words, the seller can use a Group-Buying discount to âhireâ informed customers to work as âsales agentsâ to encourage novice customers to join the group.
Group-Buying: A New Mechanism for Selling through Social Interactions
Xiaoqing Jing Georgia Institute of Technology Jinhong Xie University of Florida
This Version: 01/09/2011
Electronic copy available at: http://ssrn.com/abstract=1030824
Group-Buying: A New Mechanism for Selling through Social Interactions
Abstract This paper examines a unique selling strategy, Group Buying, under which consumers enjoy a discounted group price if they are willing and able to achieve a required group size and coordinate their transaction time. We argue that Group Buying allows a seller to gain from facilitating consumer social interaction, i.e., using a group discount to motivate informed customers to work as “sales agents” to acquire lessinformed customers through interpersonal information/knowledge sharing. We formally model such an information-sharing effect and examine if and when Group Buying is more profitable than (1) traditional individual selling strategies, and (2) another popular social interaction scheme, Referral Rewards programs. We show that Group Buying dominates traditional individual selling strategies when the information/knowledge gap between expert and novice consumers is neither too high nor too low (e.g., for products in the mid-stage of their life cycle) and when inter-personal information sharing is very efficient (e.g., in cultures that emphasize trust and group conformity, or when implemented through existing online social networks). We also show that, unlike Referral Rewards programs, Group Buying requires information sharing before any transaction takes place, thereby increasing the scale of social interaction but also incurring a higher cost. As a result, Group Buying is optimal when interpersonal communication is very efficient, or when the product valuation of the less-informed consumer segment is high.
1
Electronic copy available at: http://ssrn.com/abstract=1030824
1. Introduction Recent advances in technology have created many new opportunities for marketers to develop and implement innovative selling strategies (Shugan 2004), such as Referral Rewards programs (Biyalogorsky et al. 2001), advance selling (Xie and Shugan 2001), customized product and service design (Syam et al. 2005, Valenzuela et al. 2008), one-to-one promotion (Shaffer and Zhang, 2002), and probabilistic selling (Fay and Xie 2008, 2011). In this paper, we explore another interesting selling strategy, Group Buying, which has generated increasing attention in recent years because of the facilitation of information and communication technologies. While many different forms of Group Buying exist, a common characteristic is that the seller offers discounted group rates to encourage individual customers to purchase through buying groups. As an example, consider a specific Group-Buying offer posted recently by an online service provider, BrowserCam, on Fundable.com, a social network website that helps people to organize different group activities.1 This Group-Buying activity started on 07/13/2007 at 02:08 AM and ended on 07/30/2007 at 10:36 PM (with a deadline of 08/07/2007 at 11:59 PM). Here is the offer posted by the seller:
What is BrowserCam? BrowserCam is an online service that offers cross-browser website testing through screen captures and remote access via VNC...The normal annual price for 'BrowserCam Complete' is $499.95 USD which includes unlimited access to the Capture service and unlimited access to 60-minute Remote Access sessions on Windows and Linux machines and 30-minute sessions on Mac machines. Through this website (Fundable), BrowserCam is offering an AMAZING deal. For the cost of only one day of BrowserCam's service, you get access for an entire year. Yes, exactly the same service as you would normally pay $499.95 USD, for only $25 when 20 others register as a group…Be quick because these group actions usually complete within only a few days. How it works You simply pledge $25 via PayPal or Credit Card (but no payment is made at this stage), then, when the total of 20 people have made a pledge, Fundable will process your payment. BrowserCam is then contacted and they activate each individual’s account.
BrowserCam has initiated many similar Group-Buying activities on social network websites since 2005. From communications recorded online, one can observe some interesting interactions between the members who have signed up and potential members. The following are some examples:
1
https://www.fundable.com/groupactions/groupaction.2007-07-13.7251291800
1
“We’re getting closer, but we need about 17 more people to close this up and get a great deal on the best package for the Browsercam premium annual subscription…Hey look. What do you have to lose?” “I’ve just finished using the 24-hour free trial and I have to say that this is an amazing service for web developers. My CCS was malfunctioning in quite a few browsers and I would have never known. Sign up, this is an excellent service and you won’t regret it!” “Hi All, Just thought I would share something that may really help a lot of people when designing their Mambo Templates. BrowserCam (http://www.browsercam.com) lets you preview your website in a number of different browsers and operating systems in one click…Any questions? Feel free to ask; I’ll answer as best as possible…” “…. It basically does screen captures of web pages using many different browsers so you can quickly see the results of CSS and web pages…Fundable is designed to provide a safe way to allow groups of people to pool money to purchase things…nobody is actually charged until the purchase is fully funded…For those who do lots of website or skin design, I've found BrowserCam to be a really valuable tool, and this is a relatively inexpensive way to get an account, so I invite you to join in!” “Only 4 spots left... call or email everyone you know today. I think we can wrap this thing up if we just get 4 more people in.”
