on Online Group Buying in China

LEI Hong-zhen, LU Huan

International Business School of Shaanxi Normal University, Xi’an Shaanxi China luhuan@snnu.edu.cn

Abstract：Since 2008,the first Online Group Buying website was established in the U.S. ,which integrate online shopping and offline group purchasing in a new e-business model and began to spread all over the world. In china, however, most published researches on online group buying are made by qualitative analyze，Although there have been lots of papers on online shopping, these achievements still have not been fully applicable in this new e-commerce model—“Groupon”. This study use the structural equation model to explore the relationship between customer satisfactions, brand trust and customer loyalty. The result told us that at the present stage, customer satisfaction can not directly lead to customer loyalty，brand trust is needed as an intermediary variable which to link them. As while as online group buying websites want to elevate customer satisfactions, they requires lots of investment in brand marketing to delivering brand image to customers, and finally promote satisfactions convert to customer loyalty.

Key words：online group buying; customer loyalty; brand trust; groupon

1. Introduction

Since 2000, a few of Group Purchasing websites have emerged in China, just like Fence Net, China Group Purchasing Net and other websites, which organize customer groups to merchandise with business in low price. However, until 2008, the first Online Group Buying website was established in the U.S. which based on the model named Groupon, this business model integrate online shopping and offline group purchasing in a new e-business model which now began to spread all over the world. January 2010, China's first Online Group Buying website start operating, before the end of 2010, the number of Group Buying websites in China has reached 16. By the end of 2013, there have been 6246 group buying websites, although many of them have closed. By taking the advantage of IM, SNS, MicroBlog and other social networking tools, Online Group Buying model creating a feeling of scarcity of goods, to boost the spending power of user participation. With the formatting of internet users’ habits, online group buying sites have been shift from the model copy to the phase of customer loyalty building.

Customer loyalty is important for online retail business to gain competitive advantage and to increase their profit, but from Academic perspective, there is no common understanding on the factors which affecting online shopping customer loyalty [1-2].Some scholars believe that the perceived value, customer satisfaction, customer loyalty and trust are likely to have impact on customer loyalty[3]. But from retail business practice, the relationship between customer values, customer satisfaction, customer trust and customer loyalty are very complex, and the relationship would be changed if there is any interference from outside[4]. Online Group Buying is a new e-commerce model, research achievement on online shopping can not be fully applied. In this paper, from industry research and documentation reading, we build up a conceptual model of customer loyalty at the beginning. Questionnaire survey on consumers, using structural equation modeling to test the relationship between customer satisfaction, brand trust and customer loyalty, we find out problems in China's online group buying sites, and offer related advice to China online group buying sites on how to making a marketing strategy.

2. Literature review and conceptual model

(1) Customer Loyalty

From the Traditional marketing view, pursuit for financial Maximize is the fundamental purpose of enterprises, which requires customers continue to buy their products or Service. Therefore, customer loyalty is always measured by customer behavior, such as the purchase ratio, the willingness to buy, and repeat purchasing behavior. Dick et al (1994) regard that when a customer with positive attitude and repeat patronage behavior to providers, that means they have a real and lasting loyalty[5]. Loyal customers not only prefer to use the service, but will also recommend the service to others. Hallowell (1996) puts the customer loyalty as a brand emotional to products or organizations[6]. Tucker (1999) defines customer loyalty as repeat purchases more than three times[7]. Loyalty can also be measured by customer's first choice of next time purchase, and this reflects the different attitude toward to different Online Group buying sites. (2) Customer satisfaction and customer loyalty

Customer satisfaction is an evaluation of experience and judgments on a particular service, which is often considered as the antecedent of customer loyalty. Anderson and Sullivan (1993) believe that satisfaction is the basic factor of consumer behavior, if there are more retention, consumers will be more satisfied with it[8]. Ravald et al (1996) put that keep loyalty and get profits are closely related to create value for consumers. Cronin (2000) thought that satisfaction is an emotional response after experience a range of services[9]. Sirdeshmukh (2002) said that consumers will remain loyal if they can get more value from this shop than its competitors[10]. Harris et al (2004) research on the customer satisfaction and customer loyalty of online shopping, and believe that online purchasing satisfaction affects customer loyalty and behavioral loyalty apparently[11]. Therefore, we propose:

Hypothesis 1: Customer satisfaction has a positive impact on customer loyalty. (3) Customer satisfaction and customer value

In the concept of customer-oriented, customer satisfaction is most closely related to customer value, most researchers believe that they are complementary paradigms, not mutually exclusive. Jones et al (1995) believe that customer value drives customer satisfaction directly, and deliver superior value to customers is the only way to obtain customer satisfaction for organization[12]. Fornell et al (1996) on the basis of emotion cognitive, argue that value judgments will affect the satisfaction level and satisfaction is the result of customer value[13]. Stalk (1997) thought that customer value describes the relationship among organization, customer and services, customer satisfaction represents the customer reaction from a particular product[14]. Parasuraman et al (2000) presented the main driving factors of customer value is consist of product quality, service quality and price[15. Lam (2004) believe that customer value is a balance between obtain and expend, customer satisfaction is a positive emotion[16]. Holbrook (2006) defines customer value as an overall evaluation of the product or service[17]. Choosing goods and services from different online group buying sites, the standard is which one can bring them the optimal value. Therefore, we hypothesized:

Hypothesis 2: customer value has a positive impact on customer satisfaction. (4) Brand trust and customer loyalty

In online shopping, consumers rarely have the opportunity to view product before purchasing, if there is not enough trust, transactions between consumer and business will stay in the requirements phase. Weiying Zhang (2010) thought that in a globalized market, consumers don’t product, producers don’t consume, trust is the most important element in market economy. Moorman (1993) defined trust as a confidence on their trading partner and willingness to rely on[18]. Morgan et al (1994) believe that customers trust makes customers pay great attention to the trade relations with service providers, thereby creating a sense of commitment and loyalty[19]. For lack of effective constraint rules, online consumers will keep away from the web retailers which are not being trusted, and complexity will lead them to avoid from buying online[20]. Therefore, retailers strive to build customer trust can reduce consumer perceived risk, and enable consumers to have faith in retailers’ future

behavior[21].if Online Group Buying sites can create and maintained consumer trust, then they can reduce consumer perceived risk and enhance consumer loyalty[22]. Therefore, this study hypothesized: Hypothesis 3: brand trust has a positive impact on customer loyalty. (5) Brand trust and brand image

Under the environment of e-commerce, the physical distance between buyers and sellers arising risks of uncertainty which makes trust particularly important. Lau (1994)thought that a good brand image, is conducive to format brand \"rumors\" among consumers which comes from corporate marketing strategy to please consumers, let consumers feel that business is real concern and focus on them , thus contributing to building consumer trust [23].Keller (1993) thought that brand image is an important component in brand equity, define brand image as brand association based on Customers memory , attributes, attitudes and interests constitute the brand association[24].Shaping the brand not only help consumers to identify the level of service quality, but also drive business to success[25]. Communications and advertising are seen as an important method which helps Corporate to reduce customer perceived risk[26]. Zeithaml (1988) believe that in an increasingly homogeneous product market, a good brand image can improve the quality of consumer awareness, and also reduce the cost of information gathering[27]. Joon (2002) thought there was lots of factors impact on improves online consumer confidence, including web interface design, corporate reputation, information quality, transaction security, etc[28]. Ways to enhance customer brand trust through deliver brand image are also useful in Online Group Buying sites. Therefore, this study hypothesized:

Hypothesis 4: brand image has a positive impact on brand trust. (6) Intermediary effectiveness of brand trust

Oliver (1980) pointed out that satisfaction is an emotional status which was lead by expectations and perceptions, product and service satisfaction is the primary motivation for continue use[29].However, a survey from United States by Bain shows that nearly 65% to 85% customers will Shift to another sites despite they are satisfied with the initial company's products[30]. In virtual retail market, lack in trust is considered to be the biggest obstacles which restricted buyers and sellers online trading smoothly. Chaudhuri et al (2001) found that consumer satisfaction can format consumer brand trust, the higher customer satisfaction, the deeper customer trust. Ranaweera et al (2003) believe that between the satisfaction and loyalty, \"customer trust” plays the intermediary role, and that means customer satisfaction impact on customer loyalty through customer trust[31]. Therefore, this study hypothesized:

Hypothesis 5: customer satisfaction has a positive impact on brand trust.

