Does relationship marketing matter in online retailing.docx

《Does relationship marketing matter in online retailing.docx》由会员分享，可在线阅读，更多相关《Does relationship marketing matter in online retailing.docx（12页珍藏版）》请在淘文阁 - 分享文档赚钱的网站上搜索。

1、J. of the Acad. Mark. Sci. (2016) 44:206217 DOI 10.1007/s11747-015-0429-6 ORIGINAL EMPIRICAL RESEARCH Does relationship marketing matter in online retailing? A meta-analytic approach Varsha Verma & Dheeraj Sharma & Jagdish Sheth Received: 13 September 2014 / Accepted: 26 January 2015 / Published onl

2、ine: 17 February 2015 # Academy of Marketing Science 2015 Abstract Building on the meta-analytic model suggested by Palmatier et al. Journal of Marketing, 70, 136153, (2006), this study extends the relationship marketing framework to the domain of online retailing to identify what strategies help bu

3、ild relationships with online customers. Specifically, this meta-analytic study identifies key antecedents and conse- quences of relationship marketing in online retailing. The study also examines the relationship between the four mediatorstrust, commitment, relationship quality, and relationship sa

4、tisfaction and the antecedents and consequences of relationship marketing. Similarity and seller expertise were found to have the strongest impact on relational mediators, and word of mouth was the most critical outcome of relationship marketing efforts. The model proffered in this study will motiva

5、te hypotheses to be examined by future re- searchers. The model also helps managers to identify the key drivers of relationship marketing in online retailing. Keywords Relationship marketing . Customer relationship management . Loyalty . Online retailing V. Verma (*) Department of Marketing, Indian

6、Institute of Management-Ahmedabad, D34, R18, IIM New Campus, IIM, Vastrapur, Ahmedabad, Gujarat 380015, India e-mail: varshaviimahd.ernet.in D. Sharma Department of Marketing, Indian Institute of Management-Ahmedabad, Wing 9, IIM Old Campus, IIM, Vastrapur, Ahmedabad, Gujarat 380015, India e-mail: d

7、sharmaiimahd.ernet.in J. Sheth Charles H. Kellstadt Professor of Marketing, Emory University, Goizueta Business School, 1300 Clifton Road, Atlanta, GA 30322-2710, USA e-mail: Relationship marketing in online retailing Online retailing has grown exponentially in the last 10 years. Online retail sale

8、s have grown every year since 2000; in the past 5 years, global online retail sales have increased 17% yearly from $236 billion in 2007 to $521 billion in 2012 and are expected to reach $1248.7 billion by the end of 2017 (Kearney 2013; MarketLine 2013). The consumers of today are increasingly sophis

9、ticatedthey look up, analyze, and compare product features, prices, payment options, shipping information, and return policies before making an online pur- chase (Burke 2002; Song et al. 2012). Beyond computers, with increased web access and simple user applications, con- sumers can now not only acc

10、ess product information on their mobile phones but also make purchases. Retailers are quickly recognising the need to offer persuasive online propositions in order to attract potential consumers (Caruana and Ewing 2010). To this end, online retailer websites have evolved into information storehouses

11、 containing product information, im- ages, videos, recommendations, and consumer reviews. Many retailers have tried to use online social networks in an attempt to form some form of relationship with their consumers (McWilliam 2012). Parallel to its commercial success, e-commerce or online retailing

12、has garnered interest from marketing researchers (Grewal and Levy 2009), who have tried to apply the models of traditional retail and have developed new ones when others have proved invalid in the online domain. This research in online retailing is, however, fairly recent; a meta-analysis (19972003)

13、 conducted by Straub et al. (2005) shows only 49 academic articles related to the B2C markets. In another study, Hwang et al. (2007) found only 61 articles related to e- commerce in their list of top five marketing journals (Journal of Consumer Research, Journal of Marketing, Journal of Retail- ing,

14、 Journal of Marketing Research, and Marketing Science) between 1996 and 2005a decade. Hence, we can conclude J. of the Acad. Mark. Sci. (2016) 44:206217 207 that research in e-commerce has a broad scope but is still fragmented. In the domain of e-commerce technology, re- searchers have focused on a

15、multitude of aspects, such as mar- keting strategy, technology adoption, online store formats, buy- er behavior, mobile commerce, and CRM. However, despite these developments, research indicates that online retailers find it more difficult to build a relationship with consumers as com- pared to bric

