‘Advice on how to improve interactivity and functionalities of websites
in the Dutch FMCG business’
Author: Cyril Westerhof
Date of completion: 14
thAugust 2009
Author: Cyril Westerhof
University of Groningen
Faculty of Economics & Business
Management summary
This report evaluates the actual state of online advertising in the Dutch FMCG market. Nowadays, with growing internet marketing opportunities this part of all marketing tools should be investigated more. A company has several online advertising possibilities available, but in this report the brand website is looked at. When I started my internship at Heinz Foodcompany the brand managers were struggling to create a website for their brands. During several meetings questions were asked about how to build a website for a FMCG brand and what this website should look like.
Sparked with interest I started my thesis and was dedicated to solve the issue of creating a brand website. My interest was not limited to one company, yet I wanted to know how this is currently dealt with in the Netherlands. A literature study gave insight into how a commercial website could work for a brand. It seems that specific websites need to be created for different FMCG brands. This is part of the outcome of the evaluation of 100 websites in the Dutch FMCG business. To be able to generalize the findings to the FMCG market, a division was made. FMCG companies were divided into Food and Beverage, Personal care and Household care products.
A commercial website in the FMCG business is aimed at creating brand awareness among consumers. Sometimes the objectives differs or a website is given less priority. This report suggests ways on how to handle this challenge. Companies can achieve success by creating interactivity through special functions on the website. Research that has been done into online advertising suggests using website interactivity as a fire starter for commercial success.
Form the sample of 100 websites 3 clusters are formed. First, there is the cluster (1): `Clear choice helpers` (Characteristics: high on personal choice helper, high score on website design). These websites focus on providing product information and website design, there is not a lot of interactivity going on through these websites. Next, there is the cluster (2) `Brand advertisers` (Characteristics: high customer support, high on marketing research, low on personal choice helper, good score on advertising, promotion & publicity, real-time communication). Here a lot of effort is put into providing a service platform for the consumer and creating and maintaining consumer relationship is one of the objectives. Finally, the cluster (3): Information providers (Characteristics: low customer support, low on customer research, low on personal choice helper, lower score on website design) is formed. These websites are only effective for information services and provide some special business information. Providing more interactivity factors on their commercial website could significantly improve these websites. It can be stated that for each cluster several improvements can be formulated. Functions like downloading of
software, communication (real-time chat) and new product proposal are not present on
most FMCG websites. Including these functions in the brand website would definitely improve consumer focus and could play an important role in consumers’ brand preference. In general, FMCG companies can do a lot more to build a proper relationship between the brand and the consumers by offering certain website functions.
Preface
Table of content
1. Introduction ... 7
1.1 Internet usage ... 9
1.2 Problem statement ... 10
1.3 Research Questions ... 11
2. Function and Design of websites ... 13
2.1 Types of website and functions ... 14
2.2 Development in design ... 16
2.3 Design preferences ... 17
2.4 Interactivity and brand preference... 18
2.5 Interactivity of a brand’s website ... 19
2.6 Conclusion ... 21
3. Brands and the website ... 22
3.1 What is the Brand? ... 22
3.2 Brand equity ... 23
3.3 Brand preference ... 24
3.4 Different brands ask for different websites ... 26
3.5 Online Communities and other tools ... 28
3.6 Web 2.0 opportunities ... 29
3.7 Where to concentrate on? ... 29
3.8 Conclusion ... 31
4. What can be measured to differentiate between websites? ... 32
4.1 Measurement of performance... 32
4.2 Measurement of website performance ... 33
4.3 What determines website quality ... 34
4.4 Conclusion ... 35
5. Research design ... 36
5.1 Evaluation form for websites ... 36
5.2 FMCG Categories ... 38
5.3 Research method ... 39
5.4 Cluster analysis ... 40
5.5 Alternative insights into FMCG websites ... 41
6. Results of research ... 43
6.1 Overview of outcomes ... 43
6.2 Results per function ... 44
6.3 Website functions ... 44
6.4 Characteristics per cluster ... 46
6.4.1 Results of brand websites ... 49
6.5 Results of alternative insights ... 50
7. Conclusions and recommendations ... 53
7.1 Current use of website functions ... 53
7.2 Discussion of results... 55
7.3 Conclusion of alternative insights: Big vs. Small brands and their website ... 56
7.3.1 Hedonic vs. Utilitarian brands and their website ... 56
7.4 Opportunities for companies ... 57
7.5 Limitations of study and directions for future research ... 58
8. References ... 60
8.1 Literature ... 60
8.2 Websites ... 64
9. Appendices ... 65
Appendix 1 : Scores per category ... 65
Appendix 2: Mode overview ... 69
Appendix 3: Cluster per brand website ... 70
Appendix 3: Cluster per brand website ... 70
Appendix 4 : Dendrogram of the cluster analysis ... 74
Appendix 5: Agglomeration Schedule ... 74
Appendix 6: Graphical presentation of Agglomeration Schedule ... 76
Appendix 8: Cluster formation in percentages ... 82
Appendix 9: Overview of websites per cluster ... 83
Appendix 10: Overview Mean scores per function ... 85
Appendix 11: Website categories ... 89
Appendix 12: Top 100 biggest brands ... 92
Appendix 13: Average interactivity scores per factor (Big & Small) ... 95
Ghose & Dou (1998) claim that websites are becoming important mechanisms for marketing communication. It is therefore vital to understand what factors affect the attractiveness of a firm’s website. The two researchers mention this because success, measured by website popularity, can be heavily dependent upon this factor. When attractiveness is mentioned this entails numeral aspects that will be further dealt with in the following chapters of the literature study. What Ghose & Dou (1998) report should also give an insight into the opportunities there are both in scientific (interactive functions related to market performance) as in practical viewpoints (investing in website promotion by product mangers). They mention that “in fact, with the rapid increase in the usage of corporate web sites as firms communication vehicles, the competition in building
interactive websites to attract more surfers is likely to get more intense. Ghose & Dou
These branded website gives a good overview of what is possible with a commercial website.
