E-commerce websites and online customer reviews in France: analysis of current strategies and suggestions for improvement

(1)

UPPSALA UNIVERSITY Department of Business Studies Master Thesis Spring Semester 2012

E-commerce websites and online customer reviews in France: analysis of current strategies and

suggestions for improvement

Authors: CAO Yuan DEBEUF Benjamin Supervisor: OLSSON Ulf Date of submission: 05/25/2012 Revised version submission: 08/15/2012

(2)

ii

Abstract

E-commerce distribution channel experiences a dramatic development nowadays. France witnessed a rapid growth rate of online sales in recent years and now ranks as the second largest market in Europe in terms of turnover. On E-commerce websites, customer review system is considered as an efficient tool of E-Word of Mouth, enabling users to write recommendations which will influence potential purchasers. This paper discusses the crucial factors of customer review system. Also, a tool to evaluate review system is elaborated with five criteria such as accessibility, quality, design, interaction and control. Ten case studies of French E-commerce firms are presented according to those criteria in qualitative study.

Opinions from French customers are collected through online questionnaires in quantitative study. Researches made from supply (firms) and demand (customers) sides show that accessibility and quality are the main concerns for users and often weaknesses in current review systems. Also, the credibility of reviews is questioned by customers. Focusing on these aspects, the paper aims to give suggestions for designing an ideal customer review system to firms in French e-commerce industry.

Keywords:

E-commerce websites, customer review system, E-Word of Mouth, User Generated content, online shopping

Acknowledgement

We would like to thank the teachers of Uppsala Economic Department for their helpful advice, particularly our supervisor Ulf Olsson. Our families and friends also supported and helped us during this study. Last but not least, we appreciated that 407 respondents answered our survey and provided valuable information for the study.

(3)

iii

1. Introduction and research question

1.1 Background and Research question

Most of traditional retailing shops experienced a downturn in the period of global financial crisis, unlike online shops. France typically illustrates this phenomenon. It witnessed a rapid growth of 24% in 2010 and exceeded Germany to be the second in Europe just after UK in terms of total online sales revenue (FEVAD¹ report, 2011). However, if this distribution channel offers new opportunities for firms, it also strengthens consumers’ power. With development of internet, firms begin to market through social media tools and offer more and more possibilities for interaction to their customers. Consequently, users openly criticize brands and products on corporate websites, blogs, forums or social media. Consumers’ voice, satisfied or not, is louder.

Directly presented on product pages, customer reviews represent one of the aspect of this consumers’ voice. Previous researches have shown that consumers’ reviews and voice became an important tool for firms to communicate and to provide transparency. Literature shows the usefulness of consumers reviews (Mudambi & Schuff, 2010) and the impact on sales that reviews can create (Zhu & Zhang, 2010), (Öğüt & Onur, 2012). Half of users who visited online-shop malls consider consumer reviews important in their buying decisions (Park, Lee & Han , 2007).

Research question:

Customer review system remains a delicate tool to manage. As customer review system reflects users’ opinion, both positive and negative information about products and companies are presented. Objectives can be conflicting for firms: putting forward positive reviews to enhance sales while allowing critics, ensuring control of reviews while remaining transparent.

Further, firms are often victims of fake reviews from competitors who aim to damage their reputation. In order to reduce these threats, companies must optimize their review system.

Consequently, the study was developed around the following research question, applied to French market:

 How can e-commerce firms improve their customer review system to meet customers’

expectations?

1 French Federation of E-commerce and Distance Selling

(6)

2 Several stages have to be considered before achieving this goal. Firstly, it is essential to study the current situation of customer review system in France, analyzing companies’ strategies and policies. Secondly, French customers’ perception towards customer review systems of E- commerce websites should be investigated. The last stage consists in giving suggestions after comparing firms’ and customers’ side.

1.2 Definition of terms

It is necessary to give some definitions about certain terms used in this study.

Customer review: The definition of customer review is “a report about a product written by a customer on a commercial website to help people decide if they want to buy it” (Cambridge Business English Dictionary). A slight difference exists between customer and consumer reviews because customer is the buyer when consumer represents the user. In this thesis, the topic focus on customer review in a general scope of view from E-commerce websites. The term “review system” refers to customer review system generated by E-commerce websites.

UGC (User-generated content): Defining UGC is a difficult task because no widely accepted official definition exists. Generally, UGC is understood very broadly to include all content put online by users, whether it was created by them or not (Valcke & Lenaerts, 2010). Often UGC is monitored by companies or website administrator in order to prevent from offensive content or insulting language as well as protecting copyright issues.

EWOM: The most useful definition of electronic word-of-mouth (eWOM) is "Any positive or negative statement made by potential, actual, or former customers about a product or company, which is made available to a multitude of people and institutions via the Internet."

(Hennig-Thurau et al. 2004)

PEOU: Perceived ease of use (PEOU) is defined as “the degree to which a person believes that using a particular system would be free of effort” (Awad & Ragowsky, 2008). The higher PEOU degree one review system has, the easier it is perceived to be used by customers.

