How does e-wom affect online purchase intention?
Assessing gender differences in customers’
perception of online customer reviews for different
product types
Amsterdam, July 2, 2014
Thesis seminar Business studies
Supervisor: Dr. H.H. (Hsin-Hsuan) Lee
Academic year: 2014, Semester 2, Block 3
Student name: Xiaoyu Yang
2
Abstract
E-wom consumer review is playing an important role influencing customer’s choices and purchase intention. Existing research indicates that the received usefulness of online customer reviews is different for males and females, while consumer perception of online reviews will be affected by review valence (negative or positive review). Volume of dedicate study of the effect of e-wom consumer reviews on purchase intention is at scarce; while the limited amount of research thus far has focused largely on the implications of view valence and gender difference, little is done on that of product type.
The primary objective of this research study is to investigate the impact of e-wom product reviews on online shopping purchase intention. Meanwhile, the considerations of review valence, gender gap and product type are embedded with the investigation to identify the influence such elements have within the overall interaction between e-wom and purchase intention. In particular, the study aims to confirm the moderating role played by gender gap within the mechanism. An online experimental survey has been conducted and the data collected has been processed via an ANCOVA analysis to text five hypotheses constructed based on current literature.
Conclusions have hence been drawn to suggest online consumers tend to be drawn in by discouraging e-wom comments; hedonic products are subject to a higher level of impacts from e-wom consumer reviews when compared to utilitarian products; female consumers are less likely to purchase online; the impact difference of negative and positive e-wom product reviews on the consumers’ purchase intention is greater for females than for males; and that the impact difference of hedonic and utilitarian online reviews on the customers’ purchase intention is also greater for females than for males.
Such research results facilitated recommendations for online business management which are also presented in this report.
3
Content Table
List of Figures and Tables ... 4
Foreword ... 5
1.Introduction ... 6
2.Theoretical Framework and Research Hypotheses... 8
2.1 E-wom and Review Valence ... 8
2.2 E-wom and Product Type ... 10
2.3 E-wom and Gender Gap on Online Shopping ... 12
2.4 Interaction between Review Valence and Gender ... 14
2.5 Interaction between Product Type and Gender ... 15
3.Methodology ... 17
3.1 Research Design and Subjects ... 17
3.2 Variables and Measurements ... 19
3.3 Survey Construction and Experimental Procedures ... 21
4.Research Results ... 23
4.1 Descriptive Statistics ... 23
4.2 Manipulation Check ... 23
4.3 Hypotheses Test Results ... 25
5.Discussion of Research Results ... 31
5.1 Negativity Effect and Online Retail ... 31
5.2 Product Type and E-wom ... 32
5.3 Online Shopping by Female... 34
5.4 The Moderating Role of Gender ... 35
6.Limitations and Future Research ... 37
7. Conclusion...39
4
List of Figures and Tables
Figure 1: Hypothesis Model
Figure 2: Interaction between Gender and Review Valence
Figure 3: Interaction between Gender and Product Type
Figure 4: 3 Way Interaction between Gender, Product Type and Review Valence
Table 1: Survey Sample Composition
Table 2: Age Distribution of Survey Sample Population
Table 3: ANCOVA Analysis Results
5
Foreword
This thesis was written for my Bachelor degree in Business Studies at the University of Amsterdam. While writing this thesis I got help from many people that I would like to thank here. First of all, I would like to thank my supervisor, Dr. H.H. Lee, for her kindness to be my supervisor and providing me the suggestions from her experience and professional knowledge. Secondly, I would like to thank all my friends and people that answered my survey and helped me spread the questionnaires. Finally, I would like to thank my family for their support.
I hope you will enjoy reading my thesis!
6
1. Introduction
Word-of-mouth (WOM), has been acknowledged to be the most effective and believable form of advertisement for marketers (Henricks, 1998; Misner 1999). It is explained by the power of personal referral on fundamental customer behavior, which described that consumers have the ability to exert powerful influences upon each other (Haywood, 1989). Recently, word-of-mouth marketing has attracted a great number of attentions because traditional forms of advertising appear to be losing effectiveness (Nail, 2005). With the development of electronic technologies, electronic word of mouth (e-wom) appears as a less personal but more omnipresent form of WOM (Sen and Lerman, 2009). To be more specific, electronic word-of-mouth refers to any positive or negative comments of a product or a company made by potential and actual customers via the internet (Hennig-Thurau et al., 2004). From managerial perspective, it is a significant tool helping organizations with brand building, customer acquisition, product development and other managerial activities (Dellarocas, 2003).
Among all types of e-wom, online customer review is playing an important role influencing customer’s choices and purchase intention (Huang and Chen, 2006). Previous research shows that consumer perception of online reviews will be affected by review valence (negative or positive review). Especially, customers have the tendency to think negative review is more credible (Park and Lee, 2009). This could be explained by the phenomenon called negativity effect, which means people would regard negative information more useful and informative than positive information. Besides review valence, the received usefulness of online customer reviews is different for males and females. According to previous research, females tend to rely more on other people’s opinions than males, which indicated that customer’s reviews are more influential for females than males (Garbarino and Strahilevitz, 2004). Receiving recommendations from a friend leads to both greater reduction in perceived risk and a stronger increase in willingness to purchase online among females rather than among males, which can also ease the gender gap on customers’ online shopping behaviors (Rodgers and Harris, 2009).
Thus, for managers who want to take advantages of online customer reviews, they should better understand the effect of gender differences and review valence on customers’ received usefulness of online reviews and purchase intention. In order to figure that out, Bea and lee (2010) conducted a research in which they investigated the gender differences in responding
7 to online customer reviews. This study reviewed that online consumers respond more strongly to negative reviews as compared to their respond to positive reviews; while the female online consumers are under greater influence of e-wom compared to the male ones.
However, during the experiment conducted by Bea and Lee (2010), they only used digital camera as product for testing. Whether the results they obtained are valid across various product types is still unknown. According to Park and Lee (2009), product type would affect the influence of online reviews on customer’s judgment. It is therefore essential for marketers to adapt different strategies for different product types. This form of understanding of the online consumers, in terms of their product-specific reaction to e-wom, will aid business managers in effectively conducting marketing practices for various product types.
Therefore, to fill in the gap, the goal of this study is to investigate the gender differences in perception of online customer reviews of different product types. The primary objective of this research study is to investigate the impact of e-wom product reviews on online shopping purchase intention. Meanwhile, the considerations of review valence, gender gap and product type are embedded with the investigation to identify the influence such elements have within the overall interaction between e-wom and purchase intention. In particular, the study aims to confirm the moderating role played by gender gap within the mechanism.