The above messages seem to suggest that, in order to successfully reach the group size requirement before the deadline (e.g., 20 people within 25 days in most BrowserCam Group-Buying offers), the signed-up group members play an active role in disseminating product information and persuading others into joining the buying group. Similar Group Buying and associated information-sharing activities can be commonly observed in many on- and offline communities, including J-body Forum, Ls1Tech, and ThirdGen. Early third-party Group-Buying service providers (e.g., Mercata and MobShop) generated a lot of attention but failed to survive the dotcom crash. Recently, with the rise of online social media and social networks, numerous Group-Buying websites (e.g., Groupon.com, LivingSocial.com) have appeared and have created a new “buzz” in the business world (The Wall Street Journal, March 23, 2010, New York Times, November 27, 2010). It is also interesting to note in countries with a collectivistic culture where high value is placed on social interactions, such as China, the scale and scope of Group Buying is already quite broad (The Wall Street Journal, Feb. 28, 2006, The Wall Street Journal, May 12, 2008). While we have observed considerable variation in the scale, scope, and even success of groupbuying activities over time, across different countries, and across different product/service categories in the market, how the seller benefits from this non-traditional buying and selling format is for the most part still an open question (iMediaConnection, June 01, 2010). This paper explores the potential profit advantage of Group Buying as a marketing tool to exploit social interaction. Consumers often differ in their levels of product information/knowledge (e.g., product awareness, knowledge about product technique features, functions, installation, usage, and services). Such an information gap may result in
2
differences in their product valuations. We argue that Group Buying can be used to motivate informed customers to market products to less-informed customers via knowledge/ information sharing. In other words, the seller can use a Group-Buying discount to “hire” informed customers to work as “sales agents” to encourage novice customers to join the group. To advance our understanding of this unique selling format, we formally model Group Buying as a mechanism to motivate interpersonal information- and knowledge-sharing, and explore when and how a seller can gain from Group Buying compared with traditional individual selling strategies. We then investigate the relative advantages/disadvantages of Group Buying compared with another form of social interaction that is often observed in practice and studied in the literature: Referral Rewards programs, under which the seller uses monetary rewards to motivate existing buyers to spread product information and thereby increase sales. Our results reveal that, when there is a moderate level of consumer information heterogeneity in the market, Group Buying is more likely to dominate a traditional Margin strategy that seeks a high profit margin by selling to high-valuation consumers only, and a Volume strategy that seeks a high sales volume by charging a low price. This occurs because a very wide information gap between expert and novice customers implies a high cost for the seller to expand its market though customer sales agents, and a very narrow information gap implies a limited potential for profit margin improvement through social influences. This finding suggests that Group Buying may not be an effective selling format for new and/or technology-intense products, for which a significant information gap may exist between expert and novice consumer segments. In the later stages of the product life cycle (e.g., for well-known products, wellestablished brands, or for low-tech products), Group-Buying may not be an attractive option either as the information heterogeneity may be too small for the seller to benefit from this type of social interactions. We also show that Group Buying differs profoundly in several important aspects from another commonly observed social interaction scheme – Referral Rewards Programs. For example, Group Buying has the advantage of generating a larger scale of social interaction. This is because, different from Referral Rewards programs which allow the customer to decide whether to make a referral after she has purchased and consumed the product, Group Buying requires the customer to make referrals before any transaction is completed and the uncertainty about the consumption experience is resolved. Such “prepurchase referral” via Group Buying, however, also creates higher cost for the seller to motivate information- and knowledge-sharing. For example, when referral takes places in the post-purchase period as under Referral Rewards programs, the seller may benefit from customer satisfaction and reduce compensation costs, since satisfied customers ask for less or even zero compensation to make referrals (Brown et al. 2005). Our analysis shows that, overall, Group Buying is a more profitable social interaction scheme compared with Referral Rewards programs when interpersonal communication is very efficient or when the novice customers’ willingness to pay (without any interpersonal influence) is sufficiently high.