1.3 Objectives

The study focused on a certain kind of tool: reviews generated by users on e-commerce websites. Consequently, reviews posted on social media, review sharing website or platforms

(7)

3 did not concern the study. In addition, websites which are specialized in gathering third party reviews such as E-opinion or hostel.com are not considered in this thesis. This thesis focused on review interfaces on online shop, directly managed by companies.

The aim of the study is to give directions to firms in terms of consumer review implementation in e-commerce industry in France. Intermediate objectives can be described as the following:

First a model aiming to better understand consumer review system mechanisms was developed based on literature. It includes visions from both company’s perspective and customer’s perspective. Secondly, 10 successful websites examples in terms of visiting rates were studied in order to better understand current e-commerce firms’ strategies. Crucial factors to evaluate customer review systems were proposed, combining with factors elaborated in previous theoretical background. In order to gain insights from customers, a quantitative study of online questionnaires was performed. It permitted to draw a picture of the use and influence of reviews by French customers. The study also tested the value granted by users to review systems, analyzing customers’ mindset and pointing out their expectations.

Finally suggestions for reviews improvements were given based on qualitative and quantitative studies.

2. Literature review

In order to link the study of customer reviews with previous researches, a literature review will be presented in this chapter. Many books, journals and articles are accessible to understand the relationship between customer reviews and online marketing. Since E- commerce websites were developed just in latest few decades and customer review system is still a novelty, many articles associated with this topic were published recently. However, previous literature about third-party consumer information system is a good introduction to the importance of customer reviews.

Early in 1978, Jolson and Bushman stated that the mass media third-party source speaks forcefully among all the information systems available to guide customer choices. When consumer initiative, personal recommendations and marketer-dominated source are not appropriate, people tend to search for “expertise” from third-party such as food critics on local newspapers. Product rating and recommendation by critics proved to be sufficient in guiding consumer choices.

(8)

4 From 90s until new century, internet technology has been developed dramatically. Firms begin to offer more and more possibilities for interaction to their customers, enabling them to post reviews and comments publicly on internet. Noise generated by customers, more networked, has more visibility (Cronin-Lukas, 2006). They control more and more the environment online and give feedbacks.

2.1 Benefits of online review system

Recommendations and customer reviews have a positive impact on customer perception of the usefulness and social presence of the website. To certain degree, such perceptions enhance customer loyalty, which is value for practitioners to design B2C website (Kumar & Benbasat, 2006). Kumar states that sometimes negative reviews even create more credibility than only positive reviews. Customers tend to trust balanced opinion of reviews.

Customer rating websites are new tools of web-marketing as well. According to a CompUSA- iPerception study, statistics show that 81% of surveyed shoppers consider customer ratings and reviews important when they are researching or planning a purchase. Further, 63% of consumers indicated they are more likely to purchase from a site if it has product ratings and reviews (iPerceptions, 2006).

As customer reviews play a more and more important role in E-WOM (Electronic Word of Mouth), it is beneficial for companies to have massive amount of comments from customers on companies’ website. However it is doubtful that a majority of customers is willing to share their experience and participate in writing reviews.

2.2 Low participation of customers

Researches made by Baird and Parasnis (2011) show that only a small percentage of customers (5%) are “Engaged authors”. These users are very active on the web, responding to comments and creating posts. People sharing occasionally their thoughts, called “Casual participants”, represent the majority of users (75%). The responding rate is quite low from online customers.

To increase customer participation, it is essential to analyze customers’ motivations for leaving reviews. Literature identified motivations for people to participate in E-word of mouth

(9)

5 as following (Hennig-Thurau et al, 2004): “economic rewards, self-enhancement, social benefits received, post-purchase advice seeking, desire to help the company, hope that the platform operator will serve as a moderator, expression of positive emotions, and venting of negative feelings, exertion of power over companies, convenience in seeking redress, concern for other consumers”. There are four main categories of contributors: “self-interested helpers”, driven mainly by economic incentives, are the largest segment, representing 34% of helpers. The second group is the “multiple-motive consumers”, which represents 21% and is motivated by a large number of factors. The third segment “true altruist” aims to help both consumers and companies and represents 27% of respondents. The last group, (17%) is mainly motivated by the concern for other consumers and can be called “consumer advocates”.

The challenge for companies consists in finding efficient methods encouraging customers to comment on their products.

2.3 Do age and gender make a difference?

Influence of customer content is still in an "early adopter" phase. The most influential consumers are 24 to 44 year, heavy internet users, and the most influenced are teenagers and young adults, up to 24, (Riegner, 2007). Previous study shows young people are easier to be influenced by online reviews, which is a good hint for firms to target marketing group. But the study does not show if those heavy users or those young reviews readers are also writers of reviews. It would be worthwhile to know people from which age group are motivated to write reviews.