The study is both academic and practical driven; since it not only fills in the academic gap, but also helping managers making use of online customer reviews. It first of all presents the theoretical framework which reviews relevant literature and outline the research hypotheses. The research methodology is then explained in great details before the questionnaire feedbacks are evaluated, presented and further discussed. A short concluding section at the end will summarize the overall research findings of this study.
8
2. Theoretical Framework and Research Hypotheses
This section presents current research findings collected from existing literature which are related to the research topic of this study. In particular, e-wom and its implications for consumer purchase intentions are discussed in terms of review valence, product type and gender gap. The purpose is to facilitate the formulation of research hypotheses, which will then be statistically tested later, based on the available literature. Essentially, this section integrates the literature review and the conceptual model to present the theoretical framework of the research study.2.1 E-wom and Review Valence
Word-of-mouth (WOM), has been acknowledged to be the most effective and believable form of advertisement for marketers (Henricks, 1998). In its conventional form, WOM is generated and delivered between consumers during face-to-face conversations regarding commercial products and services provision and consumption (Sen and Lerman, 2007). As consumers adapt to the e-commence platform made available by the advancing internet technology, e-wom emerged as a new and increasingly significant variation of WOM (Bickart and Schindler, 2001). E-wom takes a simpler form as opposed to the conventional WOM in the sense that consumers do not interact personally, but primarily exchange accounts of product/service experience and their individual evaluation of the commodities via internet (Hennig-Thurau et al., 2004). Consumer review given in the form of e-wom is commonly publically available, yet only accessed when other online consumers are actively looking for product information on the internet (Sen and Lerman, 2007). Modern business marketers, especially those for e-commence entities, have acknowledged the significance of e-wom and as a result are dedicated to promoting consumers give their reviews on the websites (Yang and Peterson, 2003). E-wom has been given so much attention by the marketers that in the most extreme cases, such as the one of internet retail giant Amazon, traditional marketing methods (such as television advertisements) have been dropped to facilitate promotion of e-wom with the budget saved (Thompson, 2003). The marketers are committed to encouraging consumers to provide product experience reviews by actively contracting existing customers and inviting them to give personal feedbacks on the products they purchase (Sen and Lerman, 2007).
9 E-wom differs from conventional WOM significantly in two aspects. On one hand, e-wom generates a substantially larger amount of consumer reviews compared to that produced by conventional WOM. The sheer volume of e-wom reviews means that when searching for and referring to such messages, consumers are never going through all the available information (Payne et al., 1988). As a result it is important, from the perspective of the marketers, to understand the underlying factors (regarding the nature of e-wom consumer reviews) that influence the consumers (Sen and Lerman, 2007). This then leads to the second difference between e-wom and the conventional WOM: the former comes from people that are not personally connected to the consumers while the latter is often associated with close personal relationships (as people would often obtain opinions about certain products and services from family members and friends) (ibid.). Regarding the effect of conventional WOM, Godes and Mayzlin (2004) concluded that consumers trust reviews from people who are strongly tied to them (such as family members). An e-wom consumer review is typically generated to offer either recommendation or disproval of a product or service. Such review valence has been at the focus of academic studies in terms of its impact on consumers and purchase intention (Sen and Lerman, 2007). Early research findings suggested that negative e-wom consumer reviews often draw more attention from consumers than positive ones do (Herr et al., 1991). In particular, academic research in consumer behaviour has produced evidence to indicate that, in general, consumers tend to be influenced more by negative information than they do by positive information (Sen and Lerman, 2007). Such is referred to as the ‘negativity effect’ and explained with the reasoning that the relative rare occurrences of negatives in a consumer’s social environment make the negatives more attention grabbling (Kanouse and Hanson, 1972). Due to this principle, negative product and service evaluations tend to influence one’s purchase intention more than positive reviews do (Weinberger and Dillon, 1980).
In addition Garbarino and Stahilevitz (2004) observed that when shopping online and hence reviewing e-wom reviews, consumers seek to lower and reduce the level of purchase risks associated with online shopping. In particular, one might be concerned with the privacy and security issues when buying online (Bartel-Sheehan, 1999). In this sense, combined with the literature findings stated earlier, it could well be the case that negative e-wom reviews could raise consumer’s concern for certain issues (such as that regarding privacy and security) and as a result have significant impact on purchase intention.
10 The above presented literature findings together give justifications for the first research hypothesis of this study regarding e-wom review valence:
Hypothesis 1: A negative e-wom review has a stronger effect on purchase intention of consumers than a positive e-wom review.
2.2 E-wom and Product Type
Purchase intention is determined by a number of factors and the most significant one is evaluation of the products and services being sold by the individual consumers (Turel et al., 2008). This process of evaluating the value of commodities takes all form of available product and service information into consideration to reach an individual conclusion. Important information to consider includes product and service features, advertising messages, as well as customer reviews (Senecal and Nantel, 2004). Within the context of online shopping, e-wom offers a comprehensive range of product information, which might not be made directly available to the consumers by the retailers/manufacturers/service providers (Lee et al., 2011). Yet the information provided in e-wom consumer reviews of products is not always utilized on in the same manner: reviews of different product types could have differentiated impacts on consumers’ purchase intentions (Park and Lee, 2009).