Hypothesis 6: customer satisfaction impact on customer loyalty through customer trust.

Based on above Hypothesis, we construct a conceptual model for this paper, as shown in Figure 1:

Customervalue H2Customer satisfactionH1Customer LoyaltyH5brand imageH4BrandtrustH3 Fig.1 conceptual model

3. Study Design

(1) Questionnaire development and design

In order to ensure questionnaire reliability and validity, we adopt mature Questionnaire relatively .Customer satisfaction and customer loyalty items are from Anderson, Srinivasan (2003). Trusted brand items mainly from

Lee, Turban (2001), customer value is reference to Kotler's triangle customer value model, Include quality, service and value; brand image reference Lau’s (1994) study, measured in three areas from the brand personality, brand communication and brand reputation. To ensure the validity of questionnaire, we combine relevant theoretical concepts and interview 10 Graduate with rich experience on online group buying, finally modify the original items and development some new items. Questionnaire is using Likert 5-point scale method, there are 37 items being asked in questionnaire, 1 means totally disagree and 5 shows full agreement. (2)Data collection

Buying Online is completed on internet, participants are mostly young people and internet users, in order to obtain more universal sample, so we use online method to fill out all the questionnaire. in 220 questionnaires, there are 197 valid questionnaires, the recovery rate of valid questionnaire was .6% (invalid questionnaire selection criteria: those who have no experience on or fill out the answer with too short time). We surveyed Consumers from big city in China, like Beijing, Guangdong, Chongqing and Xi'an, et al. In samples collected, women account for 56.6%, male 43.4%, the Samples we collected conform to the characteristic of Online Group Buying industry development.

4. Data Analysis

This study adopt SPSS 17.0 and AMOS 7.O to analysis model. SPSS 17.0 analysis factors reliability and validity, AMOS 7.O testing model hypothesis. (1) Reliability and validity test

Reliability refers to the consistency of measurement results. Results show that Cronbach's α of each factor are greater than 0.7, indicating that these variables show good internal consistency. Validity refers to the validity of the measurement results. Using factor analysis to test validity of questionnaire, KMO of all factors are greater than 0.7, while Bartlett test results also reject the Original hypothesis (sig. = 0.000), illustrate the items are suitable to factor analysis.

Tab.1 reliability and validity testing

Latent variable Customer Value Customer Satisfaction Brand Image Brand Trust Customer Loyalty

Cronbach's α 0.921 0.948 0.915 0.5 0.883

KMO 0.2 0.749 0.908 0.706 0.718

(2) Model fitting

In this study, we use structural equation model to analyze the relationship between variables. If the ratio of Chi-square and degrees of freedom is small, that means matrix fit to observe data in the model assumed. In general if the ratio is less than 2, we can assume that the model has a good fit .The results showed that the ratio of chi-square and degrees of freedom is 2.58, model is in a poor fit. When the data is completely fit the model, three indicators CFI, TLI, IFI are equal to 1, in general it can be allowed if the indicators are greater than 0.9 .The results showed that the three indicators are 0.840,0.824,0.841, indicates that the model has a poor fit. RMSEA means Root Mean Square Error of Approximation. Generally, RMSEA less than 0.08 indicate a good model, but the result shows that RMSEA is 0.111. AIC and BCC is indicators to compare streamlining of two different models, the smaller the better. ECVI also used to compare fit goodness between different models, smaller value means higher consistency. For the fit index has not reached the ideal standard, model need to be amended.

Amos provides Modification Indices for Model Building. By releasing some limited or adding some new path, can make model structure more reasonable, and improve the model fit goodness. Residuals which have not meet

the standard of Modification Indices should be free estimated, such as increasing the correlation path between Z1 and Z4, then you can reduce the chi-square value. After model revised, fit indices are all improved, as shown in Table 2:

Tab.2 the results of commonly used fit index

Fit index Before amendment After amendment

Chi (df) 823.390 (319) 535.013（311）

CFI .840 .929

TLI .824 .920

IFI .841 .930

RMSEA .111 .075

AIC 941.390 669.013

BCC 974.430 706.533

ECVI 7.355 5.227

After model revised，ratio of chi-square and degrees of freedom decreased to 1.72, which reflect the model has a good fitting, CFI, TLI, IFI were greater than 0.90, RMSEA less than 0.08. the value of AIC、BCC and ECVI are all low down after amendment. Model has a good fitting, the revised structural equation model shown in Figure 2.

