Master's Thesis, 2009, 130 Pages
2 Definitions and Explanations
2.1 Web 2.0
2.1.1 What does Web 2.0 Mean
2.1.2 Electronic Word of Mouth
2.1.4 Social Networks
2.2 Classification of Products
2.2.1 High-Involvement Products
2.2.2 Low-Involvement Products
2.2.3 Consumer Goods
2.2.4 Producer Goods
2.3 Summary of the Chapters and Meaning for this Master´s Thesis
3 Theoretical Foundations of Marketing, Marketing Research and Web 2.0 Research
3.1 Marketing and Marketing Research Defined
3.2 Comparison of Web 2.0 Research and Classical Marketing Research
3.2.2 Selection Methods
3.2.4 Secondary Research
3.2.5 Primary Research
3.2.7 Further Marketing Research Possibilities in the Web 2.0 Compared to Classical Marketing Research
3.3 Motivation of Users to Write Word of Mouth in the Web 2.0
3.3.1 Focus-Related Utility
3.3.2 Consumption Utility
3.3.3 Expert Discussions about Specific Products
3.3.4 Approval Utility
3.3.5 Homeostase Utility
3.3.6 Empowered Involvement
3.4 Summary of Chapters
4 Analysis of the Web 2.0
4.1 Analysis of Users in the Web 2.0
4.1.1 Pattern of Use on the Internet
4.1.2 Pattern of Use in Blogs and Social Networks
4.1.3 Opinion Leaders in the Web 2.0
4.1.4 Future Trends
4.2 Technical Requirements for Analyzing the Web 2.0
4.3 Analysis of Blogs and Social Networks in the Web 2.0
4.3.1 Secondary Research in the Web 2.0
126.96.36.199 Quantitative and Qualitative Analysis of Word of Mouth about Servers
188.8.131.52 Quantitative and Qualitative Analysis of Word of Mouth about Men Cosmetics
184.108.40.206 Quantitative and Qualitative Analysis of Word of Mouth about Coffee
220.127.116.11 Quantitative and Qualitative Analysis of Word of Mouth about Tea
18.104.22.168 Quantitative and Qualitative Analysis of Tractors (Caterpillar) with Twitter
22.214.171.124 Quantitative and Qualitative Analysis of Fast Food (Mc Donald´s) with Twitter
126.96.36.199 Peculiarities of Secondary Data in the Web 2.0
4.4 Primary Research in the Web 2.0
4.4.1 Corporate Blogs: Examples of Nike and Microsoft
4.4.2 Online Focus Groups and Discussions with Opinion Leaders: Example of LG
4.4.3 Online Campaigns: Example of Frosta
4.4.4 Peculiarities of Primary Data in the Web 2.0
4.4.5 Further Marketing Consequences of Negative Word of Mouth in the Web 2.0 and How to Handle It: Examples of Kryptonite and the iPod Nano
4.5 Summary of Chapters
5 Implementation for Practice
5.1 Data Privacy Protection
5.2 Ethics and Critical Acclaim
5.3 Make or Buy?
5.4 Recommendations for Web 2.0 Research in Companies
5.5 Summary of the Chapters
6 Resume and Outlook
Index of Tables and Figures
List of Tables
Table 1 Comparison of Classical Marketing Research and Web 2.0 Research
Table 2 Used Web 2.0 Applications 2008
Table 3 Interest to Write Posts and to Publish Them on the Internet 2006 - 2008
Table 4 Awareness and Use of Weblogs 2008
Table 5 Membership in Social Communities/ Networks 2008
List of Figures
Figure 1 Relation of High- and Low-Involvement Products in Combination with Consumer Goods and Producer Goods
Figure 2 Marketing Research as the Basis of Marketing
Figure 3 The Process of Data Collection
Figure 4 The Marketing Research Process
Figure 5 Statistics
Figure 6 The Process of Significance Test
Figure 7 Classical Marketing Research and Web 2.0 Research as Substitutes and Compliments
Figure 8 Internet Users and Total Population
Figure 9 Pyramid of Lurkers, Intermittent Contributors and Heavy Contributors
Figure 10 Topics of Bloggers
Figure 11 The Process of Web 2.0 Analysis
Figure 12 The Value Creation Process
List of Abbreviations
illustration not visible in this excerpt
“When I took office, only high energy physicists had ever heard of what is called the Worldwide Web… Now even my cat has its own page” (Bill Clinton, 1996)
Only 20 years ago, mostly big companies had access to the Internet in order to exchange information for the worldwide production and transfer of financial services. The big change has come with new data based applications and an enlargement of broadband connections in the beginning of the 21th century: It is now possible to provide bigger amounts of data in the Worldwide Web. Nowadays, the Internet is a platform where private users can enter personal contents and information. Blogs and forums have become popular in everyday-life.
In Germany, 38% of the 1006 respondents older than 16 years are registered in online communities. In the strongest group from 16 to 30 years, even three out of four Internet users are members in those social networks, according to the representative study “Internet 2009- Wer macht was im Worldwide Web” by PriceWaterhouseCoopers in Frankfurt (cp. Westfalenblatt, 27 / 28, June 2009, p. 6). Web 2.0 applications have made a change in communication: From the passive one-way communication of the Web 1.0, where users could online read information by experts, to bi-directional communication, where people actively work on the contents of the pages. The Internet has become a social Web: Users with the same interests find each other and discuss everything they think about, for examples about products. Instead of asking friends about their product experiences, they use online communication platforms for the search and the exchange of information. In doing so, a large variety of opinions for every kind of product can be found in the blogs and forums. Analyzing the electronic word of mouth communication is the direct way to customer opinions:
“Imagine, you can overhear like a fly on the wall millions of people talking about your company, your marketing concepts, your advertisement and your products- customers, employees, competitors, partners and the media. Imagine further, you could use these news for adapting perfectly on the customers´ wishes- what you want, how you want it. (..). This and more provide blogs” (Wright, 2006, p. 20).