In their work investigating the impact of e-wom consumer reviews on purchase intention, Sen and Lerman (2007) made the distinction between two product types: utilitarian and hedonic. On one hand, utilitarian products are consumer durables, such as dishwashers and printers, which are bought by the consumers to serve a particular purpose with their physical features and functionalities. On the other hand, hedonic products, which typically include books, music and holidays, are purchased to facilitate consumption that is ‘primarily characterized by an affective and sensory experience of aesthetic or sensual pleasure, fantasy, and fun’ (Hirschman and Holbrook, 1982). Essentially, the purchase and consumption of hedonic products are driven by individual to satisfy emotional needs (Sen and Lerman, 2007). Adaval (2001) proposed an interesting proposition regarding consumers’ attitudes towards reviews of utilitarian products and hedonic products respectively. As the former is being sought after with a neutral mood for a practical function consumers are likely to pay attention to negative feedbacks on such commodities. Meanwhile as consumers eagerly look to buy hedonic
11 products, driven by a positive emotional state, negative feedbacks could be discounted when accessed (Adaval, 2001). What this particular theory implies is that due to the contrasting natures of hedonic and utilitarian products, consumer reviews given for these two different types of commodities differ in nature too. In particular, feedbacks and evaluation of utilitarian products are generally objective, focusing on the physical features and usability of the items. On the other hand, when reviewing hedonic products, consumers are naturally constructing and presenting personalized opinions from a much subjective perspective, integrating their individual perceptions and values into the comments (Verhagen et al., 2010). This above stated mechanism applies to the consumers when they are looking for the desired products and using other consumers’ reviews as reference points to evaluate quality of available options (Lee et al., 2011). Essentially, when searching for utilitarian products, consumers tend to use an evaluation process with objective standards. Meanwhile when looking to identify the most appropriate choice of hedonic products, consumers instead evaluate ‘through more subjective standards’ (Lee et al., 2011). Given the two different evaluation processes and sets of standards, utilitarian products are relatively easy to be evaluated prior to actual purchase, while hedonic products are harder to be evaluated before personal experiences have taken place. As a result in some cases utilitarian products are also referred to as search goods and hedonic products as experience goods (Nelson, 1970).
Lee et al. (2011) proposed that the effect of e-wom differ among utilitarian products and hedonic ones, due to the contrasting evaluation process. Essentially evaluating utilitarian products using consumer feedbacks is thought to be much more straightforward compared to doing so for hedonic products which requires more thorough research and more product information in general (Duan et al., 2008). Based on this observation, the study of Lee et al. (2011) produce the overall conclusion that e-wom tends to affect purchase intentions of hedonic products than it does that of utilitarian products. This opinion has been echoed by Park and Lee (2009), whose research work on e-wom and product type contained a hypothesis that ‘the impact difference of negative and positive e-wom information on the e-wom effect is greater for experience goods than for search goods.’ What was suggested was that purchase intention for experience goods are more influenced by e-wom consumer reviews than that for search goods are.
12 The above presented literature findings therefore together give justifications for the second research hypothesis of this study regarding e-wom and product type:
Hypothesis 2: A hedonic product review has a stronger effect on purchase intention of consumers than a utilitarian review.
2.3 E-wom and Gender Gap on Online Shopping
As e-wom is generated from online shopping behaviour, it is also important to examine a relevant element of online consumers that is the gender gap. According to Yang and Wu (2006), initially the male consumers dominated online consumer population when internet and online shopping first emerged. The most significant reason contributing to this early trend was that the computer technologies used to power internet have been primarily developed and implemented by males; as a result the males had earlier access to internet and the many innovative social/commercial features (Morahan-Martin, 1998). As revealed by Weiser (2000), in the 1990s, almost 95 per cent of internet users were males when females remained largely detached to the internet at this early stage.
This scenario soon changed over time as internet gradually becomes a common feature in people’s daily lives. The male dominance within the internet user population faded away and the number of regular and active female internet users increased continuously overtime to a matching level to the male one (Yang and Wu, 2006). Nonetheless, the much narrowed gender gap of internet usage is very much limited to access (Yang and Wu, 2006). This is to say that while there are almost equally many males and females using the internet on a daily basis, the two gender groups still demonstrate much contrasting behaviour patterns in terms of the purpose and extend of internet usage (Bae and Lee, 2011). First of all, it has been observed by Jones et al. (2009) that the males tend to use internet a lot more often than the females do. At the same time, the males possess a much stronger and collective confidence in their ability to use internet as well as knowledge of the internet-related cultures and activities (Bae and Lee, 2011).
13 Such gender gap in terms of the usage of internet and its many facilities is believed to be very much evident in the context of online shopping according to Rodgers and Harris (2003). This is to say that the males behave quite differently when they approach and conduct online shopping activities compared to the females (Bae and Lee, 2011). In particular, it has been observed by Slyke et al. (2002) that the males demonstrate a much more positive perception of online shopping experience compared to the females. As a result the males tend to carry out a great deal more shopping using internet than the females do (Rodgers and Harris, 2003). In addition, the males also demonstrate a higher level of trust of online shopping than the females tend to do, which ultimately facilitates the higher volume of online shopping of the males (Slyke et al., 2002).
Bae and Lee (2011) noted that a significant concern for consumers when they consider practicing online shopping is ‘the perceived risk of online shopping’. In particular, according to Garbarino and Strahilevitz (2004), the females normally consider the option of online shopping more risk than the males would perceive the practice to be. As a result, more females decide to carry out their shopping offline and via the traditional channel of visiting physical retail stores. Regarding this issue, Bartel-Sheehan (1999) noted that the most commonly considered risks for the female consumers are privacy and financial security. A common and essential feature of making purchases via internet is that online consumers have to provide a substantial amount of personal information, such as payment card details, home addresses and contact numbers, in order to facilitate the monetary transaction and delivery of purchased goods (Bartel-Sheehan, 1999). The females are normally more cautious and hence reluctant to give away such information as observed by (Paul, 2001). Shimp and Bearden (1982) stated that such perceived high level of risks greatly decrease the level of purchase intention.
On the other hand, according to (Kempf and Palan, 2006), the females often tend to reply more on other people’s feedbacks and recommendations when making purchase decisions. In particular, the perceived risk of online shopping is likely to be greatly reduced for a female consumer if she is given assurance about the particular product and seller by a friend (Garbarino and Strahilevitz, 2004). Therefore in the absence of such reliable information, the female consumers would remain reluctant to buy on the internet; they would prefer socially embedded WOM reviews that are normally delivered offline (Bae and Lee, 2011).
14 The above presented literature findings therefore together give justifications for the third research hypothesis of this study regarding e-wom and gender gap on online shopping:
Hypothesis 3: Female’s purchase intention is less than male’s.
The first three research hypotheses have been proposed based on evaluation of existing literature on the topics of e-wom, review valence, product type and gender gap on online shopping. This research study proposes two further points of investigation, which aim to study the interactions between review valence and gender, and between product type and gender, in terms of the effect of e-wom consumer review on purchase intentions.
2.4 Interaction between Review Valence and Gender
In section 2.1 it has been demonstrated that review valence has a significant presence within how e-wom consumer reviews are perceived to affect purchase intentions. In particular, as Herr et al. (1991) concluded: negative e-wom consumer reviews often draw more attention from consumers than positive ones do. Sen and Lerman (2007) then further stated that consumers tend to be influenced more by negative information than they do by positive information. This negativity effect of review valence implies that negative e-wom reviews could raise consumer’s concern for certain issues and ultimately raise the perceived level of risks associated with online shopping.