Keller (2008) mentions that there are certain advantages to internet marketing (like
measurability) and it does not disturb the customer during an activity and targeting can be
environment. The results that are related to the study of Muller et al. (2008) show “that from a pure managerial standpoint, results are encouraging and support current management practises of increasing investment in online spending as well as in website development and support”. As Huizingh (2000) suggests there is ample anecdotal evidence suggesting that the internet is an effective tool for commercial purposes. Now it is up to the practitioners to try to make something out of this new challenging opportunity.
1.1 Internet usage
organization develops an Internet-based system to actively communicate and interact with customers, clients, or patients. They (Brodie et al.) have also shown that the adoption of e-marketing is positively associated with performance, which supports Dan and Bens (2005) contention that firms adopting e-marketing are likely to have competitive advantage. As noted by Barwise and Farley (2005) in their surveys of firms’ advertising and marketing communication activities in seven countries (United States, Japan, Germany, United Kingdom, France, China, and Brazil), there has been a steady growth in Internet advertising and other e-marketing associated with communication in the last few years. “From 2002 to 2004, Internet-based E-marketing increased its share of spending from 7% to 8.2% while traditional media advertising and sales promotion shares of spending had slight declines (Brodie et al. 2007). This shows the relevant adoption in the business environment.
1.2 Problem statement
This introduction and the related difficulties that companies encounter, while developing online advertisements and websites, has led to the following problem statement:
How do Dutch FMCG firms use the Internet as a medium to create brand preference?
the commercial website, and this is important in understanding how a website can be set up the most effective way to create brand preference.
Conceptual framework of research:
With the following research questions the answers to the general problem statement will be found. These research questions will be discussed in the literature review.
Subsequently the answers will form the conclusions on which the recommendations are based.
1.3 Research Questions
To understand more about how a website can be used as a tool to market an FMCG brand more information is needed about the functions a website can have. Chapter 1 will elaborate on this topic.
• RQ1: What are the general functions of a website and how can they be used to
market the brand of the FMCG company?
• RQ2: What determines the quality of a website?
• RQ3: What are the objectives of a commercial website? • RQ4: How can a website influence brand preference?
• RQ5: How should a website be used in the Dutch FMCG Business?
There are different functions a website can have. There are functions that make it possible for the visitor to interact with the brand website. This enables the visitor to express certain feelings towards the brand website. This can create a relationship with the specific brand. Chapter 4 will go further into what makes a website work. There are different measures that can used to determine quality. This is related to the way a website can be
FMCG Brand interacts with customer FMCG Brand preferred by customer FMCG Brand website success and increased sales Website Type
2. Function and Design of websites
received little attention (Peterson, 1997). Nowadays that is very different; one of the most noticeable differences between e-shopping and store shopping is attributed to travel. One of the biggest advantages is the time gained and the efficient way of shopping that appeals to the modern person. Specifically, if a consumer decides to go to physical stores to gather information on products; i.e. conduct shopping activity, he or she has to spend travel time and travel cost to reach the shopping places. This will not be the case, on the other hand, if he or she chooses e-shopping (Ming-Hsiung Hsiao, 2009) When functions of the Internet are discussed certainly this aspect must be addressed.
2.1 Types of website and functions
In this paragraph the types of commercial websites will be discussed. The paragraph after that will explain more about how design plays a role in creating brand preference. The attractiveness of a website can be important for customers that see the website for the first time. What can be seen is that there are several types of websites that represent the strategy with which a company prefers to invest in online advertising as a marketing tool. Among those are ways of image building, sales channel, or a combination of both. According to McNaughton (2001) there are three different styles that websites can have. The first is an “image-building” function website, this site enforces the image and reputation of the company. Also, there are “sales-assistance” sites, these functions as a sales line for the company. Finally, there is the “integrated” site, which incorporates both above mentioned styles (Dann & Dann, 2004). What Benslamani & Yang (2006) suggested was that there were two ways in which (commercial) websites could be categorised. The first was the informational website, where there is not much focus besides “presenting their companies, describe the products and services offered, answer frequently asked questions, and provide alternative methods of contact. They also report that previous research has shown that the quality and the organization of the information provided are key attributes of such websites’ success. The other was the `complex`
website. This is a site where also other steps of the consumer process are offered, like
shopping mall, incentive site and search agent. Benslamani & Yang (2006) also report that “despite the potential benefits, studies show that a relatively small number of commercial websites are transactional. “Their complexity and their high development and maintenance costs can explain the relatively small number of transactional websites on the World Wide Web”. To better understand how brand websites can be used it is important to know what is desired by the market. Establishing a dialogue with customers, to understand and anticipate their needs, is key to offer better and personalized products and services (Lewi, 2005). Pleasant and well-designed websites are no longer sufficient for attracting customers and making them want to return. Substantial amounts are invested nowadays to develop additional services in order to deliver superior website experiences, which are able to generate revisits and enhance customer relationships. Perhaps the relatively low amount of transactional (or sales-assisting) websites can be explained by the threat of cannibalisation companies perceive (Deleersnyder et al., 2002). Levy Strauss, for example, already withdrew its internet channel after 1 year, mainly because of backlash from resellers and out of fear of getting ruined through cannibalization losses. However, Deleersnyder et al (2002) conclude that the Internet does not automatically need to be disruptive for others channels. Businesses need to see the possibilities it has and need to be creative to act on this change.
website, up-to-date prices, etc.); 5: Maintenance (updates and maintenance of website) ; 6: Metrics (measurement units, traffic, etc.).
Different objectives can result in either flat brochure websites or attractive and interactive
websites. It is the visitor that makes the final judgement whether the website is successful
or not. Many researchers link website success to interactivity (Tse v& Chan, 2004, Ghose & Dou, 1998, Hoffman & Novak 1995). So it can be assumed that in order for companies to be successful they will have to take this factor into account. Hoffman & Novak (1995) have posited that the future of advertising and communication will be represented by interactive presentations on the web. For different surfers there are different objectives. Yang et al. (2004) have suggested that for potential customers, IP web-portals are a platform that enables them to become acquainted with an organization, to explore its goods and services, and to make inquiries. For existing customers, IP web portals are a place to become more familiar with the organization, to obtain product and service-related information, to request services, and to exchange information with peer users of the same goods and services. Because interactivity seems to be an important issue here, in table 1: `Interactive functions`, the specific functions are mentioned that lead to interactivity according to Ghose & Dou (1998). These functions will later be accompanied with additional functions that will complete the evaluation form for the websites.