It is important for every company to find out about the customer´s attitude towards their products. Up to now, companies have conducted classical marketing research surveys, with paper or telephone interviews and focus groups. Classical marketing research works with calculation of probabilities and representative respondents. Web 2.0 secondary research for marketing purposes is a new way of gathering information about products. In addition to that, also primary research in the Web 2.0 in corporate blogs, discussion groups and online campaigns should be possible.
In this master´s thesis, the following questions shall be answered:
Is it possible in Web 2.0 research to interview a representative customer base? Does the Web 2.0 offer other possibilities, for example getting in contact with opinion leaders? Are there differences in customer and producer products and high- and low-involvement products concerning the quantity and quality of online word of mouth? How valuable is the information about products for the company? Which further Web 2.0 opportunities should marketing experts use?
In order to answer these questions, at first the most important terms are defined (chapter 2). Then, a theoretical foundation of marketing, marketing research and Web 2.0 research is given. In order to see the differences between classical marketing research and Web 2.0 research, both ways of researches are compared to each other. An examination of motivation of customers to write word of mouth in the Web 2.0 is another relevant point for evaluating the value of information (chapter 3). As a next step, users of the Internet and active users in the Web 2.0 are analyzed to find out, if a representative Web 2.0 research is possible. After a description of the technical requirements, secondary researches of online word of mouth about servers (HP, IBM, Dell, Asus, and Fujitsu Siemens), men cosmetics, coffee, tea, tractors (Caterpillar) and fast food (Mc Donald´s) follow. Then, possibilities of primary research are shown: Corporate blogs (Nike, Microsoft), online focus groups, discussions with opinion leaders (LG) and an online campaign (Frosta) (chapter 4). Additionally, chapter 5 contains thoughts about implementation for practice and what rules have to be considered. After a critical discussion about Web 2.0 research, recommendations for an efficient analysis are given. Finally, chapter 6 summarizes the gathered information and gives a short outlook.
In order not to go beyond the scope of this master´s thesis, the author only analyses producer and consumer goods. Secondary data from marketing research companies is used. Otherwise, it would not be possible for the author, referring to time and costs, to analyze 400,000 blogs and forums. Nevertheless, two own analyses with the tool `Twitter´ have been undertaken. Because of different offers by marketing research institutes and because of missing information, time and costs are not considered into detail in the comparison between classical marketing research and Web 2.0 research.
In order to understand the context of this work, the following chapters define the most important terms. The master´s thesis is about product related Web 2.0 research. First of all, it is defined what the Web 2.0 is and what electronic word of mouth means. Then, blogs, social networks and Twitter are explained. After that, the chapter shows a classification of products in consumer and producer products in connection with the classification of high- and low-involvement products. In the end, the relevant facts are summarized.
The first part of this chapter defines terms belonging to the Web 2.0.
`2.0´ is a term used for a version of software development. It describes literally the degree of maturity of the new Internet (cp. Knappe & Kracklauer, 2007, p. 17). The Web 2.0 is not a new version of the Worldwide Web which can be downloaded and installed from the Internet. The Web 2.0 is a change of usage on the Internet in the last years. Dale Dougherty, vice president of O´Reilly publishing company, realized that the Internet did not break down with the implosion of the dotcom-bubble in the year 2000. In contrast, it has become even more important. They called the change `Web 2.0´. In 2004, O´Reilly organized a conference at which this change was the main topic. After the conference, the term `Web 2.0´ was spread on the Internet and became the generic term for every renewal in the Worldwide Web (cp. Ebersbach, Glaser & Heigl, 2008, p. 23).
Tim O´Reilly (2006) defined the Web 2.0 as following:
“Web 2.0 is the network as platform, spanning all connected devices; Web 2.0 applications are those that make the most of the intrinsic advantages of that platform: delivering software as a continually-updated service that gets better the more people use it, consuming and remixing data from multiple sources, including individual users, while providing their own data and services in a form that allows remixing by others, creating network effects through an `architecture of participation`, and going beyond the page metaphor of Web 1.0 to deliver rich user experiences.”
To clear the term, Tim O´Reilly, among other aspects, states the following:
- Involvement of collective intelligence of users: Participatory-Internet and user-generated content are part of the new Internet. On different platforms users upload videos, pictures, information and write personal comments. With a low level of previous knowledge, people can use the easy and user-friendly tools of the Internet. For example, almost everybody can write easily posts in blogs and forums.
- Data is in focus: Years ago, data was collected with a high level of effort. Now, information like the collected knowledge in Wikipedia, personal information like addresses and telephone numbers in Xing or product information in other blogs and forums can be gained within seconds.
- Lightweight Programming Models: In order to spread data, lightweight programming models are implemented. That means that data is provided with HTTP or Web-services. This offers an access to global data which is stored on servers of big online companies like eBay.
- Software, which is usable on different equipments: People can not only use PCs, but also mobile phones or PDAs to download information from the Internet, no matter where they are. Users have access to the Internet at any time.
In the meantime, one additional relevant aspect has arisen:
- Social Web: Web based applications are for people who want to exchange information, work cooperatively together, to get in contact with other people and to communicate in a social context. The generated data contains relations between people who use these applications (cp. Ebersbach et al., 2008, pp. 24).
In the Web 2.0, a lot of services are personalized; therefore the actions of users are traceable. There is a transparency concerning actions, data, connections, relations and contents in the social Web. People remain anonymous with nicknames, but nevertheless, every step on the Internet can be sleuthed.
Another important point is that there are cross-links between various posts and articles which strengthen the contents. A collective knowledge is created (cp. Ebersbach et al., 2008, pp. 31). Holdenberg (2009, p.4) illustrated the evolution of the Internet. See appendix p. i, annex 1.