3
This paper contributes to the literature in several ways. First, it adds to the growing GroupBuying literature. Most existing research has focused on the impact of the power of buying groups on suppliers. For example, Che and Gale (1997) examine how an organized “buyer alliance” can use its power to influence the design of health insurance plans offered to its members. Inderst and Wey (2003) study the impact of size/power of buying groups on suppliers’ technology adoption decisions. Chen and Li (2008) explore whether buying groups, such as buying clubs for goods or utilities, should commit to exclusive purchase in the presence of competing sellers and how such commitment influences competition. Less attention has been given to Group Buying as a potentially advantageous marketing strategy from a seller’s perspective. For example, Anand and Aron (2003) model Group Buying as a mechanism which helps the seller to set the best price in the presence of demand uncertainty. Our paper explores another strategic function of Group Buying, that of facilitating consumer social interaction, i.e., motivating expert customers to promote the product to novice customers through interpersonal influence. Second, this paper contributes to the existing literature on quantity discounts (e.g., Buchanen 1953, Dolan 1987). The benefits for sellers to offer quantity discounts are well documented in the literature. For example, quantity discounts can be offered as a price discrimination mechanism to increase profit (Dolan 1987), as a cost-saving mechanism to shift the inventory holding cost to buyers (Buchanen 1953), and as a channel-coordination mechanism to increase channel efficiency (Weng 1995). The Group- Buying strategy studied in our paper can be reviewed as a unique type of quantity discount, which is neither offered to individuals who buy in large volume nor to channel members as suggested in the literature, but to a group of consumers who may not be related in any way before forming the buying group. Quantity discount as an information-sharing mechanism proposed by our paper has not been addressed in the literature, and recent advances in information technology, especially the Internet, are making such a mechanism increasingly practical in the market place. Third, this paper adds to the growing literature that integrates consumer social interactions into designing firm marketing strategies (e.g., see Godes et al. (2005) for a review). Various studies provide empirical evidence for the impact of interpersonal information sharing on consumer choices (Zhang 2010) and firm performance, such as sales (Chevalier and Mayzlin 2006) and revenue (Liu 2006). Others address marketing strategies that allow sellers to proactively motivate and manage social interactions. For example, Biyalogorsky et al. (2001) study how to motivate existing buyers to make after-purchase product referrals. Chen and Xie (2008) analyze when and how the seller should offer on-line consumer reviews as a marketing communication mix. Amaldoss and Jain (2007) examine how the interactions between profoundly different social groups require the seller to adjust its optimal marketing strategies, including pricing and product decisions. Guo et al. (2009) model a monopoly’s pricing decision in a durable goods market where social interactions play an important role in consumer product evaluation.
4
Chen et al. (2011) empirically demonstrate the differential impact of two types of social interactions, word of mouth and observational learning, on consumers’ purchase decision. Our research contributes to this stream of literature by building up a formal model to examine the benefit of a Group-Buying strategy in stimulating and managing social interaction. We show how the seller should design a Group-Buying offer, and when the offer can improve profits compared with traditional individual selling strategies and another scheme of social interaction (Referral Rewards programs). The main issue addressed in this paper, the social network marketing mechanism embedded in the Group-Buying selling format, is also of special practical importance. Recent advances in technology have significantly increased the importance of consumer social interactions as a market force by providing more tools for consumers to interact with and influence each other. This new trend creates both opportunities and challenges for marketers, and calls for more research on strategies like Group-Buying which allow firms to better facilitate and manage social interactions. In fact, the rising impact of Group Buying has become obvious in practice. For example, based on a recent study conducted by Experian’s Hitwise Intelligence, visits to social commerce and group buying websites for the week ended on April 17, 2010 were 72 times larger than the same week in 2009 (Hitwise Intelligence, April 23, 2010). The remainder of this paper is organized as follows: In Section 2, we first adopt a simple twocustomer setting to illustrate how a seller can use the Group-Buying selling format to stimulate social interaction and identify conditions under which Group Buying is more profitable than traditional individual selling strategies. We then extend our basic model to discuss how the comparative attractiveness of Group-Buying is influenced by the size of new customers acquirable through social interactions. We also model a Mixed Group-Buying format for which both the Group Buying and the individual selling offers exist. In Section 3, we compare the Group-Buying strategy with another commonly observed scheme of social interaction, Referral Rewards programs. Section 4 concludes the paper with a review of our main findings and a discussion of potential future research. 2. The Model 2.1 Assumptions We consider a seller facing two potential customers who differ in their knowledge of the seller’s product (i.e., one informed and one less-informed customer). Our objective is to use a simple model to illustrate the fundamental profit source of a Group-Buying strategy. Later, we generalize the basic model to discuss the impact of the size of different customer segments. Consumer Behavior. We consider two types of customers: (1) A less-informed customer with a
5
non-negative product valuation V0 ,2 and (2) An informed customer with a valuation V0 I , where