Awad and Ragowsky (2008) made a study about online shopping behaviors according to genders. Their results show that the effect of trust on intention to shop online is stronger for women than for men. That is to say, women tend to have more e-loyalty than men, once they trust the brand. Referring to contribution of online reviews, women tend to write personal issues based on cooperation and network-oriented collaboration. Women write reviews for the purpose of participating and helping other users in the community. Men tend to post longer reviews and more informative message than women, for self-promotion and social standing purposes. In addition, it is expected that an increase in PEOU will have a stronger effect on online trust for women than for men.

(10)

6

2.4 Balance of positive and negative reviews

Consumers do not consider positive and negative reviews equally (Zhang, Craciun & Shin, 2010). For example, when consumers are exposed to negative information about a product, they can categorize the product as low in quality. Negative information may therefore be considered more useful for decision making purposes and is consequently given greater weight than positive information. According to prospect theory, one experience of loss appears to be greater than the gains (Kahneman & Tversky, 1979). In this sense, a negative WOM message could have a stronger influence on a customer’s brand evaluations and purchase (Fukuyama, 1995), (Wright, 1974).

That is not to say, firms should censor and delete negative comments. According to Kumar, sometimes negative reviews can create more credibility than only positive reviews. Customers tend to trust balanced opinion of reviews. Companies should also present the most persuasive reviews first. If customers’ interest is raised by other forms of communication, they may be less influenced by negative WOM (Khare, Labrecque& Asare, 2011). To reduce the amount of negative WOM, companies should design the review system in a better way such as keyword searching or review recommendations. Also, marketers should offer incentives to motivate satisfied customers to share their positive experience. When comments are positive, marketers should highlight the popularity of the information (number of customers) and the benefits of conformity (follow the main stream).

Intelligent marketers will design a good review system to gather positive reviews, meanwhile, prevent negative reviews, in order to reach a balance of opinions. Then it is worthwhile to study what kind of review system is attractive for customers and what factors of website will influence customers’ purchasing choices.

2.5 Key factors to better present review system

Online consumer reviews have an effect on consumer purchasing intention. Consumer-created information is more credible than seller-created information from the perspective of trustfulness. There are some key factors to present review system in a better way in order to attract customers and increase online sales. Quantity and Quality of reviews are important factors which will influence consumer purchasing choice (Park, Lee & Han, 2007).

(11)

7 Quantity: The number of on-line reviews of a product may be taken as representing the product’s popularity. Customers tend to relate numbers of reviews to numbers of consumers who have bought the product (Chen & Xie, 2008). But quantity of reviews is only one element of showing interaction between E-firms and customers. There are other modern tools such as possibility to share on social media and ask friends’ opinion.etc.

Quality: The content of on-line reviews varies from short to long and from subjective to objective (Chatterjee, 2001). A high-quality review is more logical and persuasive with evaluation based on the facts about a product (Park, Lee & Han, 2007). Different variables can influence the perceived helpfulness of customers’ reviews (Mudambi & Schuff, 2010):

“review extremity, review depth, and product type”. Moderate reviews or lengthier reviews will be more adapted to influence consumers, depending on the kind of product (Park, Lee &

Han, 2007). Quality of online consumer review content can be evaluated in terms of relevance, reliability, understandability and sufficiency.

The credibility of information is influenced by 3 factors: argument strength, source credibility, and confirmation with receiver’s prior belief (Cheung et al, 2009). People tend more to accept information that is similar to their own opinion. Recommendation consistency and aggregate ratings seem to be determinant for influence of reviews. If the reader is less knowledgeable on the topic, reviews have more influence on him.

Design of review systems contains many details such as attractive box and positions on the page.etc. “No more than three colors should be used for the body of text and no more than two should be used for the heading text” (Ivory, 2003). Colors, fonts and graphics of review systems are part of web-design and they should be closely related to company webpage.

Because slight differences of colors and fonts may affect customers’ perception of companies.

“Unstructured, meaningless content on a page results in users experiencing difficulty finding information and navigating around the web site” (Badre, 2002). Methods are existing to improve the presentation of customer reviews (Liu, Karahanna & Watson, 2011), with different goals: to help consumers in making accurate choices, to lower the cognitive costs of making these choice, to help consumers have an unbiased understanding of the product, and to construct a set of evaluative criteria. Retailers could improve the presentation by four different ways: categorization (by attributes), overview (tag system), linkage (to related reviews) and filtering (sort reviews by star ratings or helpfulness)

(12)

8 A complicated web-design may lead customers to lose their interests and quit review system.

It is essential to design a review system which can catch customers’ attention and provide suitable useful products information. Previous literatures only concentrate on structure of review systems instead of its contents. Good design of review system should contain keywords searching, filter of reviews.etc. Those functions are also related to PEOU (Perceived Ease of Use) to customers, which should be considered as Accessibility of a review system.

2.6 Model of customer perception to purchasing choice

Zhu and Zhang (2010) develop a model which illustrates the process from customer perception to purchasing choice in online shopping. In the first stage, customers acquire products information through online reviews. This builds a basic cognition of product quality, cost and condition for customers. Thus, various consumers have different internet experience which leads to different preference as well. In addition, other factors, such as competition, business models (e.g., business-to-consumer, consumer-to-consumer), or even the online review system’s design (e.g., how ratings are displayed, how easy it is to rate an item), may affect consumers’ reliance on reviews. In the second stage, customers build judgment or reliance based on the characteristics from first stage. Only when consumer’s reliance is sufficiently high enough, they will enter the third stage of purchasing. In the process of choice making, many factors such as presentation of website or usefulness of customer review could easily influence final purchase choice.