Electronic word of mouth, abbreviated eWOM, is according to Gwinner, Gremler, Hennig-Thurau and Walsh (2004, p. 39) “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”. Internet communications about products are directed towards multiple individuals, are available to other consumers for an indefinite period of time, and are anonymous (cp. ibid.). Word of mouth takes place in many ways: On Web-based opinion platforms, in forums, news groups, blogs, etc. Since it is the “voice of the consumer” (Knappe & Kracklauer, 2007, p. 24) and reflects customers´ experiences, it is extremely credibly. 63% of consumers trust other consumers and their recommendations concerning buying advice. That is why online consumers inform themselves preferentially in forums and blogs to prepare their purchases (cp. Budak, 2009, p. 7). Nine to ten million product related comments from consumers about consumption experiences, product discussions and buying advice are available to other Internet users on Web-based consumer-opinion platforms (cp. Gwinner et al., 2004, p. 40).
Weblogs (short: blogs) are open diaries in the Worldwide Web. The name arises from log (=diary, protocol) and Web. This term was created by the philosopher and programmer Jorn Barger in the year 1997. He said, it is a “Webpage where a Weblogger logs all the other Webpages she finds interesting” (Borger, 1997, in Ebersbach, Glaser & Heigls, 2008, p. 58). The first definition means that a blog is a collection of different links. Nowadays, a blog is more like a diary open for everybody: Users write autobiographical experiences mostly in first-person-perspective and chronological order. In contrast to private diaries, blogs are directed to an audience. The articles are subjective, commentarial and up to date. Bloggers add to their articles videos, pods or pictures. It is also common to link other blogs of other users. Moreover, users can comment the blog-entries of the author which may lead to discussions about the written entries (cp. Ebersbach et al., 2008, pp. 56).
In the meantime, a huge number of different types of blogs have been created in the Worldwide Web: Diary blogs, newspaper blogs, critical watch blogs, war blogs, hobby blogs, photo blogs, video blogs and so on. Another important kind of blog is the corporate blog: Companies use blogs for marketing and public relation purposes. Blogs give the opportunity to get in contact with employees and, even more important, with customers (cp. Ebersbach et al., 2008, pp. 59).
In the focus of public interest are only a few blogs from assorted bloggers. These bloggers are called `A-blogger´ or `A-lister´. They have more than 1,000 visitors per day. The popularity of a blog depends on the personality of the blogger, his intellectual comments, a good way of writing, extraordinary thoughts and critics to current topics (cp. ibid., pp. 68).
Social networks, also called social communities, are built by people with similar hobbies and affections. After having registered, users can create their own profile, naming their interests, profession and hobbies, and showing personal pictures. The creation of a social community is based on the snowball effect: A new user registers, and invites other friends or business partners to the social network. This way thousands of contacts are created within a short period of time (cp. Ebersbach et al., 2008, pp. 79). “Virtual communities are social get-togethers, which appear in the Worldwide Web if enough people have longer lasting public discussions and if they bring in their feelings so that a network of personal relations arises” (Rheingold, 1994, p.16). According to Stanley Milgram in 1967, every person in the world is connected with another person in the world through six acquaintances. This phenomenon is called “Small World Phenomenon“ (Milgram, 1967, in Ebersbach et al., 2008, p. 81). The same applies for the Internet and for social networks where relationships can be shown on the profile site of the users. Users like to stay in contact with people, to re-find old friends and to get to know new friends in social networks.
Twitter is the biggest platform for micro-blogging on the Web. It was introduced in March 2006. In Mai 2009, already 9.4 million people used Twitter. 7,500 new users register every day (cp. Schmid-Wilhelm, 2009, p. 6). Twitter was initially planned for short messages by mobile phone. That is why the messages in Twitter were limited to 140 characters. Meanwhile, the users can write more than 140 characters for example in insisting links with `Twit Longer´ (cp. Schmid-Wilhelm, 2009, p. 10).
Twitter contains also social elements: One can follow the messages, called tweets, of other users and one can be followed by other users. Messages are updated via the Web page, mobile phone, blackberry or various other third party applications. However, once written, the tweets cannot be changed afterwards. Twitter is similar to blogging, except the messages are broadcasted in real time to other readers (cp. Lost art of blogging, 2008, pp.1). The main question originally has been “What are you doing”. According to Schmid-Wilhelm (2009, p. 21), this question should be changed into “What is interesting (or useful) for you at the moment?” By now, creative users write down quotations, insert links of Youtube and communicate with each other. The updates are listed chronologically on the user´s personal profile. Twitter offers different tools. For example, in `TweeSpeed´, tweets per minute are listed: http://www.tweespeed.com/. In `Twitter Advanced Search´, one can look for keywords: http://search.Twitter.com/advanced. `Ask 500 people´ is a short survey in which 500 Twitter users can be asked to a certain topic: http://Twitter.com/ask500people. In `Tweefeel´, the mood of Twitter users about a certain keyword is shown: http://www.Tweetfeel.com
The following chapters show a classification of high- and low-involvement products and consumer and producer goods. It is important to know that consumer goods and producer goods can be either high- or low-involvement products.
Involvement is the “degree of perceived personal importance and personal interest, which arises through a stimulus in a certain situation” (Kuß & Tomczak, 2000, p.66). In other words, involvement is the intensity of reflection, which a consumer spends when buying a product or service. There are four different types in the buying process. Two types are classified in high-involvement and two types are classified in low-involvement.
Concerning the high-involvement buying process, the purchase entails a lot of money, time, effort and consequences for the buyer. That is why people are willing to gather information about high-involvement products. They compare and deliberate the products.
- Extensive buying decision process: The customer makes a big effort in the identification of relevant decision criteria and the comparable alternative products. Decision criteria are personal value, social reputation and monetary risks (cp. Knappe & Kracklauer, 2007, pp. 40). Cars, for example, are expensive and they are seen as a status symbol. Before buying a new car, the customer tries to get as much information as possible (cp. Wikipedia, Low-Involvement-Produkte). Other examples of products are expensive watches, costly machineries and travels. Concerning this master´s thesis, servers and tractors are suitable examples, since the monetary risk is high.
- Limited buying decision process: The decision criteria are set, but there has to be a comparison of the alternatives. Laptops or mobile phones are examples (cp. Knappe & Kracklauer, 2007, pp. 40). Regarding the master´s thesis, (men) cosmetics, coffee machines and successful brand products like Mc Donald´s fast food, Nike sneakers and LG mobiles belong to that category.