2.2 Development in design
material (documents, rules, protocols) to the web (webification). Last step here is the placement of personal stories onto the web. Forums and different ways to give one’s opinion (blogs) are part of this step (Dann & Dann, 2004). Tse & Chan (2004) suggest that one of the functions of a website should be to provide interactivity, because this is what makes the website score higher on popularity polls. They have identified 6 different types of interactive functions. Out of these the following seemed to be most important: Online two-way communication, Customization, Content variety and Job search. They also mention the research done by Ghose & Dou (1998). This study shows that
interactivity affects website performance. Tse & Chan (2004) have gone deeper into these
functions and have elaborated on the different types. The internet provides companies with a new way of promoting their offerings; Stewart & Pavlou (2002) have used the following to illustrate this. They point out that the real power of the internet and the characteristic that differentiates it from traditional media is its potential for interactivity. This supports the idea of Tse & Chan (2004) who also found that 1 of the 4 most successful functions of a website is `Online two-way communication`. The definition of interactivity is given as: `A form of communication that is two-way and real-time, that involves human-computer interaction via a computer mediated environment such as the internet, and in which the responses and behaviour of both parties will affect their future responses and behaviour` (Tse & Chan 2004).
2.3 Design preferences
With the swift development and increasing use of Internet as both an information seeking and an electronic commerce tool web-users interface studies grow in significance (Zhang & von Dran, 2000). Researchers have suggested that people have certain preferences concerning website design. There are several factors that are important here. Agosto (2004) mentions eight website evaluation criteria related to biological sex (girls are respondents here): collaboration, social connectivity, flexibility, motility, contextuality, personal identification, inclusion, and graphic/ multimedia concentration. It seems that people use the Internet for social connectivity, or for creating and maintaining social connections, especially through e-mail. The participants also expressed preferences for
in navigation (especially woman). Motility also proved to be a participant design preference. Websites that enable physical control of site elements are positively evaluated. Another element that seems to be important is contextuality, people strongly prefer narrative frameworks (relevant content and story-telling). This increases interest in factual information and invites people to keep reading. Another issue is that personal
identification is also a preferred website design aspect. Information that relates to
personal lives is positively evaluated. Lastly, graphic/ multimedia concentration is important. The participants of previous research preferred sites with plentiful graphics, video clips, audio clips, etc. over less adorned sites (Agosto, 2004). There are also other factors that are important when discussing website design. Zhang & von Dran (2000) mention that with the real challenge is to identify and develop design factors that can make a website usable and serviceable, avoiding users frustration or dissatisfaction; create more stimulating, visually pleasing, comprehensive, and commercially viable websites. Also, helping to attract users to a website, maintain their interest in the website, and encourage them to return to the website again. According to Zhang & von Dran (2000) there should be basically 2 types of website design factors: hygiene factors (presence make a website useful and serviceable, whose absence cause user dissatisfaction) and motivator factors (contribute to user satisfaction, and add value to the website beyond hygiene factor). The clearly identified hygiene factors include: technical aspects, navigation, privacy & security. The clearly identified motivators are: enjoyment, cognitive outcome and credibility. These factors make users feel they are involved, either cognitively or emotionally while interacting with the website. They are the factors that contribute to one’s satisfaction with a website when present or considered. Zhang & von Dran (2000) also mention that website types may effect a user’s perception on particular hygiene or motivator characteristics. Designing an entertainment website requires different hygiene and motivator factors than an education website.
2.4 Interactivity and brand preference
interactivity plays any significant role in affecting the overall quality of their websites”. How is this all related to market performance? Hoffman & Novak (1995) point out that the web frees customers from their traditionally passive roles as receivers of marketing communication, giving them much greater control over the information search and acquisition costs, and allows them to become active participants in the marketing process. According to Upshaw (1995) the interactive nature of the web offers marketers new opportunities to create stronger brand identities that have the potential to translate into brand loyalty. This implicates a positive relation between a clear presence on the web and brand preference through loyal customers. With the help of internet functions, consumers can react and interact creating a dialogue outside managerial control (Travis, 2001). Through these developments consumers will have stronger relationships with their brands. According to Upshaw (2001) other researchers have found similar findings; “the level of interactivity on a website is critical in converting visitors from interested contacts into interactive customers. Furthermore, researchers suggested that improving the overall quality of online retailers is essential to customer satisfaction in the rapidly growing world of B2C e-commerce (Liu & Arnett, 2000; Janda e al., 2002; Park & Kim, 2003). This research relates to online retailers an not to brand websites, but the meaning and relevance of customer satisfaction is the same in both forms.
2.5 Interactivity of a brand’s website
Table 1: Interactive functions
1. Online form: for customer feedback 2. Online form: inquiries
3. Online form: comments 4. Downloading of software 5. Online problem diagnostics 6. Order status checking 7. Site survey
8. Product survey
9. New product proposal 10. Key word search 11. Personal choice helper 12. Virtual-reality presentation 13. Dealer locator
14. Electronic coupon 15. Online ordering of goods 16. Online contest/prize 17. Push media
18. Interactive job placement 19. Electronic post card 20. Surfer postings 21. User groups 22. Games
23. Multimedia presentations
2.6 Conclusion
3. Brands and the website
In this chapter the relation between a website and a company’s brand is discussed. The brand is part of the strategic goal that companies have when creating a website. Certainly a brand is one of the first things (should be!) that a customer sees when entering a commercial website. What is a brand exactly and how should this be incorporated into a commercial website? A company wants everybody to love their brand, but how can this be realised through a website? This chapter will discuss these issues and attempts to clarify the relation between a website’s functions and the resulting brand preference. Because it is widely known that FMCG brands fall into the category of low-involvement products, a closer look will be made on the issue of involvement and how a company should react to this when evaluating the functions of a website.
3.1 What is the Brand?
According to the American Marketing Association (AMA) , “a brand is a name, term, sign, symbol or design or a combination of them, intended to identify the goods and services of one seller or a group of sellers and to differentiate them from those of competition” (Keller, 2008) . So this means that technically every time a marketer creates a new name, logo, or symbol for a new product, he or she has created a brand. Another way of explaining what a brand is, is done by Aaker (1997). She relates the definition of a brand to the personality traits of humans: Sincerity, Excitement, and Competence, Sophistication and Ruggedness. She reports that practitioners view brand personality as a key way to differentiate a brand in a product category, as a central driver of consumer
preference and usage, and as a common denominator that can be used to market a brand
company’s website should clearly carry over to people. To be able to do that it should be known what brand preference means. This is something that is achieved when people, because of certain association they have with the brand start preferring one product’s brand to another.