Figure 1, Conceptual Framework, Zhu & Zhang (2010)

(13)

9 This model provides a general outline for how different elements on websites will influence customer purchasing choice. However, the model does not specify how review system will affect customer choice. In addition, this model is based on customer’s side and does not take into account companies. For example, it does not mention what policies firms use to control information from online reviews and how those policies will affect on customer’s perception.

The present study will also take into consideration companies’ side.

2.7 Conclusion

According to literature review, we can conclude that the main challenges for E-commerce firms are low participation of contributors, and balance of positive/negative reviews.

Participation is influenced by age group. Women and men have different motives for writing reviews. In order to encourage people to contribute more to review system, firms should develop a review system which considers these factors. Literature shows quality and quantity of comments, design of review system are important factors link to helpfulness of reviews.

Based on these findings, key factors of review system were developed, gathering five main aspects: Accessibility, Design, Quality, Interaction and Control. Accessibility stands for PEOU (Perceived Ease of Use) including providing user-friendly tools both for reading and writing reviews. Design contains visual attractiveness of reviews such as organization of text and keyword presentation. Interaction factor includes possibilities for customers to communicate with friends, review writers or even firms. Control stands for the ability of managing reviews by firms such as having a customer account and review verification. Firms should improve their review system according to these five main factors.

The contribution of this thesis is that it develops a new model based on the one way customer online purchasing choice model from Zhu and Zhang (2010). The new model contains both views from E-commerce firms and customers. In addition, the figures mentioned in literature reviews are general statistics. Their applicability to French market remains a question. By quantitative study from French customers, we can draw a better conclusion from the perception of French E-shopping users, which is more credible to French E-commerce firms as well.

(14)

10

3. Model

According to the former chapter, the following model has been elaborated.

E-COMMERCE USERS Review Readers/Contributors

CUSTOMER REVIEW SYSTEM

E-COMMERCE FIRMS

•Provide information to consumers

•Transparency and trust gain

Contributors

•Economic rewards

•Self

enhancement

•Suggestions to others

•Complaints

Value (utility for customers) Evaluated by 5 Factors

Accessibility/Quality/Design/Interaction/Control Readers

•Gain of information before purchase

Influence on purchase

decision and will to contribute

EFFECTS

MOTIVES MOTIVES

DEMAND SIDE SUPPLY SIDE

Feedback on products or services

Chart 1. Thesis Methodology Flow Chart

The research structure will follow this model: the left side (Chart 1) represents E-commerce users from Demand Side. This thesis gathers ideas and opinions from E-commerce users such as customer review readers and contributors. The empirical finding results to understand customer behaviors such as their motivations to write and read reviews before purchasing. On the other hand, it is supposed to understand reasons of customers’ participation and driving forces to encourage customers to share their experience on e-commerce review websites.

Although literature review tells us that motivation for readers and contributors are gaining information, economic rewards. Etc, the situation might be different in the case of French market and the results might various from woman to man, young adults to old people. To have an idea of E-commerce users’ preference is one of the purposes of this thesis.

The right side (Chart 1) representes firms of online shopping websites. Obviously, the motivation of managing an online review system for an E-commerce firm is that product review is an efficient tool of consumer marketing. On one hand, review system can provide basic information of products such as pictures, function and price to end-users which is an

(15)

11 efficient marketing tool at the same time. On the other hand, good review system reveals management ability of a company and its transparency policy to the public.

The bottom part (Chart 1) of this methodology represents the third entity of the tools which firms and consumers are interacting with. From cases studies of 10 most visited online shopping websites in France, we conclude their management of customer review system into five main domains: Accessibility, Quality, Design, Interaction, Control.

Construction of the model

The previous model was elaborated on the basis of theories found in the literature review.

Users of customer review system can be separated in two different parts: readers and contributors. Readers’ motives of use come from a research of information which influences their purchase decision (Zhu&Zhang, 2010 and Iperception). Contributors will use review system for different purposes such as social status, help to the community or rewards (Hennig-Thurau et al, 2004). When it comes to the review system itself, 5 factors can characterize its utility or value for users. Awad & Ragowsky stated that PEOU was an important phenomenon that could enhance the dynamism of review systems. This can be related with the first factor: accessibility. Several authors insisted on the fact that quality should remain a priority for review systems (Chatterjee, 2001), (Park, Lee & Han, 2007), (Mudambi & Schuff, 2010), (Cheung et al, 2009) and should be considered as the second factor. Thirdly, the appearance of the system is preponderant as well, as stated (Ivory, 2003)(Badre, 2002)(Liu, Karahanna & Watson, 2011). Design should consequently be the third factor.