I 0 is the information gap between the two segments. The information gap among potential customers generally exists in practice for various reasons, such as consumer heterogeneity in searching for and absorbing product information (Alba and Hutchinson 1987), and the seller’s inability to spread product information to all potential customers (e.g., due to budget constraints or ineffectiveness of mass advertising). The information gap may prevent the less-informed customer from fully appreciating the benefit of the product, which generates a valuation gap I . Here, α measures the impact of unit information on the customer’s product valuation and may differ across product categories. 3 We assume that information shared with the expert customer will have a positive impact on the less-informed customer’ willingness to pay (WTP).4 Key notations are summarized in the Appendix. Seller Behavior. The seller may adopt a traditional individual selling strategy, i.e., transacting with an individual customer at a posted price. There are at least two such individual selling strategies: (1) a Margin strategy, under which the seller seeks a high profit margin by charging a high price and selling to the informed customer only, and (2) a Volume strategy, under which the seller expands the market to both segments by charging a low individual price. The seller may also consider a Group-Buying strategy, i.e., transacting with individual customers at a discounted price only after receiving orders from a certain total number of consumers. Under this strategy, the seller uses the discounted group price associated with a certain group size requirement to motivate the informed customer to “recruit” the less-informed customer. 5 To purchase the product at the discounted price, the informed customer must share information with less-informed customers (e.g., the discounted price offered to the buying group for the service package of BrowserCam motivates the registered customers to disseminate information about the package, such as its function, quality, usage methods, and price promotion, in order to persuade others to join the buying group). The price discount can be viewed as compensation paid to the informed customers for their information-sharing effort. The required compensation, denoted as I , increases with the agent’s effort level in information 2
Note the less-informed customer may be completely unaware of the product’s existence (i.e., V0 0 ). The parameter, , is assumed to be positive. When is negative (i.e., the informed customer has a lower product valuation than the less-informed customer), the seller will obviously not benefit from information sharing. 4 For example, if the informed and the less-informed customers differ only in their knowledge (but not in their preference/taste) about the product, information-sharing can have a positive impact on the less-informed customers’ WTP by serving as (a) a product awareness function (e.g., introducing the product to customers who are unaware of the product and have a zero product valuation); and (b) a value appreciation function (e.g., making less-informed customers aware of various functions of the product and how to benefit most from its usage). In the Technical Appendix, we relax this assumption to allow uncertainty in the impact of interpersonal information sharing and to show that our main results still hold qualitatively. 5 Note that our model does not require the seller to directly identify informed customers and offer them incentives to disseminate product information. Rather, the informed customer self-selects to join the group and work as a “sales agent” after observing the offer posted by the seller. 3
6
dissemination I (i.e., I 0 is the total amount of information shared with the new customer), and the interpersonal information sharing inefficiency (i.e., how difficult it is for the customer “sales agent” to reach and educate new consumers). To sum up, the seller considers the following three selling strategies: I. Margin strategy (setting a high price PH and selling to the informed customer only) II. Volume strategy (setting a low price PL and selling to both the informed and the lessinformed customer individually) III. Group-Buying strategy (setting a group price PG and selling to the buying group only if the group size reaches two) An important difference between the Group-Buying strategy and the two individual selling strategies is transaction time. Under the latter strategies, the customer with a product valuation higher than the posted price will make the purchase at the beginning of the period. Under Group Buying, it takes time for the informed customer to reach the less-informed segment and to persuade the customer into joining the buying group. Transactions are completed once the buying group achieves the required size, which, we assume, can only happen at the end of the period. We capture the impact of such a time delay by a discount factor 0,1 , which is assumed to be the same for both the customer and the seller. A
smaller discount factor, , corresponds to a larger discount rate, which captures a higher degree of impatience for both the customer and the seller. To ensure that market expansion is potentially desirable, we assume that the marginal cost is zero.6 2.2 Group-Buying vs. Individual Selling Strategies Individual Selling Strategy
Given that the seller sells to the informed customer only under the Margin strategy, but to both customers under the Volume strategy, the optimal price ( PH* , PL* ) and the corresponding profit ( H* , L* ) for the two individual selling strategies are straightforward, as given in (1): * * Margin Strategy: PH V0 I H V0 I 0 * L* 2V0 Volume Strategy: PL V0
(1)
Group-Buying Strategy
Under a Group-Buying strategy, the seller faces an optimal contract design problem. The seller specifies a Group-Buying price along with a group size requirement (i.e., PG and “two” in our basic model), which serve as a contract to “hire” the informed customer as a “sales agent” to exert a certain level of effort in information dissemination (i.e., to share information 0 I I with the less-informed customer). The optimal solutions (the price PG and the information-sharing level I ) are subject to two 6
A positive marginal cost will not change the results qualitatively.