3.2 Brand equity
The term `brand equity` emerged in the 1980’s, there has been a burgeoning interest in the subject among marketing academicians and practitioners. In the past decade, researchers have focussed a tremendous amount of attention on the brand equity construct, which refers to the incremental utility or value added to a product by its brand name. Yoo and Donthu (2001) used a consumer-based behavioural view of brand equity and the defined brand equity as consumers’ different response between a focal brand and an unbranded product when both have the same level of marketing stimuli and product attributes. They also report that researchers have found that a product’s brand equity positively influences future profits and long-term cash flow, consumers’ willingness to pay premium prices, merger and acquisition decision making, stock prices, sustainable competitive advantage and marketing success. Furthermore, they suggest that brand equity consists of four dimensions: brand loyalty, brand awareness, perceived quality of brand and brand associations, as proposed by Aaker (1997) and Keller (2008). A website should, in order to influence consumers, use these dimensions to contribute to build their own brand equity of the website. To understand how this can be done, first these dimensions should be explained. To start, brand loyalty refers to the tendency to be loyal to the focal brand, which is demonstrated by the intention to buy the brand as a primary choice. Brand awareness is the ability for a buyer to recognize or recall that a brand is a member of a certain product category. The perceived quality is the consumer’s judgement about a product’s overall excellence or superiority. Aaker (1999) defines brand
associations as anything linked in memory to a brand and brand image as a set of brand
the retailer, or the consumer. From the perspective of the trade, brand equity contributes to the overall image of the retail outlet. It builds store traffic, ensures consistent volume, and reduces risk in allocating shelf space. None of this is meaningful, however, if the brand has no meaning to the consumer. In other words, there is no value to the company
only if there is value to the customer. For low-involvement products a brand name might
serve as a `halo`-effect through which consumers can make a quick assessment of the brand. The research of Cobb-Walgren et al. (1995) also suggested that brand equity
increases both consumer preference and purchase intentions. This makes it important for
FMCG businesses to create a positive brand image among as many people as possible. A website can potentially help achieving this goal.
3.3 Brand preference
Figure 1: (Source: Yoo, C., 2008, Unconscious processing of web advertising: effects on implicit memory,
attitude toward the brand, and consideration set. Journal of interactive marketing volume 22, No. 2, spring 2008)
3.4 Different brands ask for different websites
Functional products are subject to negative, problem-solving, purchase motives (Rossiter
information to access by clicking on various page links. Here the difference becomes clear between Internet as a marketing tool and other tools like radio, television and print. The differentiating effect will be addressed further on in this report.
The elaboration likelihood model (ELM) stipulates that there are two routes to persuasion (Petty, Cacciopo, and Schumann, 1983; Petty & Cacciopo, 1984). The central route is used for high-involvement products. Consumers are influenced by strong, product-related arguments. For low-involvement products, the peripheral route is used. In this case, the consumers are less focussed and are influenced by peripheral cues (Chattopadhyay, 1998). If a consumer enjoys these peripheral elements (such as games, competitions,
audiovisuals, etc.) he or she will be more likely to be a repeat visitor of the website. In
situations in which consumers pay conscious attention to the advertisement, in the case of games etc. The ads are processed through unconscious means (Yoo, 2008). This makes clear that these interactive brand websites do not cause sales to increase directly, but in stead they have another effect on the longer term. Practitioners should therefore also not be discouraged when they do not see a significant rise in sales after initiating a brand website project. In line with this conclusion Yoo (2008) mentions that if an objective of a Web ad is to induce direct responses, more than 99% of Web ads are wasted. This provides a comforting thought for those brand managers missing the direct sales.
3.5 Online Communities and other tools
What can online communities do for the brand? Algesheimer et al. (2005) and Kozinets (1999) suggest that by creating online communities a sense of belonging is formed; subsequently people will also feel attracted to the company and the brand. Phelps et al. (2004) propose that if a company wants to invest in viral marketing the first thing is to set a clear target group before this viral will go life. This will optimise the amount of people that will forward the email. Also, Kuenzel & Musters (2007) investigated whether social
influence plays a role in low-involvement products and whether low involvement
game. Though a majority of visitors may recognize that games are part of the firm’s advertising (Ducoffe, 1996), they may still love to play because games provide escapism, diversion, and possibly emotional release. These examples form a clear vision of what
interactivity is all about when related to a brand’s website.
3.6 Web 2.0 opportunities
As Kangas, Toivonen & Back (2007) suggest more can be done to interact with the customer. The way companies use the Internet by advertising only is long gone, companies can now interact and ask the consumer directly what they want. The Mona website (of the Campina dairy products company; www.mona.nl) is a perfect example of an organisation that is interacting with their consumers about what they want and how they want it. The website gives consumers the possibility to create their own pudding dessert, by means of well-designed and funny - “Holland’s next toptoetje” - production process). Once you have created the dessert of your personal preferences, you can also cats a vote for another dessert, which is then evaluated on popularity and the dessert is given the chance to shine in the grocery shopping stores. Social media is particularly based on user generated content (UGC), which may be:
• New content, people could send their photo’s, videos, music or text. • Modified content, such as compilations, videos mixes or mash-up services
• Categorised content, such as playlists, reviews or keywords (also classified as
metadata)
Of course, by using these techniques to promote a company’s brand, opportunities exist to form certain communities. Virtual Communities that have a positive attitude on the brand can create many brand fans.