To these 3 factors we can add Interaction and Control. First, social networks websites became popular a few years ago. As a consequence, within an internet platform, review system should provide sharing possibilities and integrate social media. Consumptions, review shared not only on the website but also through social media to maintain E-WOM.

On the other side, firms are providing review system in order to give more transparency, pre- buying information and consequently reasons for buying. According to Peppers and Rogers (2012), honesty can become a competitive advantage. Allowing consumers to review your products is one of the main steps to demonstrate “trustability”. In this angle, firms gain transparency, and this factor can become a key selling point for customers.

(16)

12 How to evaluate a review system?

The aim of the thesis is to propose axis of improvements for review systems towards French customers. Therefore the research performed in the qualitative and quantitative studies initially permitted to evaluate the current influence of review system on consumers. Most importantly the value given to current review systems was evaluated.

4. Research Design, methodology and argumentation

The primary purposes of the study are to analyze E-commerce firms’ policies towards consumer reviews available on the web and to give suggestions for improvement. The attractiveness of this industry for the study can be explained by its relative novelty and the important interaction, due to the use of internet, which can be either an advantage or an inconvenience. A current picture of strategies was elaborated and discussed, insights from customers were gained through questionnaires and improvements were proposed to conclude the study. The methodology used in order to respond to the research question will be presented in following part. The study provides analysis and comparison of different policies towards consumer reviews in the industry of E-business.

Why France?

The study focused on French market for different reasons. First, France is the second biggest market of e-commerce in Europe in terms of turnover and market growth. According to FEVAD report in 2011, the number of online shopping websites in France raised from 64100 to 81900 with an increasing rate of 27.77% in year 2010. That is to say, this dramatic rising number of e-commerce websites is to meet potential demand from e-commerce users in France. Secondly, choosing the country of one of the authors’ nationality was a necessity in order to gather precise and relevant information, through the questionnaire and online data research. The choice of France was also motivated by the fact that one of the authors has some professional experience within French e-commerce websites, consequently knows the industry and has contacts within these firms to better understand reviews’ issues. The question of consumer reviews also recently emerged in France due to the occurrence of numerous scandals concerning false reviews. Consequently, certification standards will appear at the end of the year, in order to better identify review authors and to limit censorship on firms’ side.

Improving these review systems could hopefully permit to sustain the growth of e-commerce in this country, enhancing credibility of e-shops and enabling to attract more and more customers.

(17)

13 For the purpose of having a complete image of the problem, the research strategy was based on a multiple-method: mixed model-method. A Case study about E-commerce firms and their current use of reviews and quantitative research on customers were completed. Using both qualitative and quantitative approaches offered complementarity in the investigation, providing a better understanding and permitting generalization.

4.1 Qualitative data

4.1.1 Sample

The collection of qualitative data was performed with the case study method. The most adapted method to analyze and evaluate policies of French E-commerce websites in terms of customer reviews.

To reach that goal and gain a broad image of the industry, ten companies of different activities were selected for the study. The selection criterion was the following: all companies belong to the top 15 of French e-commerce websites (in terms of average number of visitors per month, FEVAD report, 2011) and provide a customer review system for goods (which excludes Ebay.fr, Vente Privée and Darty for instance). An exception has been made for Yves Rocher, which is not in Top 15, but stands for the leader in the cosmetics industry in e-commerce in France. In this way, the cases gained diversity, presenting different industries.

Figure 1: Ranking of Websites for the last quarter 2011, in terms of unique visitors (FEVAD report, 2011)

(18)

14 Selected companies overview

Amazon.fr (B2C/C2C) Historically selling cultural products, this company became one of the world leaders of e-commerce, diversifying its offer while expanding internationally

Fnac.com (B2C/C2C) Originally a brick and mortar store chain, specialized in cultural goods, Fnac activities has today 3 main poles (cultural goods, electronics and small appliances, travel and entertainment)

PriceMinister.com (B2C/C2C)

On the model of Amazon, online store with a broad offer of products (from books to fashion retailing)

Cdiscount.com (B2C) Online store with a wide range of activities (cultural goods, electronics, home appliances, food or travel). The main characteristic of this website relies in its positioning, very aggressive on promotions

La Redoute.fr (B2C) Specialized in clothing industry on the internet, leader in France Pixmania.com (B2C) Stands out by its wide presence in Europe and wide range of offer

on the model of Cdiscount.

3 Suisses.fr (B2C) Main competitor of La Redoute on the online clothing market Decathlon.fr (B2C) Specialized in distributing sporting goods (main competitors

Intersport) with a worldwide presence

Leroy Merlin.fr (B2C) Specialized in construction, DIY and gardening goods

Yves Rocher.fr (B2C) Leader in cosmetics on e-commerce channel, their brand identity revolves in offering natural and organic products

4.1.2 Method

To understand firms’ current strategies, case study is the most suitable research method.

Choosing ten French leading B2C websites gives a comprehensive view of customer review system in France from the supply side. A selection of different industries collected various and complete information of E-commerce websites. Comparison between ten websites revealed strength and weakness of different review systems.

How to evaluate online reviews system?

