A quality function deployment approach to web site development: Applications for electronic retailing, G Ioannou

Tags: House of Quality, design attributes, retail store, attributes, requirements, site features, navigation features, Nielsen, Electronic Commerce, Quality Function Deployment, International Journal of Electronic Commerce, World Wide Web, Internet retail stores, Van Nostrand Reinhold, Web Development, Web Outsourcing, Spiller, L. L. Schkade, Journal of Retailing, design possibilities, Web audiences, Web audience, QFD methodology, web site design, Web Site Development, G. Van Treeck, Visualization & Usability Center, Navigation, customer requirements, store traffic, retail site, customer requirement, Internet Advertising Bureau, Software Development, Software Product Development, Journal of Computer Mediated Communications, Forrester Research, Journal of Computer Mediated Communication, Software Technology, Georgia Tech Research Corporation, Internet retail store
Content: A Quality Function Deployment Approach to Web Site Development: Applications for Electronic Retailing
George Ioannou (1) [email protected]
Katherine C. Pramataris (2) [email protected]
Gregory P. Prastacos (1) [email protected]
(1) Decision Sciences Graduate Program (2) eLTRUN (Electronic Trading Research Unit) Athens University of Economics & Business 47A Evelpidon & 33 Lefkados Str., 113 62, Athens, Greece http://www.decision.aueb.gr, http://www.heltrun.aueb.gr
Abstract While the development, maintenance and promotion of a web site requires a high investment, the value and end-benefits associated with it are not always materialised. The reasons for this mismatch are in most cases associated with the difficulty in capturing web-site customer requirements and translating them into specific design attributes and characteristics of the site. Issues of web-site design and technology usage become even more critical as businesses attempt to exploit the boom in electronic commerce and marketing over Internet. Now more than ever, the promise of electronic commerce and online shopping will depend to a great extent upon the interface of the customer and the on-line media, i.e., upon how people interact with the computer. In this paper we address the issue of capturing and integrating "the voice of the customer/user" into the development process of a web site, by applying the Quality Function Deployment (QFD) methodology, which has been used very successfully in other industry sectors. QFD is a structured approach for product design that integrates customer requirements into a tree of interrelated coupling functions that can proceed down to the controllable variables of a system. We propose the adoption of QFD to web site design, development and dynamic adjustment by considering a site as a typical product with attributes, production characteristics and measurable controls that can reveal and trace the level of satisfaction of customer needs. Furthermore, we demonstrate how QFD can be used to guide design decisions for web sites, using electronic retail sites as a case field.
Keywords:
Web site design & development, Quality Function Deployment (QFD) methodology, (Electronic) e-Retailing, Internet Retail Store Design
1. Introduction: Current Issues and Trends in Web Site Development In recent years, we have witnessed a tremendous growth of the Internet and the World Wide Web. There are currently approximately 3.6 million World Wide Web (WWW or Web in short) sites on the Internet, up from 250,000 in mid-1996 (e-land, 1999). Business enterprises -from multinational conglomerates to solo entrepreneurs- are staking their presence in the Internet, all poised to become pioneers in what promises to be the frontier of electronic commerce (Kalakota and Whinston, 1996). Yet, in spite of estimates that Internet users world-wide are close to 195 million (NUA, 1999), on-line business is still at its infancy. Apart from the obvious difficulties with bandwidth and security, and technical issues that could eventually be resolved, there is the more probing question of what value is being created by information technology in general, and on the
Web in particular. Certainly, one cannot expect real progress if it is simply the digital replacement of conventional channels such as newspaper ads, TV commercials, phones, and fax (Ho, 1997). On the other hand, the costs associated with hosting a Web site vary greatly depending on the scope of the site itself and the management process of setting it up and maintaining it throughout its lifecycle. The annual operating costs of internally managed Web sites range from $200,000 to $3 million. In addition to the running cost of Web sites, the cost of designing the site content can be steep. According to a Forrester Research report (Forrester Research, 1997), the expenditure on third-party Web site designers is expected to reach $10 billion in 2000, from $582 million in 1996. The same report states that the average cost of developing the content of a Web site is $267,000. Last but not least, the cost of promoting a web site may be equally or even higher to the above, with just online advertising revenues estimated at $693 million for the first quarter of 1999 (Internet Advertising Bureau, 1999). The above figures indicate that while the development, maintenance and promotion of a web site requires a high investment, the value and end-benefits associated with it may not be clear up-front. In other words, Web site development projects, as most Information Technology (IT) projects, incorporate