3.7 Where to concentrate on?
engaged with games, competitions and other peripheral cues than with the brand. Companies could downplay the peripheral cues and concentrate more on the brand, but this could have a negative effect on the attractiveness of the website. Marketers should make a decision whether to concentrate on the brand or other things. Research done by Gong & Maddox (2003) suggests that online advertising improves user’s brand recall, changes their attitude toward the brand, and increases their purchase consideration. But also they mention that there are several factors that moderate the advertisement effectiveness. Among those is `issue involvement`. Consumers pay more attention to advertisements if they are specifically interested in the product. Research has been done about measurable effects of advertising research, such as attitude change. Research reports that these effects are moderated by the frequency of exposure and other intervening variables such as issue involvement and initial attitude. Another reason to invest in online marketing is the fact that the engaged state that the web encourages seems to result in a higher initial attention to advertising Gong & Maddox (2003). Situational involvement affects advertisements recall, but enduring involvement was found to account for most of the variance on brand recall and other marketing ratio’s. This result is in line with the research of product involvement on advertising effectiveness done by Petty & Cacioppo (1986). The more people are interested in a specific product the more they want to know about it. Finally, also in line with a 1997 IAB study, researchers concluded that banner advertisements do work and the web is good for brand
building and consumer communication. This shows again that in order to build a brand,
3.8 Conclusion
4. What can be measured to differentiate between websites?
In order to convince management of web-advertisement expenditures, things needs to be quantified and measured. In order to do this more insight is needed in what the goal is of the online investments and what factors quantify this. This chapter will provide more insight into how this can be done.
4.1 Measurement of performance
issue is effectively communicating the corporate message. As Muller et al. (2008) have suggested the environment in which the internet is being used has evolved the last couple of years. “In this new environment, a well designed website is no longer sufficient. Delivering a satisfying and valuable experience online is important and may be key to increasing favourable brand perceptions and purchase intent (Flores, 2004). Another interesting fact is that online activities can flourish when offline activities are undertaken in correspondence.
4.2 Measurement of website performance
First off, Yang et al. (2004) reported that there is currently no established conceptual foundation for developing and measuring the (service) quality of the Websites in general. This adds to the demand for creativity and ability to pick out those factors that are important for each brand website. Because in this report only informational/promotional websites are relevant there is no need for a tool to fit all websites. As Muller et al. (2008) report traditional tools measuring traffic performance such as number of visitors to a website and page views are no longer sufficient specifically for brand websites. As they found out a website describing the technical or functional characteristics of the products offered may decrease total consumer search. According to Muller et al. (2008) a successful brand website must clearly offer more than product description to maximize consumer experience and branding. They also mention that the focus of attention in the literature on the internet is increasingly directed to website design and the identification and classification of different website dimensions, such as entertainment, interactivity, ease of use, information content and quality (Chen & Wells, 1999; Cheuing & Lee, 2005; Eighmey, 1997; Ghose & Dou, 1998; Shankar, Smith and Rangaswamy, 2000; Wu, 1999. One of these dimensions is interactivity. Researchers have even found that purchase intention seems mainly affected by similar variables, including interactivity (Wu, 1999; Yoo & Stout, 2001). So this seems an important factor of a website in order to create positive market results. Next to interactivity other measures have been found to have influence on the way people react to a website. The most prominent one seems to be
overall navigation satisfaction which has a positive influence on website loyalty
the research performed it became clear that several quality factors determined customer satisfaction related to online retailers. Lin (2007) takes several quality variables that are relevant for the online retailing business, not all of these are so for a brand website. Website quality dimensions according to Lin (2007) are system quality, information quality and service quality. Only system quality is related to the research in this report. The other two are related specifically to online retailers and that is beyond the scope of this research. The relevant quality dimension exists of website design and interactivity, these are directly related to the 23 factors mentioned by Ghose & Dou (1998). They cover the most important factors that make a brand website to a success. They mention 5 dimensions of interactivity; customer support, marketing research, personal-choice helper, advertising/promotion/publicity, entertainment. Out of these dimensions they have distilled 23 factors that form a concrete measure to a website’s interactivity level.
4.3 What determines website quality
Cao et al. (2005) mention that there is no particular body in literature that specifically addresses the concept of e-commerce website quality. Therefore, they use the TAM method to predict system acceptance, which predicts user acceptance by two factors: 1)
perceived usefulness, and 2) perceived ease of use. This does not directly relate to how
managers should set up their brand website, but still gives an idea of what is expected from a system that needs to be operating with people. To be able to deliver a commercial website that suits the needs of the customers DeLone and McLean (1992) proposed six major dimensions of website success: systems quality, information quality, use, user satisfaction, individual impact and organizational impact. However, as Cao et al. (2005) hurry to mention these six dimensions do not include the human interaction aspect (Li, 1997). To be able to connect the website’s effect to what customers expect is to be a connection can be made to customer satisfaction literature. “Service quality is commonly defined as how well a delivered service level matches customer expectations (Parasuraman et al. 1985). Preferably, the 5 dimensions of Ghose & Dou (1998) should be used to determine what customers expect from a commercial website. The WebQual
measurement made up by Liu & Arnett (2000) identifies 4 factors that are critical to
design quality. Again this is related to e-commerce business, a brand website does not specifically need to be an e-commerce website.
4.4 Conclusion
5. Research design
To be able to answer the question (problem statement): `How do Dutch FMCG firms use
the Internet as a medium to create brand preference?`, a field research is done to see just
how much Dutch FMCG companies use functions of websites to communicate their brand. As mentioned before the descriptive research will incorporate 100 websites that promote their company’s brand. These websites are all categorized and than evaluated on their interactivity and other important factors. Based on previously discussed literature a systematic approach is formed to evaluate these 100 websites. The outcome of this research will show results on how functionalities are currently used in the Dutch FMCG business.