16、k and mortar retailers (Chen et al. 2008). We believe that concepts and practices of relationship mar- keting (RM) may be useful in establishing strong relationships with online retail customers (Bendapudi and Berry 1997). The relationship marketing concept has been useful in developing new definiti

17、ons of concepts such as trust, commitment, close- ness, and relationship quality (Morgan and Hunt 1994; Gronroos 2009; Hennig-Thurau et al. 2002). Past research has also identified various instruments that aid in relationship marketing efforts such as direct marketing, database market- ing, customer

18、 partnering, CRM, services marketing, and more to achieve the objectives of customer satisfaction, loyalty, and customer retention. Various industry applications of relation- ship marketing have been identified, such as implementation programmes and new industry practices (Das 2009). Although a plet

19、hora of literature exists in relationship mar- keting in both B2B and B2C contexts, literature on the appli- cation of relationship marketing to the online retail context has emerged only in the last decade and still remains fragmented in terms of the variables being investigated. A comprehensive mo

20、del is needed to identify the various aspects of relationship marketing that have been investigated and those that need further scrutiny. This paper takes a meta-analytic approach to discover what works and what does not in establishing relationships with online customers. We employ the meta- analyt

21、ic model proposed by Palmatier et al. (2006) as the basis for understanding how the relationship marketing concept may be applicable in online retailing. We chose Palmatier et al. (2006) model as it is predicated on a comprehensive meta-analysis of extant literature in the domain of relationship mar

22、keting and identifies the key concepts that impact relation- ship marketing. In the study by Palmatier et al. (2006), 97 published and unpublished manuscripts on RM were analysed, and 637 correlations were identified to calculate the pairwise estimates. In our study we adapt this model to online ret

23、ailing and identify the key concepts being researched in the domain. Our study sets out to (a) conduct a thorough review of the empirical studies pertaining to relationship mar- keting in the domain of online retailing; (b) identify what strategies help build customer relationships in the online do-

24、 main; (c) identify the consequences of relationship marketing in online retailing; and (d) identify the potential gaps that would motivate hypotheses for further research. A model of relationship marketing in online retailing Among the various constructs in relationship marketing, Palmatier et al.

25、(2006) propose the Relational Mediator Meta Analytic Framework, in which they identify 18 constructs. They classify nine antecedents as customer-focused, seller- focused, and dyadic antecedents. Commitment, trust, relation- ship satisfaction and relationship quality are identified as customer-focuse

26、d relational mediators. Outcomes are classi- fied as customer-focused, seller-focused, and dyadic out- comes. They also identify four moderators, service versus product based exchanges, channel versus direct exchanges, business versus consumer markets, and individual versus organisational relationsh

27、ips. Based on this theoretical model, Palmatier et al. conduct a meta-analysis to finally develop the causal model. We used this meta-analytic framework as the basis of our research. Based on our meta-analysis of empirical research in online marketing, we identify the following rela- tional model fo

28、r the online retailing context (Fig. 1). Fig. 1 Relationship marketing in online retailing Customer Focused Antecedents Relationship Benefits Dependence on Seller Mediators Consequences Seller Focused Antecedents Relationship Investment Seller Expertise Commitment Trust Relationship Satisfaction Rel

29、ationship Quality Expectation of Continuity Word of Mouth Customer Loyalty Dyadic Antecedents Communication Similarity 208 J. of the Acad. Mark. Sci. (2016) 44:206217 We identify 13 constructs in our RM framework for online retailing. Six antecedentsrelationship benefits, dependence on seller, relat

30、ionship investment, seller expertise, communi- cation, and similarityare further classified into customer- focused (relationship benefits, dependence on seller), seller- focused (relationship investment, seller expertise), and dyadic (communication, similarity) antecedents. Further, as in the origin

31、al model, in the context of online retailing also we iden- tify commitment, trust, relationship satisfaction, and relation- ship quality as customer- focused relational mediators. Commitment is defined as Ban enduring desire to maintain a valued relationship (Deshpande et al. 1993). Trust is defined

32、 as Bconfidence in an exchange partners reliability and integ- rity (Morgan and Hunt 1994). Trust and commitment are the most frequently studied constructs, though they have been studied as antecedents, mediators, and even consequences. Trust has often been identified as an antecedent to commit- men