.23.80.58.49.53.34.44.32.36.42e1a11e2a21e3a31e4a411.39e5.1111.611.66e6a61e7a71e8a81e9a911.001.321.46a5Brand Image1.831.741.72.251Z11.04.26.08111.09.07e10.23e11.37e12a10a11a121.111.001.09Brand Trust1Z2.96.03Z51.29.12.37.16111Customer Loyalty1.00.92.90a25a26a27111.18e25.38e26.61.24e27e13.09e14.17e15a13a14a151.081.001.04CustomerSatisfaction.951.16Z3.99.93.96Customer Value.92.1.47.781.97.701.061.00Z4a16e161.35a17e17.281.46a18e181.49a19e19-.08a20e201.49a21e211.88a22e221.38a23e231.40a241.47e24

Fig.2 revised structural equation model

(3) Hypothesis testing

coefficient estimates on optimal path model is shown in Table 3, path coefficient from customer satisfaction to customer loyalty is 0.115, CR is 1.267, P is 0.205, this indicate that the probability of path coefficient equal 0 amounts to 20.5%, larger than the significance level at 5%, H1 can not passed. In addition, other relationships between potential variables are all passed the test. From the test we know that Customer value has a positive influence on customer satisfaction, brand trust has a positive influence on customer loyalty ,brand image has a positive influence on brand trust, and customer satisfaction have a positive impact on brand trust. Therefore, the hypothesis H2, H3, H4 and H5 has confirmed. Hypothesis of Customer satisfaction to customer loyalty have not substantiated, but it does not influence the further test of H6. In AMOS, the mediation effect can be measured by direct and indirect effects. Standardized direct effects and indirect effects shown in Table 4:

Tab.3 estimates the optimal model for the path coefficient

Hypothesis

H1 H2 H3 H4 H5

Tab.4 Standardized direct effects and Standardized indirect effects

CS BT CL

Direct effect is the direct impact from one variable to another, such as \"customer satisfaction → customer loyalty\satisfaction→brand trust→customer loyalty.” If the direct effect is greater than the indirect effect, that means mediating variables has not working. If the direct effect less than indirect effect, means intermediate variables has play an intermediary role. In Table 4 , the direct effect from customer satisfaction to customer loyalty is 0.117, indirect effect is 0.358(0.358> 0.117), so H6 (customer satisfaction→brand trust→customer loyalty) is Confirmed, brand trust is the intermediate variables between customer satisfaction to customer loyalty.

CV

.5（.000） .000（.362） .000（.425） CS .000（.000） .404（.000）

BI

BT .000（.000） .000（.000） .885（.000） CL .000（.000） .000（.000） .000（.000） .000（.000） .5（.000）

.117（.358） .000（.571） relationship CS→CL CV→CS BT→CL BI→BT CS→BT

Estimate

.115 .954 .962 1.041 .366

S.E. .091 .094 .119 .205 .056

C.R. 1.267 10.151 8.063 5.081 6.549

P .205 *** *** *** ***

RESULT N Y Y Y Y

Note: \"***\" Indicates significant at 0.01 levels

Note: Figures in brackets is the indirect effect

5. Conclusions and recommendations

Although hypothesis 1 has not been verified, but it is also meaningful, it shows that in china, Online Group Buying sites can not build customer loyalty easily if they just rely on enhance customer satisfaction. Meanwhile, the verification of brand trust mediating effect has a great practical significance, which indicates that only by establish brand trust first, customer satisfaction would be transform into customer loyalty effectively. From the path coefficient view of customer satisfaction on brand trust, each unit customer satisfaction increased, brand trust will increase 0.37 units. This means that in order to build brand trust through customer satisfaction, Online Group Buying sites should also take a lot of brand marketing investment at the same time. From path coefficient view, each unit brand trust increased, customer loyalty will increase 0.96 units. In china, Online Group Buying websites are all using similar interface and selling similar products, so it is hard for customers to choose commodity from thousands of providers. With incomplete information, consumers in the market just want to seek products in low price, this lead the online group buying industrial looks like perfectly competitive market. For the first choices have little impact on the second choice, consumer’s buying behavior will be single-game if sites haven’t build brand trust.