The selected companies review systems were evaluated with the following criteria:

accessibility, quality, design, interaction and control.

(19)

15 The different criteria were detailed with 5 items in each category. These items were elaborated on the basis of the main factors, they are the practical illustration of each factor.

For instance, the number of clicks to access a review naturally illustrates the aspect of accessibility.

In order to evaluate each review system, the following coding method was used: one point was counted depending on the presence of each sub-criterion on the website. The addition of points gives a mark for each criterion, with a maximum of 5.

Data was analyzed through spider web graphs, characterizing the review system of each firm.

Analytic tool: Radar Chart

The use of spider-web graphs was motivated by the fact that it allows the representation of 5 quantitative variables such as the 5 factors chosen (accessibility, quality, design interaction and control). This kind of graph is adapted and clear for data presentation. What’s more, it allows characteristics to be compared for each firm (the scale remains unchanged and possibility to superimpose each graph in order to compare).

Radar charts are a useful way to display multivarite observations with an arbitrary number of variables (Chambers et al, 1983). We use it in our study because of we do not know the difference of importance between five driving forces (Accessibility, Quality, Interaction, Design, Control) before empirical finding. We can not know which factor is more important than others so we suppose they are equally important variable. This might be a limitation of analysis because of our restricted marketing academic knowledge and limitation of researching time. And the fact that we choose an informative data analysis method might be also one of the limitations of this thesis.

The radar chart is composed of several dimensions including lines. A line is drawn connecting the data values for each spoke. So the plot of each chart looks like star. The star represents a single observation. In our qualitative study, the performance of customer review system in each E-commence firm is considered as a star plot. We scored them according to the criteria we made in 4.1 Qualitative Data Method/ 4.1.2 Method.

Typically, radar charts are generated in a multi-plot format with many stars on each page and each star representing one observation (NIST/SEMATECH, 2003). It is somewhat easier to

(20)

16 see patterns in the data if the observations are arranged in some non-arbitrary order (if the variables are assigned to the rays of the star in some meaningful order) (Friendly. M, 1991).

According to our empirical researching such as observing each firm’s website and registering as a E-commerce user for each website, we give points to each firm as in Table 1(Results of Quantitative Study) as being a client of those ten companies. We use our personal energy and effort to experience each website’s customer review system. We sent some fake complaining information to test the ‘Control’ factor including information transparency of those companies.

The result is showed in the Chapter 5 including the big table and some other graphs. Every red word in Table 1 means one negative point which represents 0 point for this sub-criterion of certain company. Every black word in Table 1 means one positive point which represents 1 point for this sub-criterion of certain company. For example, ‘Necessity of having a customer account’ for Amazon is ‘Yes’ (In Red), which means that’s inconvenient for customers to leave their comments. So we give Amazon 0 point for this sub-criterion under ‘Accessibility’

factor. If we look through ‘Accessibility’ of Amazon, it has three black words and two red words which mean Amazon has 3 points as showed in the Chart under 5.2.2 Retailers with a cultural goods offer background (Accessibility 3 in the Radar Chart of Amazon). It is also showed in Summary of marks Amazon/Accessibility (Table 2, under 5.1).

Accessibility

This first factor is essential for review systems. The objective consists in facilitating users’

participation, providing adapted and user-friendly tools, both for reading and writing reviews.

In order to evaluate accessibility, five items have been studied:

-number of clicks to access a review and number of clicks to write a review (prior to the access of product page)

The less access is time consuming, the more users will see reviews. In e-commerce industry, reducing the number of clicks to purchase facilitates sales. Applying this principle to reviews should motivate users to write reviews.

-help to prepare content

Some websites offer fixed-frame structure to create a review. For instance, different questions or topics will help contributors in different manners: help them to provide a structured review and assist them in being more specific on one aspect of the product (design, technical characteristics…).

(21)

17 -necessity of having a customer account

Having a customer account on a website is necessary to write a review for security reasons.

However, creating an account can be considered time consuming or complicated, and simply discourages users to write reviews.

-filter of reviews (by age, gender…)

Certain review system can provide a filter in order to make the comments posted on the website more readable. For instance, a user interested in the opinions of people with the same profile will be able to rank and select reviews or reviewers by criteria such as gender, age, date or mark.

Quality

The second priority for review systems relies in quality. In order to ensure the credibility of information and enhance review quality, firms should provide several tools developed hereafter. Doing so, trust of users will be improved.

-review evaluation

Marking reviews can be a quality indicator. Usually, reviews can be evaluated by readers, according to their usefulness. The score published can be the following “30 users out of 50 find this review useful”.

-reviewer evaluation

Reviewers can be marked as author, which can reveal the quality of reviews.

-identification of reviewer

Seeing reviewers’ profile also helps the user to judge quality of review. On the reviewer profile page, much information is available such as age, date of registration, or number of comments posted. For instance, in the cosmetic industry, revealing the color of hair/eyes on this kind of profile will help user to judge the relevance of the product and review for them.