5.1 Evaluation form for websites
Table 2: Evaluation form
Function of website Measure
1. Customer support *(Ghose & Dou, 1998) 1. Online form: for customer feedback 2. Online form: inquiries
3. Online form: comments 4. Downloading of software 5. Online problem diagnostics 6. Order status checking 2. Marketing research (Ghose & Dou, 1998) 7. Site survey
8. Product survey 9. New product proposal 3. Personal choice helper (Ghose & Dou, 1998) 10. Key word search
11. Personal choice helper 12. Virtual-reality presentation 13. Dealer locator
4. Advertising/ promotion/ publicity (Ghose & Dou, 1998)
14. Electronic coupon
Also: Tse & Chan (2004) 15. Online ordering of goods 16. Online contest/prize
17. Push media (email newsletter) Also: Tse & Chan (2004) 18. Interactive job placement 5. Entertainment (Ghose & Dou, 1998) 19. Electronic post card
20. Surfer postings Also: Algesheimer et al.(2005) and Kozinets (1999) 21. User groups
22. Games
23. Multimedia presentations 6. Communication extra (Tse & Chan, 2004) 24. Online two-way communication 7. Website design factors (Dahlén, 2003) 25. Navigation flexibility (presence menu) Also: Evans & Wurster (1999) 26. Informational content
Dahlén (2003) and Evans & Wurster (1999) 27. Credible content
5.2 FMCG Categories
The evaluation will show to what extent Dutch FMCG companies make use of included factors on their brand websites. All 100 websites will be evaluated according to the same factors. This will determine how they use websites currently and how they can improve this in the future. Recommendations are proposed based on literature research and the outcomes of the actual research of 100 FMCG websites. Based on the following table companies’ websites have been chosen to participate in the research. To come to an overall distribution of types of products the following table is taken as a starting point (Source: report of PWC on http://www.ibef.org ). This way an even division is made in different categories that form the FMCG market. Namely, (1)`Household Care`, (2)`Food
and beverages` & (3)`Personal Care`.
Table 3: FMCG Categories explained
Category Products
Household Care Fabric wash (laundry soaps and synthetic detergents); household cleaners (dish/utensil cleaners, floor cleaners, toilet cleaners, air fresheners, insecticides, and mosquito repellents, metal polish and furniture polish).
Food & Beverages Health beverages; soft drinks, staples/cereals; bakery products, (biscuits, bread, cakes), snack food; chocolates, ice cream, tea, coffee, soft drinks, processed fruits, vegetables, dairy products; bottled water, branded flour, branded rice, branded sugar, juices, etc.
Personal Care Oral care, hair care, skin care, personal wash, (soaps); cosmetics and toiletries, deodorants; perfumes; feminine hygiene; paper products.
5.3 Research method
This report contains descriptive research. The functionalities of FMCG websites is documented according to 27 interactive and design functions. The 27 factors are used as a measurement tool (see: table 2) in this research. These evaluation factors are, according to Ghose & Dou (1998): Customer support, Marketing research, Personal choice helper, Advertising/ promotion/ publicity and finally Entertainment. Also, Tse & Chan (2004) propose to measure interactivity by using 6 broader categories: Online ordering, Online two-way communication, Customization, Connection links, Interactive Job search & Content variety (the last one not taken as a measure because practically impossible to
track). Next to these points also design factors are taken into account. For a complete
overview of these factors see the evaluation form. Design factors are included according to 2 types of website design factors: hygiene (presence make a website useful and serviceable, whose absence cause user dissatisfaction) and motivator (contribute to user satisfaction, and add value to the website beyond hygiene factor). The clearly identified hygiene factors include: technical aspects (not included), navigation (flexibility), privacy & security (not included). The clearly identified motivators are: enjoyment (already
included by entertainment), cognitive outcome (informational content) and credibility
functions and how these are present in the 100 websites is shown in the chapter Research results (chapter 7). After this, the points per function are looked at, in order to decide which functions are present most frequently. This is done by simply counting the points that a website has gathered by using certain functionalities. The total sum of these points is used to decide how interactive a website actually is. This way it becomes clear how each function is used, if it is used at all, in brand websites. Here it becomes clear how some categories differ in the usage of website functions. This can be noticed when points per category are compared. Results of FMCG segments are discussed next. Also, the results of the cluster analysis is discussed in the next chapter. This analysis gives a good insight into how these websites can be clustered together based on their characteristics. Finally, results of brand websites are put in an overview, presenting how the 100 websites relate to each other and which website belongs to which cluster. This gives more information about the extent to which a website in interactive.
5.4 Cluster analysis
looking at the dendrogram (see Appendix 4) and the agglomeration schedule (see Appendix 5) of these scores a decision can be made about the amount of clusters. The graph based on the agglomeration schedule also shows a clear steeper line at the third division of variables. In this report 3 clusters are used to cluster websites together. Selecting more clusters would be possible but not desirable. In this report defining several characteristic groups is more important that forming most homogenous groups. Therefore, the amount of 3 clusters is chosen. The research is exploratory in nature and tries to label the websites as much as possible in a general way. Based on these clusters relevant recommendations can be made. All this information is used to come to answers to the research questions that are formulated to gain insight into the interactivity and functionalities of Dutch FMCG websites.
5.5 Alternative insights into FMCG websites
To be able to gain more insight into the websites and their interactivity, the websites have been categorized even further. This in done by dividing the websites into different groups, so that the websites can be better evaluated. The first division is done into `big` and `small` websites. Off course this is done by a certain measurement (see: appendix 12), here the Foodmagazine Brand top 100 (Information Resources 2004) has been used. This is a list of the top 100 biggest brands based on average sales.
6. Results of research
6.1 Overview of outcomesOutcomes of the interactivity per website factor are evaluated. To present the outcome in a clear overview a relative score is calculated. The current score is divided by the maximum score (all factors present on website). This way it becomes clear what functions are used in which frequencies.
Table 4: Score per factor over all 3 FMCG categories*
All 3 categories/ factor
0 10 20 30 40 50 60 70 80 90 100
HC F&B PC HC F&B PC HC F&B PC HC F&B PC HC F&B PC HC F&B PC HC F&B PC HC F&B PC Customer support Marketing research Personal choice helper Advertising, promotion, Entertainment Communication extra Website design factors Category total S c o re %
Table 4: *HC = Household Care, F&B = Food & Beverage, PC = Personal Care: this table shows potential of website category in relative terms. A relative percentage is calculated based on score and maximum score.
6.2 Results per function
The score of different website functions is the outcome of this research. Table 4 shows the frequency of usage of website functions. The scores show that as a category `Food
and Beverage` has the highest score in points, thus the most interactive and design
functions are present on websites that operate in this part of the FMCG business. Also, when all scores for each function are added together (see Appendix 2) it shows that in total a few functions seem to be most frequently used. These are Online form / feedback (interactivity factor) and Credible content (design factors). This indicates that these are most present when all categories are taken into account. Per category a distinction is made between interactivity factors and design factors. This is done because both factors have different goals and therefore these should be assessed separately. Also, this gives a better insight into what functions are used most frequently. Per category (see Appendix 2) it shows that in the `Household Care` category Personal choice helper (40 points),
Informational content (also 40 points) and Credible content (40 points as well) score
highest. In the category `Food & Beverage`, Online form/ feedback (32 points) and
Navigation flexibility (37 points) are most used. For `Personal care` these are the
functions: Online form/ feedback (32 points) and Credible content (39).