Based on customer satisfaction perspective, Online Group Buying sites should established selection and monitoring mechanisms on supplier, and strengthen the relationships with them. Result shows that the factor loading of “Online Group Buying sites usually offer some super-value products\" with \"customer value\" is larger than others, \"will get the same service as full-price customers \" with \"I believe it has a consistent quality as what described online” is followed, factor loading of items on the “customer service” with\" customer value \" is lower. From factor loading we know that consumers are sensitive to price and quality, but they are not concerned about purchasing experience strongly. From the view of path coefficient, customer value increase one unit, customer satisfaction will increase 0.95 units. Price and product quality are both dependent on the

relationship between product suppliers and online group buying sites. Online group buying sites would try their best to meet customer needs in the purchase process, but it is hard for them to monitor and guarantee during consumption process. Low price is the main features in online group buying, when price turn into supplier's marketing tools, it will greatly worsen the consumer experience. Products can not be consumed smoothly and customers would not be well treated, if there are potential services or products need to buy before consume. Only by establish good cooperation mechanism with suppliers, online group buying sites can enhance customer value, and make customer satisfied.

Based on the perspective of brand trust, Online Group Buying sites should set up their own brand and creates brand trust among customers. Study shows that factor loading of \"I have some knowledge and understanding on this website\" with \"brand image\" is larger than others, \"I think the site has a higher position in the market\" with \"I think the brand is well-known\" followed by, items reflect the Brand personality such as \" focus industry \"with\" focus on local” has a lower factor loading on \"brand image”. From Factor loading we know that consumer are most concern about brand awareness and market position owned by Online Group Buying sites, but they have little attention to the differential positioning. From the view of path coefficient, brand image increased one unit; Brand trust will increase 1.04 units. Brand image establish is most based on market visibility and market position, which will not conflict with Brand personality. Online Group Buying sites which tends to be large and wide coverage, can attract all types of customers for the purpose of Mindless consumption. Sites focus on a certain field or a kind of product, the professional will let customers generate the sense of safe and trust. Sites targeted in regional are easier to cooperate with local businesses, it will be more efficient on services and goods realize. Different between perfectly competitive market and monopolistic competition market is that in monopolistic competition market, products are homogeneous but difference. Therefore, brand image establish is the only way to shifted Online Group Buying sites from perfect competition into monopolistic competition. Through diverse means to deliver brand image, build brand trust, and then online group buying sites can deliver consumer concept to actual and potential customers by their own Brand equity.

[1] Anderson R E, Srinivasan S S．E-Satisfaction and E-Loyalty A Contingency Framework[J].Psychology and Marketing，2003，20(2)：123-138.

[2] Luarn P, Lin H H．A Customer Loyalty Model for E-Service Context[J].Journal of Electronic Commerce Research，2003，4(4)：156-167.

[3] Srinivasan S S，Anderson R，Ponnavolu K. Customer loyalty in e-commerce：an exploration of its antecedents and consequences [J]. Journal of Retailing ，2002 ，（78）：41-50.

[4]WANG Xu-hui，XU Jian. Study on online customer loyalty under the moderating role of switching cost [J].China Industrial Economics，2008，（12）：113-123.

[5] Dick A S，Basu K. Customer Loyalty：Toward an Integrated Conceptual Framework[J].Journal of the Academy of Marketing Science, 1994，22（2）：99-113.

[6] Hallowell R. The relationships of customer satisfaction, customer loyalty, and profitability: an empirical study[J]. International Journal of Service Industry Management, 1996,7(4):78-86.

[7] Tucker R B. Earn Peoples Loyalty：Here s How Building the loyalty of your supporters is the most important thing you can do[J].Nonprofit World，1999,17(9)：36-39.