7
constraints. The first is the participation constraint for the sales agent (i.e., the informed customer). That is, PG must be sufficiently low to compensate the informed customer for taking the effort to “recruit” others and delay her/his purchase until the group size requirement is satisfied.7 Given the required compensation I , and the discount factor , this constraint can be expressed as (V0 I - PG - I ) 0 . The second is the incentive compatibility constraint for the sales agent. The agent’s effort level is not directly observable by the seller, and the specified information sharing will be implemented only if it is the agent’s best choice. Given a group size requirement, the agent will choose the minimum informationsharing level that is sufficient to persuade the required number of less-informed customers into joining the buying group.8 This constraint can be expressed as (V0 I - PG ) 0 , which reveals that there is a oneto-one correspondence between the group price and the agent’s effort level. The seller’s problem, therefore, is to find the optimal group price (and a corresponding information-sharing level) that satisfies the above conditions to maximize its discounted profit: 9 x G 2 PG PG , I
s.t.
(V0 I - PG - I ) 0 (V0 I - PG ) 0
(2)
0 < I I
Lemma 1 summarizes the optimal solutions under the three selling strategies. (See the Appendix for proofs of all lemmas and propositions presented in this paper.) Lemma 1 (Group Buying and Individual Selling: the Basic Model) The optimal solutions of the three strategies are given in the following table. Selling Format Group Buying Margin strategy Volume strategy
Price I PH* V0 I PG* V0
PL* V0
Sales
Profit
2
G* 2 (V0
1 2
H* V0 I
2
2 I)
L* 2V0
7
Without loss of generality, we assume that, whenever the customer is indifferent between two decisions (either information exchange or joining the buying group, they will choose the one preferred by the seller. This holds true throughout the paper. 8 In a more general setting when there exist uncertainty in the impact of information sharing (i.e., the product may or may not be a match for the novice customer) and/or the hassle cost for the novice customer to engage in information sharing, the agent should choose the minimum effort level to ensure the less-informed customer will participate in information sharing (and join in the buying group only if the product is a match). Please see details of the model description and analysis in the Technical Appendix. Our results hold qualitatively in this general setting. 9 Note that Problem (2) is presented as a typical Principle-Agent problem between the seller and the informed customer (or customer “sales agent”). Since the constraints for the informed-customers are set based on the behavior rationality of the less-informed customers, satisfying these constraints will ensure the participation of the less-informed customer under the optimal solution.
8
As shown in Lemma 1, under Group Buying, the seller can charge a higher group price and obtain a higher profit (a) when the information gap is larger ( ( PG* ) more valuable ( ( PG* ) ( ( PG* )
0 , ( G* ) 0 ),
I 0 , ( G* ) I
), (b) when the information is
and (c) when the information sharing is more efficient
0 , ( G* ) 0 ).