These two buying decision processes are extremely important for the customer. He spends relatively much time in gathering information because besides the financial risk exist social and psychological risks. It is connected to the social status and the pursuit of reputation. Information resources are used well-directed and the gathered information is checked carefully (cp. Knappe & Kracklauer, 2007, p. 41). Not only the information research, but also loyalty and commitment are criteria for high-involvement products: “Consumers who are more involved with a particular brand, are more committed and hence, more loyal to that brand” (Quester, Karunaratna & Lim, 2001, p.1). Rosen (2009, p. 142) is of the opinion that the increased risk of losing money and time affects the communication of the customers. In risky situations, for example the purchase of a high-involvement product, people turn to experts for help.
Low-involvement products are often part of everyday life. They are low in price and do not require services (cp. Wikipedia, Low-Involvement-Produkte). Purchases are not important for the customer and thus they are made with little thoughts. Quester, Karunaratna & Lim (2001) affirm in their research that respondents do not appear to attach too much personal interest in a product that is of low cost, ordinary and inconsequential in nature (cp. ibid., p. 4). In the low-involvement category, sources of information are used accidentally, for example information texts on advertising prospects or packaging are read. Examples of low-involvement products are consumer goods, foods and mass products like toothpaste, milk, sugar and bread because there are no big quality differences (cp. Wikipedia, Low-Involvement-Produkte).
The following two types belong to the low-involvement buying decision process:
- Habitual buying decision process: It is characterized through habitual behavior of the consumers. The product categories and brands are already known and the customer acts routinely. The engagement of reflection is low. Examples are products of every-day-life (cp. Knappe & Kracklauer, 2007, p. 41). Regarding this master´s thesis, tea and frozen products represent low-involvement products.
- Affective buying decision process: The customer buys spontaneously and without a previous gathering of information. He/she buys directly at the point of sale because of stimuli of certain offers (cp. Knappe & Kracklauer, 2007, p. 41).
All goods, which are not used for production but are sold to end consumers, are consumer goods (cp. Cantner, Hanusch & Kuhn, 2002, p. 8). Consumer goods are bought by individuals or households for private consumption. Consumption goods are divided into consumer durables (heavy goods such as washing machines, refrigerators, furniture intended to last three or more years) and consumer non-durables (soft goods that are not expected to have a useful life more than three years such as clothing, towels, other textile goods and goods made from other soft material including flexible plastic, fur, leather and vinyl). The transaction occurs between a company and a consumer. This is called `B2C´ (Business-To-Consumer) (cp. BusinessDictionary.com, 2009, `consumer goods´, `consumer durables´, `consumer non-durables´, `B2C´). Thus, consumer goods are also often called B2C-products. In this master´s thesis, men cosmetics, coffee, tea, Mc Donald´s fast food, Nike sneakers, LG mobile phones and Frosta frozen foods represent consumer goods.
Producer goods are also called industrial goods and investment goods. They are used for production and replace or raise the capital stock of the company (cp. Cantner, Hanusch & Kuhn, 2002, p. 8). Producer goods are heavy equipment such as excavators, forklifts, generators, metal-forming or metal-working machines, vehicles, which require a relatively large investment. They are bought to be used over several years. The transaction occurs between two companies. This is called `B2B´(Business-To-Business) (cp. BusinessDictionary.com, 2009, `capital goods`, `B2B´). Thus, producer goods are also called B2B-goods. In this master´s thesis, servers and tractors represent producer goods.
These first introductory chapters outlined important definitions for the master´s thesis. They explained that the Web 2.0 is the `interactive Internet´, in which users state their opinions, give buying recommendations and talk about problems in blogs, forums and in Twitter. The user generated content is called electronic word of mouth.
Moreover, the chapter defined high- and low-involvement products as well as consumer and producer goods. The following figure 1 shows that consumer and producer goods can be either high-involvement or low-involvement products.
illustration not visible in this excerpt
Figure 1: Relation of High- and Low- Involvement Products in Combination with Consumer Goods and Producer Goods (own illustration)
As an assumption, there should be more discussions about high-involvement products in Web 2.0, no matter if consumer or producer goods, than about low-involvement products. This assumption is going to be analyzed in chapter 4. In order to gain a full understanding of marketing, marketing research and Web 2.0 research, the next chapter gives important information.
Marketing is the generic term for marketing research and Web 2.0 research. Firstly, the chapter 3.1 defines marketing and marketing research in order to give an overview about the whole topic. Secondly, the chapter 3.2 compares classical marketing research and Web 2.0 marketing research with each other. It examines the topics information, selection methods, statistics, secondary and primary research as well as products. Moreover, the chapter 3.2.7 points out further marketing and marketing research possibilities in the Web 2.0. It is crucial to see which opportunities a Web 2.0 research offers and where its use is restricted. Afterwards, the chapter 3.3 illustrates different motives of users to discuss product related information on the Internet. It is necessary for companies to understand the reasons of customers in order to grade the value of information. Furthermore, the understanding of motivation enables companies to encourage their customers to more electronic word of mouth.
For a company, it is important to build customer relationships based on customer value and satisfaction. Kotler and Armstrong (2004, p. 4) define marketing as the following:
“Marketing is managing profitable customer relationships. The twofold goal of marketing is to attract new customers by promising superior value and to keep and grow current customers by delivering satisfaction.”
Marketing is critical to the success of every organization no matter, whether it is large or small, for-profit or non-for-profit, domestic or global (cp. Kotler & Armstrong, 2004, p. 4). Marketing has to be understood as “satisfying customer needs” (Kotler & Armstrong, 2004, p. 5). Thus, marketers have to understand consumer demands, to develop and to innovate products with superior value, to set an optimal price, to distribute and to promote them effectively in order to sell these products easily. Summarizing its characteristics, marketing is a “social and managerial process by which individuals and groups obtain what they need and want through creating and exchanging products and value with others” (ibid., p. 5).
