肖俊极.docx

1、肖俊极Are points like money? An empirical investigation of reward promotion effectiveness for multi-category retailersABSTRACTPoint-based frequency reward programs are widely used by retailers as a sales promotion strategy. To promote a specific product category, retailers offer more favorable reward r

2、atios so that members can earn extra points. This paper examines the impact of reward ratio variations on sales in a multi-category setting, and compares the effectiveness of the reward and price promotion strategies. The authors estimate a market basket model using scanner data of member purchases

3、from a large retailer in four categories, grouped into two pairs. They find that increasing the reward ratio in a category positively affected its choice probability for three of four categories, but that the elasticity is smaller than price elasticity. Rewards promotions also impact the choice prob

4、ability of non-promoted but closely related categories, and even of categories that are only loosely related. As forms of sales promotion, price discounts and reward promotions can substitute for each other. The authors construct and compute two measures, the rate of substitution and the retailer va

5、lue of promotion points, to quantify the effects of substitution. Keywords: Sales promotion, frequency reward programs, shopping behavior, elasticity, rate of substitution.“But what is a mile worth? Airlines sell them to credit-card firms at an average of just under 2 cents a mile; their value when

6、used to buy a ticket or to upgrade to business class can be anywhere between 1 cent and over 10 cents per mile.”- The Economist, 2005a“Yet the dollar has been superseded not by the euro, nor by the yen or yuan, but by another increasingly popular global currency: frequent-flyer miles.”- The Economis

7、t, 2005bFrequency reward programs have become commonplace among firms in many industries, leading to a proliferation of reward pointsthe most notorious being frequent-flyer miles. In a 2005 article, the Economist boldly claimed that “the dollar had already been toppled as the worlds leading currency

8、”, in favor of frequent-flyer miles. In recent years, however, the retail sector (including prominent businesses such as Costco and Macys) has surpassed the travel and hospitality sectors in the prevalence of frequency reward programs. The 2009 U.S. Loyalty Marketing Census, conducted by the researc

9、h agency Colloquy, reports that as of 2008 the retail sector accounted for the largest fraction of memberships at 39%, with 701 million individuals participating in loyalty programs (Ferguson and Hlavinka 2009). Retailer rewards have become far more sophisticated than the traditional offer of “buy o

10、ne get one free”. In general, the reward programs adopted by airlines, credit card companies, and retailers are now very similar in that members receive and accumulate points based on their purchases, and may exchange them at any time for benefits largely proportional to the number of points redeeme

11、d (Berman 2006). As with airline miles, retailer reward points can be treated as an alternative currencyat least within the retailers location. But what is a point worth?There are several perspectives on evaluating reward points. First, as illustrated by the above quotes from the Economist, retailer

12、s usually impose rules on the “exchange rate” between monetary payments and reward points in both earning and redemption. Second, the consumers subjective perceptions of reward points may not be consistent with their actual monetary value (Drze and Nunes 2004). We propose a third approach to underst

13、anding the relationship between monetary prices and reward points. Rather than focusing on the retailer or consumer, this research studies the effectiveness of rewards points in generating short-term sales. In other words, we look at reward promotions as a form of sales promotion in comparison. The

14、marketers operating frequency reward programs have come to realize that the acts of awarding and redeeming points also represent opportunities to implement new marketing strategies, especially ways to enhance the perceived value of the program. One widely adopted strategy is to provide more generous

15、 terms for earning points over a limited period of time and/or for a limited category of purchases. In the following discussion, we use the term “reward ratio” to refer to the number of reward points a consumer earns for every unit of money spent on particular purchase. For example, the Canadian dru

16、gstore chain Shoppers Drug Mart offers consumers a reward ratio 20 times of its usual ratio over certain long weekends. Such promotions speed up the accumulation of points for consumers, providing more savings on future purchases. Thus, reward promotions are similar to price reductions in that they