33、t. Relationship quality is the Boverall assessment of the strength of a relationship (Crosby et al. 1990). Relationship satisfaction is defined as the satisfaction of the consumer from the overall relationship (Crosby et al. 1990). Finally, we iden- tify three consequences of RM as expectation of co

34、ntinuity, word of mouth, and customer loyalty. The definitions of the various constructs are presented in Table 1. Method We conducted a literature search in various scientific data- bases in order to identify studies pertaining to relationship marketing in online retailing. Ebsco, Elsevier Science

35、Direct, Proquest, and Google Scholar search engines were used to search abstracts and keywords. We searched for each construct present in the model offered by Palmatier et al. (2006) along with one of the following search terms: online retail(ing), in- ternet retail(ing), electronic commerce, and e-

36、commerce. The initial search generated more than 100 empirical studies that were examined for the constructs. To be included in the anal- ysis, each study needed to meet the following criteria: the study was conducted in the online retailing setting, the study reported the sample size, and the study

37、 reported the Pearson correlation coefficient or a test-statistic that can be converted to correlation. Based on these criteria, a total of 50 empirical studies were identified from the last decade (20002013) that provided a total of 153 causal relationships (Table 2). Empirical studies in online re

38、tailing have used multiple constructs and variables with similar definitions. To organize them as per Palmatier et al. (2006) framework, we coded, using standard procedures, the various antecedents, mediators, and consequences according to the definitions offered by Palmatier et al. (2006). Statisti

39、cs were coded based on the results reported in each study and included sample size, means and standard deviations, correlation values, F-tests, and t-tests. Out of the 153 causal relationships identified originally, we used 131 relationships in the model. The final list of studies used in the empiri

40、cal analysis is available on request. The first step in the analysis involved converting the effect size values to correlations (r). Correlation was taken as the primary metric as it is a scale-free measure and is easy to interpret. In order to include as many effect sizes as possible, we included s

41、tudies using regression and structural equation modelling (SEM) (Peterson and Brown 2005). To convert coefficients of regression (beta) we used the formula r = 0.98+0.05, where equals to 1 when the coefficient is non-negative and 0 when it is negative (Peterson and Brown 2005). For SEM r-values, we

42、took direct effect r-values as is. Where there were indirect effects present, we incorporated the same and calculated the total effect size. The effect sizes across studies were integrated by applying the Schmidt and Hunters Bare Bones approach (2004) that accounts for sampling error to the r-values

43、. While Hunter and Schmidt suggest that the Bare Bones approach may be defi- cient, other researchers have demonstrated that applying cor- rections for all artefacts can be inaccurate especially when the number of studies is small (Spector and Levine 1987). Since we identified very few studies (12 i

44、n some cases) for some relationships in our model, a more conservative approach was necessary (Allen et al. 2004). While r-values were corrected for sampling error, they were not corrected for measurement error. Durvasula et al. (2012) recommend that researchers should report disattenuated effect si

45、zes. This is because disattenuating effect sizes increases the effect sizes and, assuming reliabilities are about equal, all effect sizes will increase by about the same amount. This we think will result in increasing the cut-off values for categoriz- ing the effects into small, medium, and large bu

46、t not change the category to which the effect size is assigned. No correc- tions were made for other artefacts; these corrections are gen- erally required when researchers wish to aggregate the studies and analyse the multivariate causal model. Online retailing as an area of study is a relatively yo

47、ung field in the domain of marketing. The objective of this study was to examine whether the constructs and relationships pertaining to relationship marketing are of relevance in online retailing, and to develop a framework indicating the key rela- tionships being examined by researchers, which woul

48、d moti- vate hypotheses to be examined by future researchers (Janiszewski et al. 2003). At this stage of research in relation- ship marketing in online retailing, a quantitative summary of the existing body of research can be a relevant contribution to literature. Schmidt et al. (1985) state that: E

49、ven with small numbers of studies and small Ns, meta- analysis is still the optimal method for integrating find- ings across studies. In the absence of such interim meta- analyses, psychologists would likely base judgments on J. of the Acad.Mark. Sci.(2016)44:206217209Table 1 Definitions of constructs included in framework Construct Definition Commonly used terms Reference Antecedents Relationship benefits The various functional or social benefits received from the exchange partner Convenience motivation, information qua