a high-risk parameter. IT professionals have attempted to tackle this issue and reduce the risk associated with software development by adopting appropriate design, development and deployment methodologies. The basic objective of these methodologies is to capture user requirements as fully as possible and then to deliver a system that meets the requirements to the greatest extent. Understanding the users' work domain has been a consistent source of interest and challenge for human-computer interaction designers. Consequently, much of the research to-date has addressed different approaches for achieving this understanding. In the 1980's, user-centred design focused on the need for designers to base their rationales on an understanding of the work domain to be supported. Since the early 1990's, work in cooperative design, user-centred design and participatory design has emphasized the role of user-participants in treating work context as an empirical source of design rationale and has addressed some of the user work domain problems facing Web designers, principally the problems of incorporating workplace expertise into the design process. Other research efforts on Web design development refer to the notion of usability, and many researchers have enumerated some of the difficulties of incorporating usability into the design of web sites (Vora, 1997). Roh et al. (1997) suggested a specific process for initiating, developing, and launching web pages that explicitly incorporates usability techniques at several points in the development cycle. In this paper we address the issue of capturing and integrating "the voice of the customer/user" into the development process from a considerably different perspective, that of treating a Web site as a typical product, i.e., a product with certain attributes that should meet specific user needs. We apply the expertise assimilated in the field of classical product engineering and design, and more specifically the steps of the Quality Function Deployment (QFD) methodology. QFD is used with success in the several industry sectors since the 1970s. In the software development industry, QFD has been employed as a methodology for general software design and development (Erikkson, 1993; Thackery & Treeck, 1990). Haag et al. (1996), discuss the adaptation and use of what they call Software Quality Function Deployment (SQFD) for software development by major software vendor firms, such as DEC, AT&T, Hewlett-Packard, and Texas Instruments. They define SQFD as a front-end requirements solicitation technique, adaptable to any software engineering methodology that quantifiably solicits and defines critical customer requirements. Lamia (1995) applied the QFD proposition to Object Oriented (OO) software design and analysis, showing how QFD can be used for the initial information collection, providing easy-to-understand structuring tools that do not require extensive training in OO concepts and methods. In this paper, we make a first attempt to explore the usage of QFD in the process of Web site development. 2
The remainder of the document is organized as follows: Section 2 outlines the major principles and mechanics of the QFD methodology and highlights the key components of QFD. Section 3 presents our approach in applying QFD to Web site development and, in particular, the structured identification of specific product attributes coupled with market requirements that are important in this environment. Section 4 takes this process further through the application of QFD for electronic retail sites. The paper concludes in section five, where future research opportunities are also put forth. 2. The QFD Methodology QFD is a structured method used to identify and prioritise customer requirements, and to translate these requirements into engineering specifications for systematic deployment throughout a company at each stage of product or process development and improvement. The concept of QFD was introduced in Japan by Yoji Akao in 1966, and was first applied at the Kobe Shipyard of Mitsubishi Heavy Industries Ltd. Since that time, QFD has become the accepted methodology for development of products and services in Japan and has enabled businesses to successfully develop and introduce products in a fraction of the time required without it (Jackson & Frigon, 1994). In the early 1980s, QFD was introduced at Xerox, and since then American businesses have exhibited substantially growing interest in using it. According to Akao (1990), QFD "is a method for developing a design quality aimed at satisfying the consumer and then translating the consumer's demand into design targets and major quality assurance points to be used throughout the production phase. ... [QFD] is a way to assure the design quality while the product is still in the design stage." As a very important side benefit, Akao (1990) points out that, when appropriately applied, QFD has demonstrated the reduction of development time by one-half to one-third. QFD is a structured process, a visual language, and a set of interlinked engineering and management charts. According to Jackson & Frigon (1994), QFD uses the following seven management tools: affinity diagram, interrelationship digraph, tree diagram, matrix diagram, process decision program chart, arrow diagram, and matrix data analysis chart. It establishes customer value using the voice of the customer and transforms that value to design, production, and manufacturing process characteristics. The result is a systems engineering process that prioritises