[8] Anderson E W，Sullivan M W. The Antecedents and Consequences of Customer Satisfaction for firms[J].Marketing Science, 1993,12 (2)：125-143.

[9] Cronin J J，Brady M K，Tomas G. Assessing the effects of quality ,value ,and customer satisfaction on consumer behavioral intentions in service environments[J].Journal of retailing,2000,76(2)：193-218. [10]Sirdeshmukh D，Singh J, Sabol B. Consumer Trust，Value，and Loyalty in Relational Exchanges[J]Journal of marketing，2002，66(1)：15-37.

[11] Harris L C，Goode M H. The Four Levels of Loyalty and the Pivotal Role of Trust A Study of Online Service Dynamics [J].Journal of Retailing，2004，80(2)：139-158.

[12] Jones T O，Sasser W E. Why satisfied customer defect[J].Harvard business review，1995，73(6)：1-14.

[13] Fornell C，Johnson M D，Anderson E W，Cha J，Bryant B E. The American Customer Satisfaction Index: Nature, Purpose and Findings[J]. Journal of Marketing, 1996,60(4):7-18. [14] Stalk G J. Customer Value：The next Source of Competitive Advantage [J].Journal of the Academy of Marketing Science，1997，25(2)：41-51.

[15] Parasuraman A，Grewal D. The Impact of Technology on the Quality value loyalty Chain: A Research Agenda [J]. Journal of the academy of marketing science, 2000, 28(1): 156-174. [16] Lam S Y et al. Customer Value，Satisfaction，Loyalty and Switching Costs：An Illustration from a Business to Business Service Context[J].Journal of the Academy of Marketing Science，2004，32(3)：293-31.

[17] Holbrook M B. Consumption Experience, Customer Value, and Subjective Personal Introspection: an illustrative photographic Essay[J].Journal of Business Research，2006，(59)：714–725. [18] Moorman C，Deshpande R，Zaltman G.Factors Affecting Trust in Market Research Relationships[J].Journal of Marketing，1993，57（1）87–101. [19] Morgan R M，Hunt S D.The Commitment-Trust Theory of Relationship Marketing[J]. Journal of Marketing，1994，58(3)：20-38. [20] Jarvenpaa S L，Todd P A. Consumer Reactions to Electronic shopping on the world wide web[J].International Journal of electronic commerce，1997，1(2)59-88. [21] Schoorman F D，Mayer R C，Davis J H. An integration Model of organizational trust[J].Academy of management review，1995，20(3)344–354.

[22] Chaudhuri A，Holbrook M B. The Chain of Effects from Brand Trust and Brand Affect to Brand Performance：The Role of Brand Loyalty[J].Journal of Marketing，2001，65（2）81–93.

[23] Lau G T，Lee S H. Consumers' Trust in a Brand and the Link to Brand Loyalty[J].Journal of Market Focused Management，1999，4（4）：341-370. [24] Keller K L. Conceptualizing，Measuring，and Managing Customer based Brand Equity[J]. Journal of Marketing，1993，57 (1)：1-22.

[25] Berry L L. Cultivating Service Brand Equity [J].Journal of the A academy of Marketing Science，2000，28（1）：128-137. [26] Davies M. Image problems with financial services[J].Management Decisions，1996，34(2):- 71

[27] Zeithaml V A. Consumer Perceptions of price, Quality and Value: A Means-end Model and Synthesis of Evidence [J].Journal of marketing，1988，52：2-22.

[28] Joon Y S.The Antecedents and Consequences of Trust in Online-purchase Decisions[J].Journal of Interactive Marketing，2002，16(2): 47-63.

[29] Oliver R L.A cognitive model of the antecedents and consequence of satisfaction decisions[J].Journal of Marketing Research，1980，17 (4)：460-469.

[30] Reichheld F F，Schefter P. E-Loyalty Your Secret Weapon on the Web[J]. Harvard Business Review，2000，7：105-113. [31] Ranaweera C，Prabhu J. The Influence of Satisfaction Trust and Switching Barriers on Customer Retention [J].International Journal of Service Industry Management，2003，14（4）：374-395.