17、stimulate sales during specified promotional periods due to the points pressure effect where “customers increase their purchase rate in an effort to earn a reward” (Taylor and Neslin 2005). This research aims to answer the following question: As a sales promotion strategy, how effective are reward p

18、romotions compared to price discounts? Most existing research on reward programs focuses on behavioral questions, in particular whether and how they affect customer loyalty and retention (Liu 2007; Bolton, Kannan, and Bramlett 2000; Keh and Lee 2006; Kivetz, Urminsky, and Zheng 2006; Lewis 2004). Th

19、e literature on marketing and customer relationship management (CRM) has dealt exhaustively with the use of reward programs in purchase tracking (Mauri 2003). Other authors have investigated their role in price discrimination (Hartmann and Viard 2008). But none of these papers has explained why reta

20、ilers would offer extra reward points at certain times. If the goal were simply to lock in members, there would be no need for temporary promotions. Researchers have also noticed an intriguing lack of empirical studies concerning the immediate effect of reward promotions on short-term sales (Bolton,

21、 Kannan, and Bramlett 2000; Drze and Hoch 1998; Lal and Bell 2003; Lewis 2004). We emphasize that the focus of this study is not reward programs per se, but the sales impact of reward promotions. Though long-term loyalty and the ability to track shopper information are certainly desirable benefits,

22、it is equally important for academics and marketers to understand and explore the other creative uses to which reward programs have been put. Here we consider their ability to boost short-term performance. This research studies the effectiveness of reward promotions in the context of multi-category

23、retailing. In a shopping trip where consumers are faced with multi-category purchase decisions, the outcome is usually a “basket” of categories rather than a single category (Manchanda, Ansari, and Gupta 1999; Russell et al. 1999; Seetharaman et al. 2005). A typical retailer carries hundreds of cate

24、gories at a single location, and the consumer response to price changes and promotional activities varies significantly from category to category (Narasimhan, Neslin, and Sen 1996). Moreover, a promotion can affect sales in nonpromoted categories. In order to design optimal pricing and promotion str

25、ategies, retailers must understand these effects and coordinate their marketing activities across multiple product categories. As such, the primary purpose of this study is to analyze the effects of price reductions and increased reward ratios both within and across categories. Indeed, this paper is

26、 the first empirical study of the effect of reward promotions on cross-category purchase behavior.While they share certain similarities, price and reward promotions work rather differently. Discounts offered in the form of reward points can be realized only in future repeat purchases. Thus, reward p

27、oints help lock consumers in for at least one more visit, a benefit that price cuts cannot achieve. If consumers do not redeem the points, which happens quite often, the outstanding balance will never become a liability to the retailer. On the other hand, reward points are a pseudo-currency that can

28、 be converted into money to pay for later purchases. Thus, it is no surprise that consumers tend to regard them as a proxy for price cuts. A sales promotion can take the form of either direct price cuts or higher reward ratios. As a point of comparison, it is interesting to measure the effectiveness

29、 of these two strategies at increasing sales. To quantify the relationship between the strategies, we have constructed a novel measure: the “rate of substitution” (RS) between price and reward ratio promotions. RS is defined as the ratio between a percentage change in prices and a percentage change

30、in reward ratios that applied together cause no change in the sales of a particular category.A retailers product categories usually belong to different price tiers (e.g., facial care products tend to be more expensive than personal hygiene products). Retailers can promote specific categories by offe

31、ring extra reward points for the same amount spent. But how does a reward point promotion in one category impact sales in other categories? To answer this question, we apply a parsimonious shopping basket model to scanner data from a large retailer. Such models allow one to analyze the interdependen

32、ce of demand across item categories in the final baskets. We can identify not just the direct effect of a reward promotion on sales of the promoted category, but its externality on categories that have similar functionality or nearby shelf locations. Our empirical results show that reward point prom

33、otions effectively increase sales of the promoted category, and also have a positive effect on sales in other categories. We consider four categories grouped into two pairs: facial care and personal hygiene. Our results indicate that complementarity is found mostly within each pair. Furthermore, we find that reward point promotions can serve as