Lemma 1 also illustrates the differences among the three strategies. While the Margin strategy charges the highest price ( PH* max PL* , PG* ), it serves only one segment of customers and leaves the lessinformed customer market untapped. Both the Group-Buying and Volume strategies allow the seller to expand the market, although in different ways. The former expands the market by offering a very low price, PL* V0 , whereas the latter expands the market by inducing interpersonal information sharing via a discounted group price, PG* V0
2 I . While the price under both strategies is lower than that under a
Margin strategy, Group Buying leads to a higher profit margin than the Volume strategy, PG* PL* , because information dissemination increases the valuation of the less-informed customer and hence market expansion can be achieved at a higher price. On the other hand, the Volume strategy leads to better liquidation because transaction is not delayed as under Group Buying. A formal comparison of the three maximum profits in Lemma 1 leads to Proposition 1, which states the conditions under which each strategy dominates the others. To ensure that Group Buying is a viable strategy, we assume that the discount rate is not too large such that, at least under some conditions, Group Buying is more profitable than the individual selling strategies.10 Proposition 1(Profit Comparison: the Basic Model) The optimal strategy is jointly determined by two factors: consumer information heterogeneity ( I ) and the inefficiency of interpersonal information- and knowledge-sharing ( ). Specifically, the optimal strategy is: (a) Group-Buying strategy when I L I I H , 2 ; (b) Volume strategy when I I L ; (c) Margin strategy when I I H ,; where I L 0 ; 0 I H if 1 , and I H otherwise ( I L , I H , 1 , 2 are defined in Appendix).
Proposition 1 is graphically illustrated in Figure 1. 11 The horizontal axis represents buyer information heterogeneity ( I ), and the vertical axis represents inefficiency of interpersonal informationand knowledge-sharing ( ). As shown in Proposition 1, the Group-Buying strategy is more profitable than the two individual selling strategies when two conditions hold: (a) Consumers have a mid-range of information 10 11
This implies that 1/ 2 . Figure 1focuses on the area where 1 because key patterns remain unchanged when is further increased.
9
heterogeneity ( I L I I H ), and (b) interpersonal information sharing is sufficiently efficient ( 2 ). While an information gap between the informed and the less-informed segments that is either too small or too large favors individual selling strategies, each occurs for different reasons. A wide information gap implies a significantly higher valuation for the informed customers compared with the less-informed customers. Hence, the seller is better off by capturing the high margin from the informed segment at the cost of foregoing sales to the low-valuation segment (i.e., the Margin strategy is optimal). When the information gap is narrow, the two segments do not differ greatly in their product valuation and, thus, expanding the market to serve both segments becomes more attractive. In order to do so, the seller has two different choices, Group Buying or the Volume strategy. Under the Volume strategy, the seller expands the market by charging a price that is as low as the valuation of the less-informed customer and does not benefit from the product information mastered by the informed customer. In contrast, Group Buying encourages the informed customer to disseminate such product information. When the information gap is very small ( I I L ), the Volume strategy is a more efficient way to expand the market since the seller may induce sales from the less-informed segment directly and promptly with a small individual price discount. The above discussions explain why, in general, a small information gap favors the Volume strategy, a large information gap favors the Margin strategy, and Group Buying can dominate both when the market presents a moderately-sized information gap. Information-sharing inefficiency, , affects the optimality of Group Buying in several ways. First, Group Buying cannot be advantageous when is too large ( 2 ). As shown Figure 1, the midrange information gap required for the optimality of Group Buying does not exist when is too large (i.e., I H I L if 2 ). Group Buying relies on informed consumers as “sales agents” to expand the market through interpersonal influences, and the attractiveness of this strategy depends on how much incentive the seller has to offer (i.e., group price discounts) to compensate for this information-sharing effort. The higher the interpersonal information-sharing inefficiency is, the more costly and less attractive it is for the seller to expand the market via Group Buying. Second, the required mid-range information gap for the optimality of Group Buying becomes wider as decreases (i.e., I H 0 ,
I L 0 ,
see Figure 1). Furthermore, when social interaction becomes very efficient, the optimality of
Group Buying is no longer subject to an upper bound of the information gap ( I H if 1 ). These results suggest that a higher level of interpersonal information-sharing efficiency (i.e., a small ) weakens the conditions under which Group Buying dominates the traditional individual selling strategies. Furthermore, the importance of the information, α , also influences the comparative profitability of the three strategies; First, a larger α generally makes the Volume strategy less attractive compared with
10
Group-Buying and Margin strategies. This is because α measures the gap in willingness to pay between the informed and less-informed customers, and the Volume strategy ignores such a gap by selling all customers at a price acceptable to less-informed customers. Second, a large α makes Group-Buying more attractive compared with the Margin strategy when the information-sharing efficiency is sufficiently high (
1 0 , i.e., the interval within which the Group-Buying strategy always dominates the Margin strategy
becomes wider when α increases).