6.3 Website functions
In the relevant table (see Appendix 1) a clear overview is given on which functions are used per category. In the section `Customer support` online form/ feedback scores are pretty high with 98 points. Indicating that most websites make use of this function. This function is also used the most in Household care. Online form / inquiries also scores very high (96 points), again Household care has the highest score. Online form/ comments scores 93 points, in which Household care has the highest score. Downloading of software is less present in websites and has only 13 points. F&B websites score highest among all categories on this point. Online problem diagnostics has an average usage score of 50 points. Household care websites score the most points on online problem diagnostics. Order status checking has no points at all. In the segment `Marketing
research`, `site survey` scores very low, only 13 points. The F&B segment only scores 2
6.4 Characteristics per cluster
As mentioned before the 3 clusters provide information on which recommendations can be formulated. Brand managers can act on this by improving their commercial websites. From the dendrogram (see Appendix 4) the amount of 3 clusters is chosen. Each cluster has specific characteristics that illustrate the current way FMCG companies use their brand website. The outcome of the cluster analyses is that cluster 2 has scored highest on
Customer support factors. Cluster 3 scores lowest on these issues. Next, on Marketing research also cluster 2 scores highest and cluster 3 scores lowest, cluster 1 forms the
average score. On Personal choice helper, cluster 1 scores highest there. The other 2 clusters have a lower score in this function. On Advertising, promotion & publicity cluster 2 has most relevant functions included. On the segment that includes functions for
Entertainment cluster 2 scores highest and cluster 1 clearly has the lowest score. On Communication extra (real-time), only cluster 2 has scored a point. On the function that
Table 4: Descriptive statistics of the clusters Cluster 1 (n=61) Cluster 2 (n=16) Cluster 3 (n=23) Variabelen Mean Std. Deviation Mean Std. Deviation Mean Std. Deviation V1: customer feedback 0,98 0,128 1,00 0 0,22 0,422 V2: customer inquiries 0,97 0,180 1,00 0 0,17 0,388 V3: customer comments 0,92 0,277 1,00 0 0,17 0,388 V4: downloading of software 0,11 0,321 0,44 0,512 0,04 0,209
V5: online problem diagnostics 0,49 0,504 0,13 0,342 0,22 0,422
V6: order status checking 0 0 0 0 0 0
V7: site survey 0,13 0,340 0,13 0,342 0 0
V8: product survey 0,26 0,444 0,44 0,512 0,09 0,288
V9: new product proposal 0,08 0,277 0,19 0,403 0 0
V10: key word search 0,18 0,388 0,19 0,403 0,09 0,288
V11: personal choice helper 0,70 0,460 0,13 0,342 0,13 0,344
V12: virtual-reality presentation 0,08 0,277 0 0 0,04 0,209
V13: dealer locator 0,08 0,277 0 0 0,17 0,388
V14: electronic coupon 0,05 0,218 0,25 0,447 0,09 0,288
V15: online ordering possibilities 0,15 0,358 0,19 0,403 0,09 0,288
V16: online contest/ prizes 0,11 0,321 0,87 0,342 0,39 0,499
V17: push media (newsletter) 0,49 0,504 0,81 0,403 0,30 0,470
V18: interactive job placement 0,11 0,321 0,56 0,512 0,09 0,288
V19: electronic post card 0,13 0,340 0,50 0,516 0,22 0,422
V20: surfer postings 0,02 0,128 0,13 0,342 0 0 V21: user groups 0,03 0,180 0,19 0,403 0 0 V22: games 0,08 0,277 0,87 0,342 0,17 0,388 V23: multimedia presentations 0,31 0,467 0,56 0,512 0,35 0,487 V24: online two-way communication 0 0 0,06 0,240 0 0 V25: navigation flexibility 0,93 0,250 0,87 0,342 0,91 0,288 V26: informational content 0,98 0,128 0,81 0,403 0,70 0,470 V27: credible content 1,00 0 0,94 0,250 0,83 0,388 Totaal 9,41 2,020 12,25 2,978 5,48 2,020
shows a higher motivation to have interaction by providing the most newsletters (v17).
Design functions (last 3 variables) show that cluster 2 is again most attractive for visitors.
So, based on the information in the tables (cluster 1, 2 and 3) a description can be formulated for each cluster. Also, this can be some sort of guidance tool to place certain brand websites in. It can be suggested that all brand websites fall within one of these three clusters. Based on that assumption, the attractiveness of certain websites can be assessed and points for improvement (in interactivity or website design) can be recommended. Table 5 gives an illustration on how each cluster could be described based on the cluster’s score.
Table 5: Cluster names
Cluster Name Characteristics (all companies: see Appendix 10)
Cluster 1 Clear choice helpers • high on personal choice helper
• high score on website design
• Most companies are in Personal care, like: Oral-B, Schwarzkopf & Dove
Cluster 2 Brand advertisers • high customer support
• high on marketing research • low on personal choice helper
• good score on advertising, promotion & publicity • focus on consumer communication
• Most companies are in F&B, like: Pepsi, DrOetker & Campina Cluster 3 Information providers • low customer support
• low on customer research • low on personal choice helper • lower score on website design • Most companies are in F&B, like:
6.4.1 Results of brand websites
6.5 Results of alternative insights
The division of bigger and smaller brands is put in a table. The bigger brands are most present in cluster 1 (Clear choice helpers), but so are the smaller brands. The bigger brands are least present in cluster 3 (Information providers). When we look at the small brands, a comparable division has taken place. Here, also most brands are in cluster 1, except for the smaller group here is cluster 2 (Brand advertisers). Overall, the big and small brand are significantly different. The big brand’s websites are significantly more interactive than the small brand’s websites. Especially the factors: online comments, site survey, new product proposal, key word search, electronic coupon, interactive job placement and games are more present on bigger brand’s websites. The variable `dealer locator` is significantly more present on smaller brand’s websites.