-different criteria of notation

The mark given in the review can be separated in different items that bring more precision to the reader. For a garment, the mark can be composed of three sub-marks: quality, size and comfort.

-organization of review, text (by characteristics/topic)

In terms of content, the organization of review is a factor of quality. Publishing the review with different sub-categories by topic for instance brings more clarity to the final user.

(22)

18 Design

An attractive and adapted design enhances review system efficiency, catching attention of users and motivating them to contribute. Generally, a good design should bring more harmony, visibility and dynamism to the review system.

-attractive presentation of reviews

Placed in a page full of information, the review must stand out to be more visible. Usually firms use different colors, separate in different boxes to improve the clarity and let users take reviews into consideration.

-represented notation of product

Marks given by reviewers are possibly the most banal and the most important item of a review system. It is essential for design to represent it in a clear and appealing manner.

-position on the page, visual

According to the user reading habits, reviews are more visible if placed on a particular part of the page. Again, more visibility should bring more dynamism and increase the effect of review system, catching more users.

-direct visual on the product page

Reviews should appear directly on the same page, under the product information to be directly visible. One more click or tab decreases the number of users reading the reviews.

-keywords presentation

An interesting and efficient way of improving reviews’ design consists in providing keywords at the top of the review. That way, the product and its main characteristics can stand out, as well as the review itself.

Interaction

Another important aspect relies in interaction, between firms and users as well as between customers themselves. By interaction is meant all the exchanges possible from an actor to another, such as information or even rewards. The objective is to communicate around products and review system, to enhance the dynamic of sharing knowledge and to create a relationship even for brand image.

-possibility to share on social media

Offering the possibility to link reviews to social media permit to create informal internet communication, from friends to friends. As seen in the literature, friend as advisor can be more influential on customers than classic communication.

(23)

19 -opinion about the comment

This refers to the possibility of giving an opinion about the usefulness of a review. Unlike in the quality part, it does not concern the score, but the presence of the following question “Was this comment useful or not?”

-incentives to write a review

In order to motivate people to write reviews, incentives from firms can be offered such as discount or gift. This way of interacting is used to enhance the dynamism of review system.

-access to the reviewer

To have a deeper conversation and feedback on products purchased, contacting directly the reviewer could help users in their shopping process. This goes beyond a simple review and establishes a relationship among users’ community.

-possibility to ask friends’ opinion

A direct link is given for users to send the product page to their friends, and ask for a comment. In that sense, friends undertake the role of advisors and reviewers in private.

Control

The last essential aspect for firms to provide a review system of quality is control. As content generated by users belongs to their websites and product pages, this kind of information becomes real selling points. It is also essential to control information, in terms of vulgarity and slurs in link with the quality. Consequently, firms have to set up several safeguards in their review system.

-having a customer account

If users must log-in to post a review, the risk of letting outsiders commenting is reduced. It enables company to better control the resource of reviews and keep customer relationship.

-possessing the product

Naturally, having tested the product you are commenting on is essential to ensure the viability of the review. Sometimes, the reviews written by non-users of products are low quality because they do not have an experience of using products. Firms should check if users commenting have the product in their order history.

-feedback/alert on the content

In order to limit vulgarity and slurs, the possibility is offered to denounce a comment to the firm.

(24)

20 -word limit

Controlling the number of words is a kind of limitation, to provide clarity and to maintain a framed format of reviews. On the other hand, word limit should not be too short because it is important to have complete customer’s opinions and encourage their co-creation.

-review verification

Last safeguard, the firm reserves the right to keep or not a review, according a delay of generally 24h before publishing on the website. Companies who have this function of review system have a high control of all the reviews. Questions of censorship and ethics are raised because firms could limit the number of negative reviews which could tarnish the image of products and decrease sales. This sub-criterion is still considered as a beneficial policy in terms of review management.

4.1.3 Limitations

The main limitation of the study relies in the fact that the number of selected companies is small comparing with 81 900 active e-commerce websites in France (FEVAD report, 2011).

However the study has focused on the leading companies of the sector, in different industries, which can give a good picture of the market, as e-commerce users are at least customer of one of the companies selected.

Also, the cases offered an external view of companies’ policies. There is a lack of internal study inside companies. On the other hand, feasibility of proposals needs to be further tested in reality of companies as well.

The five factors are also not equally important. It is assumed that they are equal in the framework, but it might be different proportion in their importance to measure a review system. For example, the accessibility part could be less important than design for customers.

Finally, the main category that could have been included is quantity of reviews. Chen & Xie (2008) and Park, Lee & Han (2007) stated that quantity remain a quality indicators for users.

The more comments are numerous, the more representative and valuable is the information.

However, it is difficult to observe and compare the number of comments from one website to another. The product offer is different, as well as the presentation, number of sales, Etc.

Different factors influence the success of a product and number of comments. More than a precise and fair evaluation of quantity, one can just notice a global trend concerning the number of comments on each website.

(25)

21

4.2 Quantitative Data

Quantitative interviews will provide information concerning users’ mindset and expectations towards reviews. In a second time, the possibility is given to propose some tracks of improvements for the company’s strategy. Combining case study and quantitative research permits to respond to the second step of the study (propose some improvements for firms).