and links the product development cycle so that it assures product quality as defined by the customer/user. Additional power derives from use within a concurrent engineering environment. QFD, as applied under the kaizen philosophy and within total quality control, is by far the most highly developed form of integrated product and process development in existence (Dean, 1992). There are several approaches to QFD; each of these approaches makes use of matrices to organise and relate pieces of data to each other. Many times these matrices are combined to form a basic tool of QFD called the "House of Quality" (Hauser and Clausing, 1988) shown in Figure 1. The House of Quality starts with a "What-How" Matrix that identifies the wants, desires, and needs of the customer. These customer requirements are shown on the left part of the House of Quality. The ceiling of the House, i.e., the horizontal axis below the roof, shows the design or technical requirements, while the body of the House visually displays the relationships between the customer requirements and design requirements. The symbols used in each box, if any, show whether the relationship between the two corresponding elements is strong or week, positive or negative. This way the House of Quality quickly reveal patterns and identifies weak points in the design requirements. The Interaction Matrix, also known as the Correlation Matrix, is the "roof" of the House of Quality. It is established to determine the technical interrelationships among the design requirements. This information is valuable as the basis for decisions regarding technical trade-offs (Jackson & Frigon, 1994). It is important to note that the results of QFD are only as good as the data that go into the house. The data come from many sources, including focus groups, telephone and questionnaire surveys, experimental tests of competitor's products, and published materials. The important consideration 3
in design is to distinguish, in the customer's point of view, the critical few aspects of the product from the trivial many and to make sure that these critical ones are done right. The basic House of Quality has been further expanded to include additional information, such as comparisons to competitors, technical assessments and target values. Figure 1: The House of Quality The House of Quality is the first of many matrices in the QFD process of translating customer needs into product requirements and system specifications. The number of matrices corresponds to the number of phases in the development process. For example, the American Supplier Institute (ASI) follows a four-phase approach, in which a QFD team deploys customer requirements into product characteristics, product characteristics into part characteristics, part characteristics into process characteristics, and finally process characteristics into production characteristics. Other approaches may include more phases, from customer requirements to design requirements, Engineering Design, product characteristics, manufacturing/ purchasing operations, and finally production/ quality controls. King (1989) and Clausing & Pugh (1991) provide more extensive and complete views than the simple "House of Quality" view. Dean (1992) views QFD as a system engineering process which transforms the desires of the customer/user into the language required, at all project levels, to implement a product. It also provides the glue necessary, at all project levels, to tie it all together and to manage it. Mizuno and Akao (1994) indicate that QFD is far more than had previously been disclosed, i.e., the mechanism for deploying quality, reliability, cost, and technology throughout the product, the project to bring forth the product, and the enterprise as a whole. Despite the many views that exist around the application of QFD, in this paper we confine ourselves to the basic House of Quality, in our attempt to apply QFD to Web site development. This, as the purpose here is not to make a detailed analysis of how QFD could be used in this new 4
context, but to give an overview of the potential that this structured approach has in an unexplored engineering environment, that of web site design. In the following section we discuss the main steps of applying QFD to the development of a web site and define the basic House of Quality. 3. Applying QFD to Web Site Development Gazing back over the history of the Web, one can clearly see distinct evolutionary stages of Web development, particularly with regard to corporate Web sites (Hewitt, 1998): 1. In the earliest stage, development was clearly the highest priority. At that time, simply having a Web presence was enough. 2. In the next stage, "artistic" design became a top priority. It wasn't enough to simply have a Web site; that Web site had to be attractive. People began experimenting with design possibilities for site navigation and decoration. Companies designed elaborate sites that included animation and other effects. However, often they paid little attention to download times. 3. Next, the content stage took hold. Many companies moved to answer their Web audience's seemingly insatiable need for information. They rushed to post newsletters, product information, and annual reports. 