On the other hand, recall that in section 2.3 it has been shown that current literature states that the males demonstrate a much more positive perception of online shopping experience compared to the females; and they also have a higher level of trust of online shopping than the females tend to do, which ultimately facilitates the higher volume of online shopping of the males compared to the females (Slyke et al., 2002). Meanwhile according to Garbarino and Strahilevitz (2004), the females normally consider the option of online shopping more risky than the males would perceive the practice to be. As a result, more females decide to carry out their shopping offline and via the traditional channel of visiting physical retail stores. Given such observations, it is reasonable to suggest that the considerably significant negative e-wom consumer reviews could further drive the female consumers away from making online
15 purchases than it would do the male consumers. In addition, in section 2.3 it was also shown that according to (Kempf and Palan, 2006), the females reply more on other people’s feedbacks and recommendations when making purchase decisions. In particular, the perceived risk of online shopping is likely to be greatly reduced for a female consumer if she is given assurance about the particular product and seller by a credible source (Garbarino and Strahilevitz, 2004). Therefore what this implies, in combination with consideration of the negativity effect of e-wom consumer reviews, is that the female consumers could react more strongly to both negative and positive e-wom reviews compared to the male consumers.
All in all, the fourth research hypothesis of this study regarding the interaction between review valence and gender is set to be:
Hypothesis 4: The impact difference of negative and positive online reviews on the consumers’ purchase intention is greater for females than for males.
2.5 Interaction between Product Type and Gender
Recall that in section 2.2, it has been shown that due to the contrasting natures of hedonic and utilitarian products, consumers use e-wom reviews of these two different types of commodities in contrasting manners. Essentially, when searching for utilitarian products, consumers tend to use an evaluation process with objective standards. Meanwhile when looking to identify the most appropriate choice of hedonic products, consumers instead evaluate ‘through more subjective standards’ (Lee et al., 2011). As a result, evaluating utilitarian products using e-wom consumer feedbacks is thought to be much more straightforward compared to doing so for hedonic products which requires more thorough research and more product information in general (Duan et al., 2008). These observations give justification to suggest that e-wom tends to affect purchase intentions of hedonic products than it does that of utilitarian products.
The female consumers’ risk averse nature identified in section 2.3 then suggest that when comparing the impacts e-wom product reviews have on the consumers’ purchase intention regarding utilitarian and hedonic products, the females are likely to demonstrate a greater
16 level of difference (as to the two product types) when compared to the male consumers. This is primarily due to the fact that upon seeing a negative review about a hedonic product, a female consumer is likely to be further deviated from making a purchase decision than a male consumer. The fifth and last research hypothesis of this study is therefore:
Hypothesis 5: The impact difference of hedonic and utilitarian online reviews on the customers’ purchase intention is greater for females than for males.
In summary, the hypothesis model of this study is illustrated by Figure 1 below:
17
3. Methodology
The last chapter presented a literature review that has been embedded within the constructed theoretical framework. Five research hypotheses have been proposed. This chapter will now demonstrate the overall research design of the study. In particular, it will illustrate how the five research hypotheses are tested utilizing on survey data collected.
3.1 Research Design and Subjects
The overall objective of this research study is to investigate if there exists significant differences of the impacts of e-wom product reviews on consumers purchase intentions in terms of gender, product type, and review valence; as well as the interactions between gender and product type/review valence (as illustrated in Figure 1 at the end of the last chapter). Five hypotheses have been constructed based on the literature review. In order to collect quantitative data so as to facilitate statistical testing of the research hypotheses, it has been decided to carry out an experiment online survey.
The online survey method is chosen to collect data due to the fact that this approach allows quick collection of a large amount of data that will be accessible for numerical analysis according to Saunders et al. (2009). Given the theoretical framework outlined in the previous chapter, there are a number of different features to be examined in cross reference therefore an adequate sample size is desired. At the same time, it is also required to collect information with standardized questions so as to ensure the consistency of feedbacks from different individual participants; therefore the survey method is considered appropriate and effective in delivering such consistency (Saunders et al., 2009). To provide further justification for using the online survey method, note that previous academic researches with similar topics, such as the ones of Bae and Lee (2011) and Park and Lee (2009) have used the same data collection method.
Saunders et al. (2009) have pointed out a couple of potential limitations to the data collection method of survey. First of all a questionnaire could be lengthy and thus tedious for the participants to complete. Second of all a questionnaire can only be filled out once by the participants. Consequently, the design of a questionnaire should ideally be efficient and
18 accurate, posing a significant challenge for the researcher. In this particular case, the number of questions contained in the questionnaire is not large due to the nature of the study; while the need to compose clearly structured survey that is accessible for the participants had been kept in mind throughout the process of survey design to optimise the quality of data collection and this research.
In terms of the hypotheses testing, as revealed by the content of the five proposed hypotheses, the process is designed with a 2*2*3 structure. There are the gender comparison, two particular product types identified for investigation (hedonic and utilitarian), and three scenarios regarding review valence (positive, negative and no review). Thus there are in total six conditions for the proposed experiment, with each condition having both male and female contributions. In particular, each of the two product types corresponds to each one of the review valence scenarios to constitute the six conditions; the three conditions regarding utilitarian product have 30 survey participants each while the other three regarding hedonic product have 54 survey participants each; the numbers of male and female survey participants are equal in each case. This is illustrated in Table 1 below.
Utilitarian product Hedonic product
Positive review No review Negative review Positive review No review Negative review Females 15 15 15 27 27 27 Males 15 15 15 27 27 27 Total 30 30 30 54 54 54
Table 1: Research design: 6 conditions, male and female’s composition Total number = 252
As for the selection of sample population to take part in the online experimental survey, it has been decided to use undergraduate students as the research subjects. According to Gallagher et al. (2001), young adults under the age of 30 are found to be the most active and committed internet users who at the same time participate in online shopping keenly. As a result, this particular group of people is ideal for the investigation proposed to study the effect of e-wom on online shopping behaviours. Research findings produced by Gallagher et al. (2001), which identify undergraduate students at universities as the heaviest internet users, provide further
19 justification to use this group of people for the study. In addition, the fact that the researcher herself is a current university student grants her relative easy access to such data generated from fellow university students.