Figure 2 shows the construct of marketing with marketing research as the base and the four Ps of marketing mix as the columns.
The term `marketing mix´ became popular after Neil H. Borden published in 1964 the article “The Concept of Marketing Mix” (cp. NetMBA.com., 2002-2007).The original ingredients in his marketing mix included product planning, pricing, branding, distribution channels, personal selling, advertising, promotions, packaging, display, servicing, physical handling, fact finding and analysis. Later, E. Jerome McCarthy grouped the named ingredients into four categories which are today known as the four Ps of marketing: Product, price, place, and promotion (cp. ibid.).The marketing mix is a set of controllable, tactical marketing tools. It consists of everything the firm can do in order to influence the demand for its product.
- Product is the good and/ or service which a company offers to the target market
- Price is the amount of money consumers pay for the offered product
- Place includes company activities that make the product available to the target group
- Promotion means activities that communicate the product and that convince the customers to buy it (cp. Kotler & Armstrong, 2004, pp. 53).
“An effective marketing program blends all of the marketing-mix elements into a coordinated program designed to achieve the company´s marketing objectives by delivering value to consumers. The marketing mix constitutes the company´s tactical tool kit for establishing strong positioning in target markets” (Kotler & Armstrong, 2004, p. 54).
The master´s thesis is about product related marketing research; that means only the one P for `product´ of the marketing mix is relevant in this context. According to Prof. Dr. Iris Ramme and Anke Schramm, the author of this master´s thesis makes no difference between `marketing research´ and `market research´. Both are methods to support solutions to actual and future problems of marketing management (cp. Ramme & Schramm, 2008, p. 12).
illustration not visible in this excerpt
Figure 2: Marketing Research as the Basis of Marketing (own illustration on the basis of E.-C. Greilich, 2009, p.7)
For making decisions, the marketers need information about the object, the environment, the consequences, etc. Marketing research supplies information to realize problems and to solve them. Marketing research can help companies to access market potential, to analyze the market share, to understand customer needs and customer satisfaction, to examine purchase behaviors, and to measure the efficiency of marketing activities like pricing, product changes, distribution channels and promotion (cp. Kotler & Armstrong, 2004, p. 134). Kotler and Armstrong (2004) say that “Marketing research is the systematic design, collection, analysis, and reporting of data relevant to a specific marketing situation facing an organization” (ibid., p. 134).
The fields of marketing research can be grouped into different categories.
Referring to the objects to be analyzed, marketing research is divided into consumer goods marketing research, producer goods marketing research and service marketing research (cp. Weis & Steinmetz, 2005, pp. 17). In this master´s thesis, the focus shall be on goods: Consumer goods and producer goods are examined. Services are not considered in this context.
The method of data collection is classified into secondary and primary research. (cp. Weis & Steinmetz, 2005, pp. 17). The following figure 3 illustrates the differences between secondary and primary marketing research. In secondary marketing research, the researchers use available data. A distinction is made between internal and external resources: Internal resources are from the company itself, external resources are available surveys from other institutes (cp. Weis & Steinmetz, 2005, pp. 35). In primary marketing research, the needed information does not exist and therefore the company has to conduct a new survey.
illustration not visible in this excerpt
Figure 3: The Process of Data Collection (source: Weis & Steinmetz, 2005, p. 53)
Another classification exists about the time period of provided information: In retrospective marketing research, information is retrospectively provided and analyzed. In adspective marketing research, the actual market situation is shown. In prospective marketing research, marketing researchers try to forecast the changes of markets in the future (cp. Weis & Steinmetz, 2005, pp. 17).
Since the deciders and managers need information about products in order to make the optimal decision, marketing research has to provide the necessary facts. Product related classical marketing research analyzes the acceptance of products, the satisfaction of the product name, the image of a product and the product life cycle. It helps to find an adequate product name, the optimal styling, the packaging, repairs, warranties, accessories and services, to support innovations, to find new trends, to analyze products of the competitors, and to forecast developments (cp. Weis & Steinmetz, 2005, p. 20). Therefore, it shall be found out, if also Web 2.0 research is capable to deliver the important information. The following chapters will provide answers.
The following chapters give an overview about classical marketing research compared to Web 2.0 research. Since marketing research is a wide field, only a selection of the most important topics is considered. Thus, topics like omnibus surveys and experiments are excluded.
This chapter compares which information about products marketers can gain through classical marketing research and Web 2.0 research.
Baines and Chansarkar (2002, p.3) formulate the purpose of marketing research as follows:
“The primary purpose of marketing research is to gather information which will allow your company or organization to make better, more informed decisions. Marketing research is closely linked to the marketing concept as it implies a customer focus, that the customer is central to the activities of the company, and the opinions of the customer are a highly valued and useful aid in decision making.”
The classical marketing research process comprises five stages, shown in figure 4:
illustration not visible in this excerpt
Figure 4: The Marketing Research Process (source: Baines & Chansarkar, 2002, p. 16)
The most crucial stage in classical marketing research is to define the problem. An example might be a decreasing number of sales. This should lead to the development of a marketing research question including sub-questions that also need exploring (cp. Baines & Chansarkar, 2002, pp. 16). In classical marketing research as well as in Web 2.0 research, marketers want to find out about the actual image of the product, about the level of satisfaction, about problems, to improve and how to innovate the products. Consequently the first stage applies for classical and for Web 2.0 marketing research.
The next step in classical marketing research is to decide about the research plan. It involves the consideration of the sample, the interviewing methods, the research methods, the time and location and finally the research objectives (cp. Baines & Chansarkar, 2002, pp. 20). The researchers have to structure the problem carefully in order to receive relevant information. The considerations should be done properly in order to avoid the effect of “garbage in – garbage out” (Ramme & Schramm, 2008, p. 8). Information has a higher value, the more detailed it is concerning the formulated problem. In the run-up for the survey, the relevance of information is hard to evaluate since it is difficult to estimate all parameters and coherences in advance. In classical marketing research, marketers can decide how detailed the information should be by adapting the questions of the questionnaire. Of course, information is more valuable the more it is up to date (cp. Pepels, 2007, p. 25). In classical marketing research, it takes time to plan and to conduct the survey, to analyze and interpret it. Thus, it might become a problem that information becomes obsolete. In Web 2.0 secondary research, the marketers have to use the information that is given in blogs and forums. In this research method, they cannot influence directly, what people talk about. In contrast to information in classical marketing research, information found in the Web 2.0 is always updated since the analysis takes only some days. Thus, it is much faster and in consequence less costly than a classical marketing research (cp. Budak, interview in August 2009).