Questions focus on customers’ opinions concerning products reviews. What do they think of current company’s strategy towards consumers’ reviews? Does the website show transparency of comments and reviews? Is the layout of product review attractive, practical or confusing?

How could it be improved? How to motivate people to share reviews? How much does products reviews influence customer choice?

Data collected through questionnaire and analyze should permit to generalize conclusions to E-commerce industry in France. It is aimed to understand the current situation and respond to the research question.

4.2.1 Sample

The questionnaire was handled through internet, to facilitate distribution and being practical in financial terms. The objective is to reach a representative sample to be able to generalize findings in the French market.

The quantitative research was based on around 400 respondents: according to the size of the market (27.7 million of e-commerce users in France in 2011, FEVAD report). With a margin of error of 5%, the adapted sample size for a ten million population or more is 384, according to Saunders, Lewis and Thornhill (2009).

The following conditions shall be respected: different ages, gender and if possible social background, they must be E-commerce Users, who have already experienced online reviews reading or writing.

4.2.2 Method

The objectives of this questionnaire are multiple. It first completes the case study, offering insights from customers. It also permits to test the value given by users with the 5 factors (accessibility, quality, design, interaction and control). Trends and expectations shall be pointed out as well as directions to improve current strategies in terms of customer reviews.

(26)

22 All the questions in the questionnaire are designed and written by authors of this thesis (Appendix I). The questions were created on the basis of the theory (5 factors) in order to match the qualitative analysis. In that way, it would be more practical to compare the current situation of review system and the expectations of users on the same subject. To justify the rationality of questions, they are designed in the purpose of understanding consumer behavior and preference.

Data was collected through social media (Facebook, LinkedIn), personal email lists, alumni of school and review forums. This method had the advantage of offering the possibility of targeting interrogated people and also to reach an adequate number of respondents quite rapidly (three weeks in our study). All the data was gathered by Google docs and analyzed through Excel by the authors of this thesis.

The questionnaire is designed on three different parts. Q1 to Q3 (Gender, Age and Occupation) are general questions of customer segmenting. It is for the purpose of analyzing empirical finding results. Q4 to Q5 (Frequency of purchasing online, Do you read reviews before shopping? Do you contribute to write reviews?) are subdivided questions to distinguish heavy users and common users of customer review system. Q7 (Trust of reviews) and Q8 (Companies are neutral in publication?) is to test French consumers’ general opinion towards E-commerce websites’ transparency policy. We want to know if French customers believe what is showed in firms’ review system. We wonder if they notice that those reviews might be propaganda from firms which are probably sent by employees of firms. The answers are divided from 1(Strongly disagree) to 7(Strongly agree) in order to segment different opinions.

Q9 (Necessity of review system for a company) and Q10 (Influence on purchasing choice) is to ask respondents’ opinion towards importance of review system. Q11 (Most important factor among 5 driving forces of review system) is to ask respondents to select one most important factor among Accessibility, Quality, Design, Interaction and Control. Due to limitation of one choice, the result tends to be ‘Accessibility’ and ‘Quality’ which is possibly to be a limitation of this thesis. Q 12 to Q 17 focus on detail question towards 5 main driving forces in order to test importance of each of them. Q18 and Q 19 are in aim of testing motivations for customers to participate or contribute to write reviews. The answers are created according to previous literature review such as economic rewards and sharing experience.

(27)

23 For the analysis, age groups are separated by decades 15-24, 25-34, 35-44, 45-54, 55-64 and +64, which is a compromise between the five-year definition of INSEE² and a larger acceptation, fitting more in terms of customer profile.

The socioprofessional categories respect the definition of INSEE. In order to gain precision, the groups of jobseekers, students and interns are added, replacing the category “other persons without occupation”.

In order to present collected date, different kinds of charts were used. Pie-charts graphs are adapted to illustrate quantitative data, representing all its diversity and proportions for each group. As for them, bar charts have the advantage to facilitate the comparison between different groups of data. It enables also to recognize the norm and the weight of each side, useful for data extraction, comparison and analysis.

4.2.3 Limitations

Conclusions cannot be expanded worldwide because the population selection is only composed by French E-commerce users. It might be different in other countries with various customer preference and habits. Also, biases may exist in the sample, due to shortcomings concerning representativeness of certain groups. Unfortunately, the study could not ensure to avoid this problem, mainly because of the non-existence of a budget for the research.

4.3 Time Horizon

This thesis is a cross-sectional study, evaluating customer reviews phenomenon in a short period of time. Firstly, there is a time limitation for tracing changes of opinions between marketers and end-users in long term. Time schedule of this research project is 5 months.

Secondly, this thesis focuses on current customer reviews management policies instead of observing customer review development in a long term. What can be ensured is that this thesis gave customer review management suggestions to companies according to end-users’

feeling and expectations.

2 French National Institute for Statistics and Economic Studies

E-commerce websites and online customer reviews in France: analysis of current strategies and suggestions for improvement