Recall that in section 2.2, utilitarian and hedonic products have been identified as consumer durables, such as dishwashers and printers, which are bought by the consumers to serve a particular purpose with their physical features and functionalities; and commodities purchased to facilitate consumption that is ‘primarily characterized by an affective and sensory experience of aesthetic or sensual pleasure, fantasy, and fun’ (Hirschman and Holbrook, 1982) respectively. Decisions have since been made to use printer (utilitarian product) and DVD (hedonic product) as the subject products for the study. Bae and Lee (2010) stated that home electronics are among the most favourite product categories for online shopping; such makes printer a commonly known concept for the survey participants. Yet such products are often sophisticated and require substantial product research using online consumer reviews when making purchase decisions (Park et al., 2007); therefore the choice of printer as one of the subject products fits the purpose of the study. On the other hand, a DVD is a common and frequently bought hedonic product the purchasing and evaluation process of which is therefore expected to be familiar with the survey participants.
The e-wom consumer reviews used in the survey have been adopted from Amazon.co.uk. The product brand and price information is not included in the questionnaire so as to eliminate the chance that survey participants are affected by such information when giving feedbacks.
3.2 Variables and Measurements
There are three variables implemented in the study, which include gender (independent variable), review valence (independent variable) and purchase intention (dependent variable). 126 female and 126 male survey participants gave their feedback via the questionnaire.
As shown in Table 1 in the previous section, the survey participants have been each assigned to one of the three groups (for both hedonic and utilitarian products). The groups which are offered no product review are the control groups while the groups with either positive or negative product reviews are the experimental groups. For those in the experimental groups, their purchase intentions of the products are measured after they have read the product
20 reviews. As in the case of the control group, the survey participants’ purchase intentions are measured without being exposed to any product review. It is important to make sure that the survey participants in the experimental groups do in fact perceive the positive and negative review valence presented to them: a positive review recommends buying the product while a negative review advise against doing so. Therefore in order to test and hence confirm that the survey participants recognizes the difference in review valence sufficiently to provide reliable response, each of the experimental group participants was asked to evaluate the review valence presented to them. This process of evaluation uses a seven-point scale that indicates level of satisfaction, with 1 being completely dissatisfied (of the product and hence advise against purchase) and 7 being completely satisfied (of the product and hence recommend purchase) (Vagias, 2006). On the other hand, the dependent variable in this study is the purchase intention of online consumers having been exposed to e-wom product reviews. It is measured once again using a seven-point scale to indicate the likelihood of making a purchase, with 7 indicating virtual certainty and 1 indicating exceptionally unlikelihood (Mastrandrea et al., 2010).
There also exist three covariates in the study which are the survey participants’ prior knowledge of the chosen subject products, existing internet use experience, and current level of online shopping experience. These factors could affect the survey participants’ perception of the e-wom product reviews presented to them during the experiment. In turn the measured purchase intention level could also be dependent upon these covariates. This is to say that it could be the case, for instance, a particular university student that takes part in this survey has plenty and substantially more knowledge of printers/DVD and the experience of buying them online compared to the other survey participants, hence hindering the quality of data collected and the subsequent hypotheses testing. Despite the significance of this potential problem, it is considered that on one hand, product information of either printers or DVD’s is widely accessible on the internet; on the other hand, university students are believed to have sufficient and consistent level of experience regarding using internet and shopping online (Bae and Lee, 2010). As a result, it is reasonable to assume that all the survey participants taking part in this experiment are highly likely to have the similar values for the three covariates. Nonetheless, control mechanisms have been adapted for the three covariates during this research study. The purpose of doing so is to make sure that the three covariates do not have significant impacts on the survey participants’ perception of e-wom product reviews. The
21 survey participants were asked to evaluate their level of prior knowledge of the subject products, internet using experience, and online shopping experience respectively. Once again, multiple-point scales are used in this case, with 1 suggesting the lowest level of knowledge/experience and 2/5/7 suggesting the highest level (Vagias, 2006).
Last but not least, there was also the need for this study to control certain product information in order to eliminate bias. In particular, brand names and prices tags could well have considerable independent impacts on the survey participants’ evaluation of the products with the presence of e-wom consumer reviews (Park et al., 2007). Therefore during the proceedings of the survey, product brand and price information had been removed.
3.3 Survey Construction and Experimental Procedures
The survey consists of six versions of questionnaires, each correspond to the six control/experimental groups as identified in Table 1. These versions of questionnaires are therefore largely similar in content, part from the fact that control groups participants are not offered with e-wom product reviews. For the experimental groups, an essential feature of the questionnaires is product information with consumer reviews (either positive or negative). The questionnaires have been constructed so that relevant information about the survey participants is first of all extracted. Such includes gender, age, internet using frequency, and online shopping experience. The last two of the three are two of the three covariates that are being controlled. Next, all basic product information aside from brand and price is then presented to the survey participants with product image adopted from Amazon.co.uk along with a screen shot of the online product reviews. The questionnaires then ask the survey participants to describe their willingness to buy the particular product having read the product information and consumer reviews, using a seven-point scale. The next question is set to make sure that the survey participants in the experimental groups do in fact perceive the positive and negative review valence presented to them. It asks the experimental group participants to evaluate the review valence presented to them using a seven-point to confirm that the review valence in terms of positivity voiced in the product feedback. Lastly, the third covariate, prior knowledge of the subject product is tested in the questionnaire as to conduct the manipulation checks.
22 The survey participants have been approached and contact with via emails. Each of the 252 who took part completed their online experiment individually. They had been assigned to the six groups as identified in Table 1. Having been presented an instruction of the nature, purpose and procedures of the study, the survey participants took their respective versions of the questionnaire. Those in the experimental groups were informed that the e-wom product reviews had been adopted from a real internet retail website and were asked to indicate their level of purchase intention individually rather than giving a definite answer regarding their purchase decision. Those in the control groups were asked to indicate their level of purchase intention without reference to the e-wom product reviews.
Due to time and resource constraints, the online experimental survey had not been pilot tested extensively to facilitate a more thorough design process of the questionnaires. Nonetheless, in order to make sure that the questionnaire questions could be easily understood by the survey participants, the questionnaires have been reviewed by friends prior to the study commenced. The feedback received had been encouraging and approved the clearance of the questions and the ease to complete the survey. Average time taken to complete the survey was under five minutes. During the actual experiment, all survey participants were asked to kindly complete the questionnaire within five minutes. This was to help maintaining an adequate level of concentration from the participants in answering the questions.