Other qualities that information in classical marketing research should have according to Pepels (2007, pp. 46) are: Reliability, validity, objectivity and significance.
Reliability is defined as “a criterion for evaluation measurement scales; it represents how consistent or stable the ratings generated by a scale are” (Parasuraman, 1991, p. 443, in Baine & Chansarkar, 2002, p. 33). It is not important, if the measured object really is what the researchers actually want to measure. Reliability refers to the extent to which the generated data would be replicated in a repeated study. One example is a test-retest reliability: At different point of times, researchers measure the same sample with the same measuring instruments. Is the test reliable, then the results are highly correlated (cp. Pepels, 2007, pp.46). Compared to information in Web 2.0 research, the gained information is not reliable. Repeating the study two weeks later, the information could be completely different because other users write their product related opinions in blogs and forums. Therefore, the sample does not stay the same.
Validity is defined by Parasuraman (1991, p. 441, in Baine & Chansarkar, 2002, p. 32) as “a criterion for evaluation measurement scales; it represents the extent to which a scale is a true reflection of the underlying variable or construct it is attempting to measure”. It is based on the extent to which the generated data is free from bias. One example is a balance: It is valid for measuring a person´s weight, but it is not valid for measuring the height. In Web 2.0 research, it is more difficult to estimate in advance, whether the information is valid or not. They have to take the information they get from blogs and forums. They can find valid information about for example the actual image of the product, but it is also imaginable that they do not find any results about the product. Particularly about low-involvement products, there is the risk that the results are not valid.
Objectivity means that information is free from subjective influences. In consequence, different individuals independent from each other should come to the same results (cp. Pepels, 2007, pp.46). In classical marketing research, the researchers try to avoid any subjective influences. In contrast, objectivity is not given in Web 2.0 research, since the researchers do not analyze the same blogs and forums and do not use the same software. Thus, different companies probably come to different results.
Significance of information implies that the results are more than just coincidentally. This quality is important to transfer conclusions of a sample to a wider population (cp. Pepels, 2007, pp.46). The significance is statistically tested in classical marketing research. In Web 2.0 research, the gained information represents the opinion of some users on the Internet. The information is not significant because this information of some special consumers cannot be concluded to the population.
Summing up, in classical marketing research, marketers are responsible that these four criteria are fulfilled in order to conclude from the survey to a population. In contrast, in Web 2.0 marketing research the four qualities can hardly be fulfilled.
To get valuable information, another choice facing the classical marketing researchers is between using qualitative and quantitative research methods. A combination of both is also possible.
Quantitative classical research methods contain predetermined, standardized questions from a large number of respondents. In other words, it is a collection of a small amount of information from a large number of people. The answers are quantified in percentages and often statistically analyzed. Quantitative research techniques address the issue of representativeness and generalization by basing the research on large samples of respondents. The researcher establishes the level to which the results will reflect the entire population by choosing the number and type of required respondents. Only superficial information can be gathered because there is no communication between respondents and researchers (cp. Baines & Chansarkar, 2002, pp. 20). The gained information is for example about the market volume, market share, customer structure, dispenses etc. The costs are relatively high because of the scale of data (cp. Weis & Steinmetz, 1991, p. 30).
Qualitative classical research methods are mostly used at the preliminary stages of a project in order to identify the basic factors affecting the defined problem. Generally, a small number of respondents is involved. In contrast to quantitative research methods, the emphasis is on obtaining rich, detailed information from a small group of people. The advantages can be found in uncovering completely new information about the motivations for customer´s behavior, attitudes, opinions and perceptions. A disadvantage is that the results are not generalizable to the wider population (cp. Baines & Chansarkar, 2002, pp. 20). Qualitative information is subjective, not quantifiable and not representative. The communication is interactive in an interview or a group discussion. The costs are lower because of a relatively small number of cases (cp. Weis & Steinmetz, 1991, p. 31).
When analyzing word of mouth in the Web 2.0, marketing researchers get qualitative information but in a high amount. Thus, the analysis of the Web 2.0 is a combination of a quantitative and qualitative research method. Users write down what they think, feel and want from products and companies. Inner feelings, motivation, attitudes, opinions and perceptions become clear. Researchers analyze thousands of posts so that a trend of the Internet users becomes apparent. The comments are not standardized and the users do not represent the whole customer base. Therefore, the results are not generalizable to a wider population. In contrast to classical marketing research, the costs of gaining a high amount of rich, detailed data are relatively low. Moreover, the Web 2.0 offers the possibility to communicate interactively with users in order to gain even more detailed qualitative information.
After having decided on the research method, data is collected. This stage involves the fieldwork and the collection of the required information (cp. Baines & Chansarkar, 2002, p. 31). For example in classical marketing research, researchers work with paper-based questionnaires, have telephone interviews, personal interviews, etc. In Web 2.0 research, data is collected through the use of software which analyzes forums and blogs for certain key words. The data collection is faster and simpler than in classical marketing research. In consequence, it is less cost-intensive.
The next step in the classical marketing research process is analyzing and interpreting the collected data. It comprises data input, analysis and interpretation of the gathered information (cp. Baines & Chansarkar, 2002, p. 32). Software like SPSS helps to analyze statistically the gathered data. Compared to classical marketing research, the findings in Web 2.0 research are in some cases difficult to interpret. Users write in the Web 2.0 comments with irony and colloquial language. Software is not able to understand it correctly; thus, an employee has to re-read the sentences. In classical marketing research, the marketers get more or less the answers to the defined problem. In Web 2.0 research, the marketers never know previously which kind of information they are going to find out. It might be possible to gain information about the image of a product, trends and the amount of comments (cp. Budak, interview in August 2009).