Information Sharing Inefficiency
Figure 1: GB vs. Individual Selling: the basic model ( 0.8, V0 10, 0.65 )
IL IH
Margin Strategy
2 Volume Strategy
1
Group Buying IL
IH
Information Gap
These findings have some practical implications. Our results suggest that Group Buying, in general, has a higher chance of dominating individual selling strategies when the market possesses a moderate level of information heterogeneity. Given that the information gap is influenced by various market/product characteristics that may change over time, our finding implies that the seller may (and perhaps should) dynamically adjust its selling strategy across the product life cycle. For example, for newly launched high-tech products, a significant information gap between expert and novice customers may suggest that the seller should adopt the Margin strategy, selling only to expert customers who appreciate the product the most. When the market becomes more mature and the information gap among potential customers decreases (e.g., because of the information disseminated through advertising and/or natural WOM), Group Buying may become a profitable selling strategy to attract less-informed consumers who do not fully appreciate of the value of the new products. As the market further evolves, information heterogeneity is further reduced and the Volume strategy becomes more attractive to reach the less-informed segment, if such exists. On the other hand, for some really new/innovative product
11
categories/brands, the information gap may first increase (e.g., because it takes time for the seller to disseminate product information and educate expert customers) and then decrease. In this case, the different strategies may be adopted in reverse order at the initial stages of new product launch. Our results also suggest that Group Buying is more likely to be advantageous when consumer social interactions are more efficient. Such efficiency can be affected by cultural and social factors. For example, a small may apply to countries with a collectivistic culture, where individuals have a higher need for information sharing and a strong tendency for conformity. Communication efficiency may also be affected by the level of trust in social interactions: A smaller is expected in a social environment where individuals have a high level of trust, which suggests that Group Buying can be a more efficient product promotion format when offered through online social networks or through websites with many loyal visitors (e.g, who are familiar with each other’s user ID and have developed trust over time). Finally, it is expected that the development of communication and information technologies will continue to increase interpersonal information-sharing efficiency (i.e., leading to a smaller ) and hence will increase the attractiveness of Group Buying relative to traditional selling strategies. To validate the generality of our main finding and derive additional insights, we extend our model in several dimensions. We present two of such extensions below and several others in the Technical Appendix. 12 2.3 Extension One: Distribution of Buyer Segments
Our basic model considers the case with equal size of informed and less-informed customers. We now extend the basic model to explore how the distribution of the two buyer segments may affect the relative attractiveness of Group-Guying. Specifically, we allow the relative size of the two segments to vary by assuming that, there are n informed customers and n less-informed customers in the market, where represents the ratio of the number of less-informed to informed customers. The other assumptions in Section 2.1 remain the same. As in the basic model, the seller faces three strategic choices. The seller sells to n informed customers only at a high price Ph V0 I under the Margin strategy, and to all n(1 ) customers at a low price Pl V0 under the Volume Strategy. When a Group-Buying strategy is adopted, the seller chooses a group price to motivate the n informed customers to acquire the n less-informed new customers. 13
12
For example, we extend our model by allowing (a) uncertainty in the impact of information sharing, and (b) both the information sharing impact uncertainty and a positive cost of information receiving for the less-informed customer. Our main results hold in each of these cases (see detailed analyses in the Technical Appendix). 13 Here, we assume that the seller sets the group size requirement as n(1 ) . We relax this assumption and offer some brief analyses about the group size issue in the Technical Appendix.
12
Assume that the n informed customers exert the same level of effort, i.e., each informed customer acquires less-informed new customers. 14 Accordingly, the compensation required by the “sales agent” for such an effort is I . Therefore, similar to the basic model, in designing the Group-Buying offer, the seller solves the following optimization problem: x G n (1 ) PG pG , I
s.t.