23
4. Research Results
Using the research methodology presented in the previous chapter, the proposed online experimental survey had been conducted to collect data. Such has then been process numerically to present numerical evidence which then formulate the discussion chapter. First of all, research results, in particular those yielded from hypotheses testing, are presented in the chapter.
4.1 Descriptive Statistics
In total, 252 people took part in the online experimental survey. Among all participants, 50 per cent (126) were males while the other half was females. At the same time, all survey participants were over the age of 21; and the vast majority of them were found to be younger than 24 years old. In particular, 57 out of the 252 (22.6 per cent) were either 21 or 22 years old; 144 out of the 252 (57.1 per cent) were either 23 or 24 years old; another 24 out of the 252 (9.5 per cent) were either 25 or 26 years old; with the other 27 survey participants aged over 26. This is show below in Table 2.
Table 2: Age Distribution of Survey Sample Population
The age distribution of the survey sample population is one that had been reasonably anticipated. This is due to the fact that data had been collected from current university students, whose age range and distribution are ones that correspond to what Table 2 represents.
4.2 Manipulation Check
As pointed out in the methodology chapter, it is important to check that the survey participants recognize the difference in review valence sufficiently to provide reliable response.
24 Therefore each of the experimental group participants was asked to evaluate the review valence presented to them. This process of evaluation used a seven-point scale that indicates level of satisfaction, with 1 being completely dissatisfied (of the product and hence advise against purchase) and 7 being completely satisfied (of the product and hence recommend purchase). This forms the manipulation check. As reflected by ANCOVA analysis results, which are presented in Table 3 below, a desired and significant recognition of review valence does indeed exist.
ANCOVA results
Source df F-value Sig.
Prior knowledge 1 1.162 .282
Online shopping experience 1 1.220 .271
Internet experience 1 1.372 .243
Review valence (A) 6 3.458 .003**
Product type (B) 1 5.142 .031* Gender (C) 1 16.722 .000*** A * B 6 1.990 .068 A * C 6 2.666 .016* B * C 1 3.961 .044* A * B* C 5 2.427 .036* Error 222 Total 252 Adjusted total 251 *p<.05, **p<.01, ***p<.001.
Table 3: ANCOVA Analysis Results
As shown in Table 3, ANCOVA analysis indicates that survey participants were effectively manipulated to recognize the presented review valence (F=3.458, p=0.003 sig).
25
4.3 Hypotheses Test Results
An ANCOVA analysis has been carried out, for which the results are presented in Table 3, to test the five research hypotheses. There are three covariates existing as pointed out earlier: prior knowledge of the products, internet using experience and online shopping experience. Control mechanisms have been adapted for the three covariates during this research study. The purpose of doing so is to make sure that the three covariates do not have significant impacts on the survey participants’ perception of e-wom product reviews. The survey participants were asked to evaluate their level of prior knowledge of the subject products, internet using experience, and online shopping experience respectively. Multiple-point scales were used in the questionnaires and the results are included in Table 3. According to these results, the prior knowledge covariate does not appear to have significant impact on consumers’ purchase intention (F=1.162, n.s.). The same conclusions are drawn for the internet using experience covariate (F=1.372, n.s.) and the online shopping covariate (F=1.220, n.s.). One possible reason for this trend is that all the participants of the survey have been drawn from the same social context. Consequently, the relevant demographic characteristics of the participants, which have impacts on the survey results, are similar in nature.
The means and standard deviations of the purchase intention, calculated through the ANCOVA analysis utilising on the survey data, are present in Table 4 on the next page. All the control and experimental groups are included in Table 4.
The first research hypothesis proposed that a negative e-wom product review has a stronger effect on purchase intention of consumers than a positive e-wom product review. In order to test this proposition, the ANCOVA analysis sets to help determining if, when comparing to indicated purchase intention (by the survey participants) in the absence of e-wom product reviews, negative e-wom product reviews have a greater impact on the consumers than the negative ones do. This implies that the ANCOVA analysis essentially compares the values of |Negativno review| and |PositivNo review|; the former indicates the effect of negative e-wom product reviews on consumer purchase intention and the latter indicates the effect of positive e-wom product reviews on consumer purchase intention. Results are that
|Negative-26 no review|=1.9881 > |Positive-No review| =1.5952, P=0.013. The first research hypothesis is
therefore supported.
The second research hypothesis proposed that a hedonic e-wom product review has a stronger effect on purchase intention of consumers than a utilitarian e-wom product review. In order to test this proposition, the ANCOVA analysis sets to help determining
Gender Product type Review valence Mean SD
Male Utilitarian Negative 4.00 1.069
No review 3.27 1.223 Positive 4.40 1.242 Total 3.89 1.247 Hedonic Negative 3.59 1.600 No review 5.00 1.359 Positive 4.48 .935 Total 4.36 1.434 Total Negative 3.74 1.432 No review 4.38 1.545 Positive 4.45 1.041 Total 4.19 1.384
Female Utilitarian Negative 2.13 1.506
No review 2.80 1.082 Positive 4.00 1.558 Total 2.98 1.574 Hedonic Negative 2.59 1.803 No review 3.63 1.245 Positive 4.04 1.652 Total 3.42 1.680 Total Negative 2.43 1.699 No review 3.33 1.243
27
Positive 4.02 1.600
Total 3.26 1.650
Total Utilitarian Negative 3.07 1.596
No review 3.03 1.159 Positive 4.20 1.400 Total 3.43 1.484 Hedonic Negative 3.09 1.762 No review 4.31 1.464 Positive 4.26 1.348 Total 3.89 1.627 Total Negative 3.08 1.695 No review 3.86 1.490 Positive 4.24 1.359 Total 3.73 1.589
Table 4: Means and Standard Deviations of Purchase Intention
if, in the two cases of hedonic and utilitarian products and given e-wom product reviews, when comparing to indicated purchase intention (by the survey participants) in the absence of e-wom product reviews, the former reacts more strongly compared to how the latter does. This implies that the ANCOVA analysis essentially compares the values of |Hedonic-no review| and |Utilitarian-No review|; the former indicates the effect of e-wom product reviews on consumer purchase intention for hedonic products and the latter indicates the effect of e-wom product reviews on consumer purchase intention for utilitarian products. Results are that |Hedonic-no review|=1.9167 > |Utilitarian-No review|=1.56667, P=0.034. The second
research hypothesis is supported.