The last step of the marketing research process is the preparation of a presentation. In classical marketing research, the findings of the study are presented effectively and free from bias (cp. Baines & Chansarkar, 2002, p. 34). Researchers use for example PowerPointPresentations to show the results of the survey. In Web 2.0 research, the results are not free from bias because the four qualities are not fulfilled. The way of presentation in PowerPoint is the same.
The chosen selection method determines who is going to be asked in the survey when gathering primary data. For a high informational value, it is desirable to test as many elements as possible of the defined population in order to receive exact data (cp. Pepels, 2007, p. 35).
In classical marketing research, as shown in figure 3, p. 19, there is the possibility to have a census or a sampling:
In a census, all objects of a statistic entity are asked. On the one hand, it offers a high level of information and avoids mistakes. On the other hand, it causes high costs and requires much time. In consequence, it is mostly ineffective and not practicable. That is why it is almost never done (cp. Pepels, 2007, p. 36). One typical example of a complete inventory count is a census of population in which every citizen is counted. Other examples are analyses of customers of a car brand, students of a university, and members of a sports club, etc. (cp. Weis & Steinmetz, 2005, p. 79).
Sampling is an important procedure in classical marketing research because it is mostly impossible to collect data from every relevant person within a population as it is done in a census. Kotler and Armstrong (2004, p. 142) state:
“A sample is a segment of population selected to represent the population as a whole. Ideally, the sample should be representative so that the researcher can make accurate estimates of the thoughts and behaviors of the larger population.”
It is the aim to approach the level of information to a complete inventory count, but with less costs and time. Because of the fast survey process, gained information is up to date (cp. Pepels, 2007, pp. 36). Examples of samplings are surveys of television viewers, image analysis and psephology (cp. Weis & Steinmetz, 2005, p. 79).
When analyzing electronic word of mouth in the Web 2.0, only already existing comments of Internet users can be analyzed. Compared to classical marketing research, the reach of consumers is global. The marketers have to enter the key words in the search bar in the language they want to get the information in. For example, if they enter English keywords, they get results from Australia, USA, UK and even from other countries when people communicate in English. The group of users is limited to certain individuals who are active in forums and blogs. There is an inequality of users in blogs and forums due to keyboard skills, writing skills, individual expression skills, creativity and knowledge. Also Internet access, broadband and technical skills arise inequality (cp. Rausch & Stegbauer, 2006, pp. 96). Moreover, the motivation and the personality of users explain why only some of them generate Web 2.0 contents. That means that the sample of customers in the Web 2.0 does not represent the population as a whole. Thus, researchers cannot make accurate estimations of the thoughts and behaviors to a wider customer base. When analyzing blogs and forum in the Web 2.0, it is not possible to classify it in a strict census, nor in a strict sampling: On the one hand, all relevant comments are analyzed and there is no selection of certain users with specified characteristics. So analyzing all comments would be a census. On the other hand, not the whole customer base is asked but marketing researchers analyze only the opinion of the group of some Internet users which are limited. This would be a sample. But in contrast to a sample in classical marketing research, the analyzed word of mouth is not representative for the whole population. In consequence, the analysis of electronic word of mouth in the Web 2.0 can neither be compared with a census nor with a sample of classical marketing research.
There are two different forms of statistics: Descriptive and inductive statistics. The following figure 5 shows the constitution of statistics:
illustration not visible in this excerpt
Figure 5: Statistics (own illustration on the basis of Pepels, 2007, p. 41)
In classical marketing research, descriptive and inductive statistics are possible depending on the sample or census. In Web 2.0 research, only descriptive statistics are possible since the respondents are not representative for the whole population.
Descriptive statistics contain all methods to describe the gathered data:
In a univariate frequency distribution, only one variable is analyzed. Aggregation and homogeneity of data can be shown. For example, the modus is the most frequent value and the median is the middle value of an ordered list of data (cp. Baines & Chansarkar, 2002, p. 130). The arithmetic mean is obtained by finding the sum of observations divided by the number of observations (cp. Baines & Chansarkar, 2002, p. 139).
Measures of dispersion characterize the variability of data and show how representative the mean is for the entity (cp. Pepels, 2007, p. 163). The range is defined as the difference between the highest and lowest values in data. Quartiles are the three values which split the distribution into four equal parts so that there is a first (or lower) quartile, a second (median) quartile and a third (or upper) quartile. Variance and standard deviation take into account how all the values in the data are distributed around the mean (cp. Baines & Chansarkar, 2002, pp. 134). For more information to descriptive statistics and the corresponding formulas, see appendix, pp. i, annex 2.
In summary, in classical marketing research as well as in Web 2.0 research, descriptive statistics is possible. In Web 2.0 research, the marketers can count the amount of relevant comments, separate them into positive and negative ones and analyze the most frequently word. However, since the comments are not transferred to a numeric value, a measure of dispersion is unusual in Web 2.0 research. Because of the scope of the master´s thesis and the missing relevance for the next chapters, bivariate and multivariate analyses are not further explained.
In contrast to descriptive statistic, inductive statistic is not possible in Web 2.0 research and is reserved to classical marketing research:
Inductive statistic allows conclusions about the generalization of samples for the whole population. Two different kinds of inductive conclusions can be made: A direct conclusion from the population to the sample (selection through selection method) and an indirect conclusion from the sample to the population (representativeness through extrapolation) (cp. Pepels, 2007, pp. 40). It is important to use a suitable sampling method in regard to the objectives of marketing research. The sample can be done with a probabilistic (objective) or non-probabilistic (subjective) method:
The probabilistic method is based on the principle that every element of the population has a calculated chance to be withdrawn. The results are representative. There are different types of random selection, for example the simple random sampling: It is a probability sampling procedure where each population element is assigned a number and the desired sample is determined by generating random numbers appropriate for the relevant samples size. The population is assumed to be homogeneous. All samples of a particular size have an equal chance to be selected. Examples are the use of an urn for choosing football teams in a tournament or the use of a drum in National Lottery (cp. Baines & Chansarkar, 2002, p. 153).