Table 6:Big and small brands
Percentage
Big brands (in top
100) Cluster 1 63,89
Cluster 2 22,22
Cluster 3 13,89
Small brands Cluster 1 59,38
Cluster 2 12,50
Cluster 3 28,13
The overview in table 7 displays a majority of smaller brands in cluster 1. Cluster 2 has exactly the same amount of big and small brands. Cluster 3 has a majority of small websites. The relatively smallest amount of big brands is in Cluster 3. See appendix 13 for more information about differences between `big` and `small` brands.
Table 7: Brands per cluster
Percentage
Cluster 1 Big 37,70
Clear choice helpers Small 62,30
Cluster 2 Big 50,00
Brand advertisers Small 50,00
Cluster 3 Big 21,74
6.5.1 Hedonic and utilitarian brands
The most Utilitarian brands are in cluster 1. The smallest group was cluster 2 among the Utilitarian brands. Information providers (cluster 3) scored 22,73% when it comes to Utilitarian brands. The biggest group of Hedonic brands is in cluster 1. Cluster 2 and 3 are equally divided when it comes to the share of Hedonic brands. Both dimensions (Utilitarian and Hedonic) have the same share (average of 76%) in clusters 1 and 2. No significant overall difference between Utilitarian and Hedonic brands is present if the total average of interactivity is considered. Although, there are some variables where there is a significant difference, these are: downloading of software, key word search, electronic coupon , online contest/ prize, interactive job placement. These variables were all more present on Hedonic brand’s websites. On the other hand: online problem diagnostics, personal choice helper, virtual-reality presentation were all more present on Utilitarian brand’s websites.
Table 8: division of brands for each brand dimension
Percentage
Utilitarian Cluster 1: Clear choice helpers 70,45
Cluster 2: Brand advertisers 6,82
Cluster 3: Information providers 22,73
Hedonistic Cluster 1 53,57
Cluster 2 23,21
Cluster 3 23,21
Table 9: Presence of Brand dimension for each cluster
Percentage Cluster 1 Utilitarian 50,82 Hedonistic 49,18 Cluster 2 Utilitarian 18,75 Hedonistic 81,25 Cluster 3 Utilitarian 43,48 Hedonistic 56,52
7. Conclusions and recommendations
When all results are taken into account then it becomes clear that the current FMCG companies are using their brand websites in 3 different ways. First, there are the companies that we can call the `Clear choice helpers` (medium interactive). These companies focus on providing the consumer with a lot of product information and are stating clear product attributes on their websites. This way a consumer knows the advantages that the product can deliver to him, a product need is almost being created on these websites. Next, there are the `Brand advertisers` (most interactive) websites. These websites focus on creating a customer relationship and are by far the most interactive. These websites provide fun & entertainment to create brand awareness. Finally, there are the `Information providers` (less interactive) websites. These websites provide the consumer with product and company information but on a factual basis. There is little interaction possible and there is no clear consumer focus on these websites. The recommendations that are given are based on the perception that website interaction is desired and useful. Off course, some companies will have no interest in improving this aspect of their website, or will not believe that this works for the company. This report delivers recommendations based on literature and findings in the research, thus generalizing some of the recommendations. A company’s website objectives should always be kept in mind when applying these recommendations.
7.1 Current use of website functions
To be able to see to what extent Dutch FMCG companies use the website functions to interact with their consumers, a closer look is taken to their way of using websites. By tracking how they use the different functionalities the websites are evaluated. It shows that customer support is higher in the Household Care segment. Although the difference is relatively low. One could say that maybe more customer support (and factual data) is needed with these products. Because household products have a negative motivator for purchase, people need to be convinced with facts and product attributes. It needs to be said that `customer support` is second highest in score, so probably companies do have a strong interest in informing the customer and providing possibilities for feedback. The
point. Probably companies is this segment value customer feedback the most. Ideas for new products are most welcome in this segment. The measurement `personal choice
helper` scores highest amongst Household Care, other 2 segments score lower. Probably,
the reason for this is the amount of products and the distinct functionalities they have. Certain customer needs can be manipulated to stimulate product purchase.
Advertisement, promotion and publicity is above all segments used in F&B. So,
special promotional activities are undertaken in the F&B segment more than the other 2 segments. This could be because food & beverage brands are harder to distinguish than household products and personal care items, so here a more aggressive marketing attitude is being displayed, to attain a place in the consumer’s mind. The entertainment level is highest among the F&B sites. The products in this segment are mostly brought to consumers to persuade through the `peripheral route`, meaning that little to no factual information can be used to market these products. Therefore, these brands stress the
enjoyment which is associated with the specific brand (Chattopadhyay, 1998).
Communication extra is only found once in this research in the F&B segment. This
7.2 Discussion of results
In general, FMCG websites can do a lot more to interact with the consumer. As can be seen in the results not all websites provide a feedback form. Although this could be a valuable tool to provide companies with consumer information. This function could also be seen as a hygiene factor, meaning that the absence of it could create visitors’ dissatisfaction. Companies should make it very clear to themselves what the hygiene factors are. This makes it a lot easier to create a website that relates to the customers’ desires. When all functions are taken into account the ones that are used the least are:
Online two-way communication, although this provides instant answer to whatever
questions visitors may have, this function is almost not used at all. Perhaps this is due to the technical restrictions there are for such websites or due to time pressure. Next, Order
status checking is also not used a service tool towards customers. Obviously, this
outcome is related to the fact that most brand websites have nothing to order, but still there are a few that offer merchandising. Then order status checking could be a handy tool to prevent time consuming customer care activities. Finally, surfer postings are also not used very often. Although these postings could create a forum for brand consumers that have specific remarks or questions, not many companies offer this function on their website. In general FMCG websites show that they are able to provide the consumers with product information and that some companies offer extra functionalities, but still more can be done to give the consumer more opportunities to interact with the brand. These examples show just how easily companies could provide visitors with functions that are very helpful or pleasant to them. This will have a positive effect on the way they perceive the brand. And perhaps next time they go shopping they will choose this brand because they were satisfied with the brand’s website. What is already being done a lot is Customer support. And within this specific function the following issues are included:
Online form for customer feedback, Online form for customer inquiries and Online form for customer Comments. In the research these points are discussed separately, sometimes