The third research hypothesis proposed that, regarding online shopping, the female consumer’s purchase intention level is less that the male consumers’. In order to test this proposition, the ANCOVA analysis sets to help determining if, in the general cases of shopping online and without considering any impacts of e-wom product reviews, the female consumers are less likely to purchase from online retailers when compared to the female consumers. This implies that the ANCOVA analysis essentially compares the mean values of the purchase
28 intention with no e-wom product reviews involved among the female and male consumers. Results are that, as shown in Table 4, 3.33 (purchase intention for female consumers) < 4.38 (purchase intention for male consumers), P=0.001. The third research hypothesis is supported.
The forth research hypothesis proposed that the impact difference of negative and positive e-wom product reviews on the consumers’ purchase intention is greater for females than for males. This implies that for the proposition to hold, two conditions need to be both met: difference of negative e-wom product review is greater for female than for male, and difference of positive e-wom product review is greater for female than for male. In order to test this proposition, the ANCOVA analysis sets to help determining if, in the two cases of being presented negative and positive e-wom product reviews, the purchase intention for the female and male consumers deviate away from the indicated purchase intention (by the survey participants) in the absence of e-wom product reviews, by different magnitudes. This implies that the ANCOVA analysis essentially compares the values of |(Female, Negative Review)-(Female, No Review)| and |(Male, Negative Review)-(Male, No Review)|; and the values |(Female, Positive Review)-(Female, No Review)| and |(Male, Positive Review)-(Male, No Review)|, corresponding to the two conditions stated earlier. The results are 2.3333>1.6429, P=0.001 (first condition) and 2.0000>1.1905, P=0.000 (second condition). Therefore both conditions are met and the fourth research hypothesis is supported. Figure 2 below illustrates the interaction between gender and review valence in terms of purchase intention.
29 This visual representation of the ANCOVA analysis clearly shows that the female values of negative and positive reviews are further away from the values of no reviews than the male values.
The fifth and last research hypothesis proposed that the impact difference of hedonic and utilitarian online reviews on the customers’ purchase intention is greater for females than for males. This implies that for the proposition to hold, similarly to the case of the fourth research hypothesis, two conditions need to be both met: difference of purchase intention when buying hedonic product with or without e-wom product reviews is greater for female than for male, and difference of purchase intention when buying utilitarian product with or without e-wom product reviews is greater for female than for male. In order to test this proposition, the ANCOVA analysis sets to help determining if, in the two cases of buying hedonic and utilitarian products, the differences of purchase intention with e-wom reviews for the female and male consumers deviate away from the indicated purchase intention (by the survey participants) in the absence of e-wom product reviews, by different magnitudes. This implies that the ANCOVA analysis essentially compares the values of |(Female, Hedonic Products with Review)-(Female, No Review)| and |(Male, Hedonic Products with Review)-(Male, No Review)|; and the values of |(Female, Utilitarian Products with Review)-(Female, No Review)| and |(Male, Utilitarian Products with Review)-(Male, No Review)|, corresponding to the two conditions stated earlier. The results are 1.9333>1.2000, P=0.002 (first condition) and 2.2778>1.5556, P=0.000 (second condition). Therefore both conditions are met and the fifth research hypothesis is supported. Figure 3 below illustrates the interaction between gender and product type in terms of purchase intention.
30
Figure 3: Interaction between Gender and Product Type
Aside from the five research hypotheses, which have all been supported by numerical evidence produced, the ANCOVA analysis also yield statistical evidence regarding the three way interaction among gender, product type and review valence. In particular,
Figure 4: 3 Way Interaction between Gender, Product Type and Review Valence
there is statistics (F=2.427, p=0.036) to indicate significant three way interaction effect, as Table 3 shows. Figure 4 illustrates the structure of this identified three way interaction.
31
5. Discussion of Research Results
The previous chapter presented the ANCOVA analysis results, which provided statistical evidence to support all of the five research hypotheses. Such research results are further discussed in this chapter. The focus is on managerial implications of the research results: the discussion further elaborate on the conclusions drawn from the online experimental survey to facilitate feasible recommendations for online retail business management.
5.1 Negativity Effect and Online Retail
First and foremost, the research results provided further support for current literature to state that the negativity effect does occur with strong significance. Online consumers tend to be drawn in by discouraging e-wom comments made by fellow consumers and hence be expected to decide against making purchase of a particular product (hedonic and utilitarian) associated with negative feedbacks. This implies that, in all likelihood, even with the presence of a considerable and dominant amount of positive e-wom product reviews, a particular commodity will suffer from having only a few unsatisfied customers who decided to voice their discontent publically.
From what this research study has managed to reveal, it remains unclear if online consumers actively look for (the presence of) negative e-wom product reviews when carrying out research and evaluation of potential purchases online. Current literature reviewed and presented in the theoretical framework of this study also lack coverage in this aspect. It can as a result be only claimed that when seeing one negative product review online, consumers are expected to read into it with more attention (than they would with a positive comment). From the perspective of online retailers as well as the product manufactures, such is a concern as it hinders the positive work conducted to generate (overall) customer satisfaction.
Based on the discussion above, it is important for online retailers and the product manufactures to minimise possibility of one customer giving a negative e-wom review. This might seem a trivial point to make, yet the emphasis is on trying to turn such negativity into indifference (as expressed by a neutral e-wom product review). So that online consumers are
32 at least less likely to be driven away from a purchase decision with a lower level of occurrence of negative comments.
A possible solution is to facilitate a communication platform between the customers (who have bought the products and are about to submit online reviews) and the retailer/manufacturer. The purpose is to allowing a last-ditch opportunity for the latter to continence the former otherwise in the event of a negative e-wom review being given. This might be achieved by either offering the unsatisfied customers incentives to change their views about the overall competency of the product (for example the purchase could be ill-informed in terms of feasibility of functionality regarding specific personal needs and circumstances, and an alternative model would excel in the individual cases); or through exceptional customer service. Such a mechanism has in fact been put into operation: the online retail platform EBay allows the buyers to communicate with buyers, to possibility remove negative comments.
It should be noted that what has been proposed above naturally would be more feasible for utilitarian products than it would for hedonic products. This is due to the fact that while the objective assessments made of the former might be altered if customers’ expectations are somehow misinformed (for example by a lack of understanding of the functionality of the products), subjective opinions regarding the latter are likely to remain vastly individualized and independent of the manufactures’ intentions and efforts.