In non-probabilistic methods, the researchers themselves decide which elements are chosen for the sampling (cp. Pepels, 2007, pp. 57). That means that the outcome is not as reliable as the probabilistic methods. One example is the quota sampling: This non-probability sampling procedure restricts the selection of the sample by controlling the number of respondents by one or more criterion (cp. Baines & Chansarkar, 2002, pp. 158). It bases on the idea that a smaller model of the population can be developed. That means a sample can be created in which all for the population relevant characters are representative. This distribution is called quota. This method is most frequently used in surveys. Demographic information like age, gender, residential zone, occupation is important in the selection process. In the end, the researchers choose people fitting to the given quota so that the sample is representative (cp. Pepels, 2007, pp. 58). Both procedures show why it is not possible to use inductive statistics in Web 2.0 research: To stress again, the active users in blogs and forums are not representative and therefore, it is not possible to conclude the gained results to the wider customer base.
In classical marketing research, the gained information can be empirically verified by a significance test which is not feasible in Web 2.0 research:
“A test of a statistical hypothesis is a procedure for deciding whether to accept the hypothesis on the basis of the significance of the observed result from the sample” (Baines & Chansarker, 2002, p. 169).
The process of significance test is shown in the following figure 6:
illustration not visible in this excerpt
Figure 6: The Process of Significance Test (own illustration on the base of Pepels, 2007, p. 66).
Statistics is based on probabilities and distributions. It gives information about how often an occurrence appears and how likely it is to occur (cp. Pepels, 2007, p. 40).
The first step is to formulate the statement of the null hypothesis and of the alternate hypothesis. The researcher assumes characteristics of the population based on the sampling data (“if…then...”, “the more… the more…”). There are two excluding hypothesis: The null hypothesis H0 is formulated that it should be rejected. The alternative hypothesis H1 is true, if the null hypothesis is rejected. The test for the hypothesis can be one-tailed or two-tailed (cp. Pepels, 2007, p. 66).
The two hypotheses are examples of formulations for a parametrical test (t-Test):
H0: The consumers evaluate the product in average from good to very good.
H1: The consumers evaluate the product in average worse than from good to very good.
The complete example is to find in the appendix, p. iv, annex 3.
The second step is to choose the level of significance. It determines the decision about the acceptance or rejection of the formulated hypothesis. Because of having tested a sample and not the whole population, one can only decide with a certain probability. The significance test proves, if a result is significant or due to a random sampling error. The level of significance is mostly set from 0,1% (very high significant), 1% (high significant), 5% (significant) or 10% (faintly significant) (cp. Pepels, 2007, pp. 66).
The third step is to choose the test statistic. For large samples with more than 30 cases and randomly taken samples, it is assumed that the distribution of sample mean is normal. See a figure of the normal distribution in the appendix, page v, annex 4.
The assumption of normal distribution allows undertaking a parametric test. Non- parametric tests, not requiring such assumptions, are therefore generally used for small samples. Moreover, the choice of test depends on how many samples are used: One sample, two independent samples, a paired sample or more than two independent samples (cp. Baines & Chansarkar, 2002, pp. 170).
 Wikipedia is a free encyclopedia on the Internet in 27 languages. Anyone can edit the articles.
 Xing is a social software platform for professionals.
 HTTP is an abbreviation for `Hypertext Transfer Protocol´. It is an application level protocol for distributed, collaborative, hypermedia information systems.
 PDA is an abbreviation for “Personal Digital Assistant”. It is a portable computer which is used for personal programs like calendar and address services.
 Abbreviation of `podcast´. A podcast is an audio or video data, which can be downloaded from the Internet.
 The awareness level of blogs is about 75% in German speaking countries. Compared to the leading countries USA, China, England and France, the quality and quantity of German blogs are far behind. Nevertheless, there is trend for an increase is visible in Germany (cp. Höckel, 2008, p. 45).
 In their research, Quester Karunaratna & Lim (2001, p. 3) take shoes/sneakers to represent the “total involvement category” and ballpoint pens to represent the “minimal involvement” product category. They found out that repeating purchase behavior of a high-involvement product is an indicator of brand loyalty, whereas repeating purchase of a low-involvement product is simply habitual purchase behavior.
 As shown in the following chapters, Web 2.0 researchers can influence the customer´s word of mouth, when conducting a Web 2.0 primary research.
 SPSS is a computer program used for statistical analyses.
 Chapter 4.2 gives further information about technical requirements.
 In classical marketing research, researchers would ask all costumers (census) or a representative mass of the customers (sampling).
 The normal distribution plays an important role in marketing surveys. Measurements such as income, profits, turnover and other interval scaled measurements follow a normal distribution. It is specified by the parameters `mean´ and `variance´. The distribution is symmetrical around its mean. The total area (probability) under the normal curve is 1. Half of the area is to the right of the mean and the other half to the left of it (cp. Baines & Chansarkar, 2002, pp. 140).
Master's Thesis, 97 Pages
Presentation (Elaboration), 37 Pages
Seminar Paper, 26 Pages
Bachelor Thesis, 40 Pages
Diploma Thesis, 63 Pages
Term Paper (Advanced seminar), 31 Pages
Master's Thesis, 97 Pages
Presentation (Elaboration), 37 Pages
Seminar Paper, 26 Pages
Bachelor Thesis, 40 Pages
Diploma Thesis, 63 Pages
Term Paper (Advanced seminar), 31 Pages
GRIN Publishing, located in Munich, Germany, has specialized since its foundation in 1998 in the publication of academic ebooks and books. The publishing website GRIN.com offer students, graduates and university professors the ideal platform for the presentation of scientific papers, such as research projects, theses, dissertations, and academic essays to a wide audience.
Free Publication of your term paper, essay, interpretation, bachelor's thesis, master's thesis, dissertation or textbook - upload now!