(V0 I - PG - I ) 0 (V0 I - PG ) 0
(3)
0 < I I
Note that the basic model in (2) is a special case of the more general model in (3). We prove in the Technical Appendix that, as in the basic model, an “information sharing” equilibrium can exist under the more general setting in (3). More interestingly, the relative attractiveness of Group-Buying strategy is affected by the distribution of buyer segments. We summarize these results in Corollary 1. Corollary 1 (The Distribution of Buyer Segments) (1) An “information sharing” equilibrium can exist for problem defined in (3), and at such an equilibrium, Proposition1 holds but with a new set of boundaries, I L , I H , and 2 . (2) The distribution of buyer segments affects the seller’s incentives to adopt Group-Buying strategy such that, a larger size of the less-informed customers enhances the profit advantage of Group-Buying relative to Margin strategy but weakens the profit advantage of Group-
Buying relative to Volume strategy ( i.e.,
I L I H 0 and 0 ).
Corollary 1 reveals that our findings derived from the basic model still hold qualitatively when allowing the relative size of the less-informed customer segment to vary. It also reveals some additional insights. Specifically, as the comparative size of the less-informed segment, , increases, Group Buying
tends to be more attractive than the Margin strategy ( I H
( I L
0 ).
0 ),
but less so than the Volume strategy
This is because, compared with the Margin strategy, Group Buying expands the market, and this
advantage increases as the size of the new market becomes larger. On the other hand, compared with the Volume strategy (which acquires new customers directly through a low price), Group Buying expands the market with a time delay and at the cost of compensating informed customers’ effort in reaching and educating less-informed customers. This relative disadvantage increases with the size of the novice
14
In the Technical Appendix, we check whether this equal information-sharing assumption may hold true in equilibrium. We found that, for the several game structures we examined, such an “information-sharing” equilibrium always exists but may not always be unique (i.e., a “non-information sharing” equilibrium under which nobody works on information sharing may appear even though the seller has offered enough group discounts). We offer some further discussion about this interesting issue in the Conclusion section.
13
segment. 2.4 Extension Two: Mixed Group-Buying
Our basic model considers a Pure Group-Buying strategy, i.e., the seller either sells through buying groups or to individual customers directly. In real markets, the seller can also consider Mixed Group-buying, i.e., adding a Group-Buying offer to the existing individual selling channel and allowing customers to choose between joining the buying group or buying individually. It is easy to see that such a mixed strategy is unable to improve profit in markets with homogenous informed customers (e.g., the basic model) where all informed customers would make the same choice. However, the seller may have the opportunity to benefit from adopting Mixed Group-Buying strategy when informed customers are heterogeneous. In this section, we extend the basic model to explore this potential benefit by allowing the informed customers differ in their willingness to delay the transaction. Specifically, we adopt a multicustomer setting as in Section 2.3 with two informed and two less-informed customers ( n 2, 1 ).15 We assume that one informed customer has a higher delay factor than the other ( 1 and 2 with 1 2 ).16 All other assumptions in Section 2.3 remain. With Mixed Group-Buying strategy, the seller offer both an individual buying offer at a full price ( PI ,) and a Group-Buying offer at a discounted group price ( PMG ), under which impatient informed customers buy individually and the patient informed customers share information with two less-informed consumers and form a buying group of size three. The seller’s problem (in choosing PI , PMG , and I to maximize its total profit) is similar to that in Section 2.2 and Section 2.3, but with three additional constraints (see the last three constraints in (4)). The first is the participation constraint for the impatient informed customer who purchases individually, i.e., V0 I PI 0 . The second is the incentive compatibility constraint for the impatient informed customer to choose purchasing individually over participating in Group-Buying and delaying the transaction, e.g., V0 I PI 2 (V0 I - PMG -
2
I ) .17 The third is the incentive compatibility
constraint for the patient informed customer to choose Group-Buying over purchasing individually, e.g., (V0 I - PMG - 2I ) V0 I - PI . Therefore, the seller has the following optimization problem:
The results in this section can be generalized to the general ( n, ) multi-customer setting as in Section 2.3. We may also allow novice customers’ delay factors to be different. Since considering this heterogeneity will not change the results, we continue to assume that novice customers have the same delay factor . 17 If the impatient informed customer participates in Group Buying, the two informed customers share the task of recruiting one less-informed customer. Therefore, the required compensation for information sharing is I . 15 16
2
14
Max
PMG , PI , I
s.t
MG PI 3 PMG
1 (V0 I - PMG - 2I ) 0 (V0 I - PMG ) = 0
(4)
0
