize design is the core of the study of probabilistic prize promotion schemes. Chen (2007) conducted research on the effectiveness of two different kinds of promotions, namely a low probability of winning a prize worth a large amount of money and a certainty of winning a gift of small monetary value. She found that when the promotion involves high-priced products, the former prize option proves more effective, whereas there is no significant difference between the two options when selling low-priced products.

Lin, Ke, and Tao (2008) found that, for high-priced products, as the prize value increases, the effect of the promotion takes the shape of a reversed U curve, and as the value declines, the effect takes the shape of a U curve. For low-priced products, the effect of the promotion was not found to be significantly different whether the prize was of a high or low value.

Some other scholars consider a probabilistic prize as an imprecise discount and have compared this with traditional sales promotions in which a discount is a fixed amount. …

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.