5.2 Product Type and E-wom
The second research hypothesis was also supported by numerical evidence to suggest that hedonic products are subject to a higher level of impacts from e-wom consumer reviews when compared to utilitarian products. This indicates that the online consumers for hedonic products would most likely be more active in looking for and replying on product reviews than their utilitarian product seeking counterparts. The most significant implication hence is that for the retailers and manufacturer (or service provider such in the case of holiday operators), attention needs to be paid to monitoring consumer feedbacks that are publically accessed and to trying to promote positive e-wom reviews while eliminate negative ones.
33 In this sense, two scenarios arise. On one hand, for hedonic products of which the product features and quality could not be altered or/and enhanced once the production process finalises (such as music and DVD), efforts should be made from the manufacture and retailer to clearly state what the product contains and who the target audience are. This is to say that by informing the consumers of the exact features of the products and the most likely user experience, miss-informed purchases and hence unsatisfactory consumer experiences could be eliminated. It does not help, for example, that a DVD has the most basic descriptions of its content to leave the customers guessing. Instead, a more thorough and honest statement might just well improve after-use evaluations even when it at the same time hinders the overall sales volume in the short term (as the accurate and comprehensive product description filters out consumers that would not enjoy the products). On the other hand, there are also hedonic products whose specific features and overall quality could be adjusted or/and improved throughout their product life cycles. A typical example of such a hedonic product is a holiday package: while the geographic and natural characteristics of a travelling destination remain consistent, the service features and quality could be controlled by the operator and its business associates. As a result, negative e-wom consumer reviews could always be avoided through effective management. However the emphasis should really be on how, without operational changes which take time to take effects, negative e-wom consumer reviews could be eliminated. A possible solution is recommended.
Essentially, consumers, when submitting their reviews online, should be promoted to state their individual expectations, preferences and purposes which facilitate their purchases and consumption of the hedonic products, along with their eventual verdicts of the experience. This is to help generate e-wom product reviews which have a significant element of objective justification regarding the naturally subjective assessment given by individual consumers. For example, a holiday maker might come back from a budget trip and simply say that the experience had been one filled with a few yet largely insignificant problems which are in fact expected (by themselves) given the economic natural of the holiday. This type of e-wom reviews lack the fair mention of the product features (which again should always be made clear by the retailer in the first place) while solely concentrates on the subjective accounts given by the consumer. Therefore by promoting entry of purchase reasons in an e-wom consumer review, the seemingly discouraging accounts of the purchases could be evaluated objectively and hence offset.
34
5.3 Online Shopping by Female
The third research hypothesis was supported to add to literature review findings that in general the female consumers are less likely to purchase online. Reasons for this trend have been identified earlier in the theoretical framework to be that, firstly the female consumers show a higher level of concern for privacy and security issues; and secondly they tend to trust WOM reviews from people they know. In accordance, two recommendations are given for the online retail to promote purchase from the female consumers.
First and foremost, while current e-wom consumer reviews tend to have the common feature of concentrating on product features and user experience, there is the need to encourage the insertion of approval of the underlying online shopping environment and facilities. This is to say that efforts should be made, in order to ease the concern for privacy and security issues, to make sure e-wom consumer reviews comments on the convenience and credentials of the online shopping platform. By doing so, the e-wom reviews would also be able to help advertising the shopping channel itself along with the products being sold. However it might be technically difficult to convince the (female) consumers that shopping online is safe and reliable in terms of privacy and financial security, due to the fact that the related threats could not be visibly eliminated. Nonetheless, it would help to make sure that the consumers do not receive marketing messages following their online purchases unless such has been willingly asked of. It would also help to practice efficient and effective online customer services, implement convenient product display to optimise the overall online shopping experience.
At the same time, due to the fact that the female consumers tend to prefer conventional WOM over e-wom that is often from people without social connections to them, it would be beneficial to establish such missing personal ties. A possible way to is this would be for the e-wom provision to integrate with the now vastly popular social media website so that the chance for female consumers to access e-wom reviews produced by people they know could be enhanced. It has to be admitted however, this recommendation might prove to be difficult to realise given certain technical and regulatory constraints. Such is beyond the scope of this research study. Nonetheless, the proposed principle is a justified one, having seen the third hypothesis proven by numerical evidence generated from the online survey.
35
5.4 The Moderating Role of Gender
On the other hand, the forth hypothesis has been proven to indicate that the impact difference of negative and positive e-wom product reviews on the consumers’ purchase intention is greater for females than for males. Meanwhile the last research hypothesis has been proven to state that the impact difference of hedonic and utilitarian online reviews on the customers’ purchase intention is also greater for females than for males. These conclusions confirm the moderating role of gender interlinking review valence/product type and purchase intention. Previous sections in this chapter have already discussed the individual relationships among gender, review valence and product type in relation to online purchase intention, along with the implications of such interactions. Additional comments are provided here to provide further managerial insights into how the research findings could be utilised on by online retail business.
Given the moderating role of gender, and the fact that the female consumers are subject to significantly more impact from e-wom in formulating their purchase intention levels, online retailers and the relevant product manufacturers need to pay particular attention on monitoring e-wom feedbacks published and promoting positive ones if they have the intention or/and obligation to target at the female consumers in particular. Specific measures to achieve such objective have been given earlier in this chapter. Meanwhile, it is also a relevant point to note that for online retailers which do not sell products that are specially intended for the female consumers, it requires further research to determine if it is cost effective to try improving regarding e-wom review provision. This is to say that for such online retailers, whether or not enhanced e-wom consumer review generation does in fact improve business performance, given the resources needed and from a purely financial point of view, depends very much on the degree of effect in relation to relevant financial outlay. It could be the case that investment committed to improve on the e-wom front is not justified in the long term, when, say, the male consumers contribute considerably more to revenue while are not overly influenced by e-wom.
In regard to hedonic products intended for the female consumers, it is important for the business managers to make sure that e-wom product reviews work in their favour. This is
36 simply due to the fact that the impact of e-wom on purchase intention level is at the most significant level among all possible scenarios. Once again there are some possible methods which could be used to help as stated earlier in this chapter. In addition, it is also recommended that consumers are invited to give specific suggestions as to how the product or/and service provision could be effectively improved to meet their expectations when giving e-wom reviews. By doing so, the retailers and product/service providers could rectify the issues identified by negative e-wom reviews and make such changes visible in the constantly updated product information. However it should be noted that doing so would need to be backed up by relevant market and consumer research from an overall and long term financial point of view, so as to make sure that the extra effort put into the business operation is justified.