4. In the current stage of Web evolution, companies have begun to see that Web content requires context. Web sites have grown and evolved substantially, and Web audiences have become more diverse and sophisticated. Consequently, today's corporate Web sites require the same kind of editorial guidance common in the "primitive" world of print publishing. In this evolutionary course of Web development and design, we propose that the application of QFD methodology can satisfy the increasing need for high quality, context-centric sites. The first step in applying QFD to the development of a Web site is to identify the "What's" and "How's" that will form the basis of the House of Quality. The "What's", also referred to as customer needs or requirements, is a list of what the customer wants or what is to be achieved. In the case of a Web site, these would be user statements about what they want the Web site to look like and offer, i.e., what are the desirable site features. On the other hand, the "How's", also referred to as design or technical requirements/ attributes, is a list of what one can measure and control in order to ensure that the customers' requirements will be satisfied. Typically, the entries on this list are parameters for which a means of measurement and a measurable target value can be established. Again, in the case of a Web site, these would be the specific site characteristics, visible or not, that satisfy the specific user needs. The range of Web site objectives can be anything from pure advertising of a company's or organization's name, to implementing customer support or online sales. It is therefore, becoming increasingly important for companies to be able to successfully attract potential customers to their own Web sites. How does a firm hope to attract target customers or compete for them especially among sites of the same nature? With the explosion in the number of Web sites, this has become a challenging question. Just as manufacturers of consumer products would like to see their products ranked high by Consumer Reports, firms would like their sites to be included among the attractive ones in such a list. 3.1 QFD "What's" 5
The factors that might be influential in improving the quality of Web sites, i.e., what does a user look for in a site that will make him/her visit and re-visit the site, are exactly the "what's" of QFD. Below, we state some of the basic "what's" that we have identified: · Availability: One of the key reasons mentioned by users for using Internet services is the 24x7 service availability. This means that users expect that any site on the web is up-and-running whenever they want to access it, 24 hours a day, 7 days a week. Availability also implies that users don't get any error messages, or transfer interrupt messages due to site server overload. · Functionality: An equally important "what" to site availability, is the ease of using a site for both expert and naive users. Functionality is often called "the key to a successful web site." It is what helps users navigate through the site, use the site features, and find the information they are looking for. Achieving true functionality requires balancing attractive graphic design, simple navigation solutions and good content that's worth revisiting. A functional web site is one where designers anticipate the needs of users, and where users can easily find what they need. · Integration of information: Even if a site is always available and easy to use, it is worthless for the user if it has no substance, i.e., it contains non-value-adding information, and/or the information is not coherent or up-to-date, it is partial and incomplete, or it is not valid. A characteristic also related to this requirement is whether a site exploits the hypertext properties of the Web and contains links to other relevant sites and sources of information. In the early years, the Web was more of an experimental environment and users understood that they would be limited to a sample of a company's services. Now, users expect comprehensive service from sites, which not only do they need to be up and available at all times, but they also need to have all the information and services users want in a certain category (Nielsen, 1999). · Complete services: Ideally for a user, a site should act as a `one-stop-shop' for a specific information topic, product or service; i.e., a user interested in a topic, product or service should access one specific site where he/she could find all the relevant information, would get value added services, preferably matching his/her profile, could purchase the product or service, could get after-sales support etc. For example, some researchers have referred to the need and competitive advantage of sites supporting the whole buying process for a consumer, on-line (Vrechopoulos et al., 1999). · "Easy" to find site: A site that has the asset of acting as a `one-stop-shop' should ensure that users can easily find it and can, thus, appreciate the service that is offered to them. The first attribute that supports the easy access to a site is its URL, i.e., its www address. The relevance between the address and the site's content, the address's ease of remembering, the previous existence of a well-known brand name, etc. can enhance the reachability to a site. Another attribute associated with this requirement refers to the use of keywords and `meta-tags' that facilitate a site's identification and inclusion in the catalogues of popular search engines. Last but not least, we need to mention that, probably the most effective practice of guiding traffic to a site is through advertisement, either on-line, on the web, or off-line, via traditional media. · Responsiveness: Users also require that sites respond to their specific needs, a requirement usually translated into customisation features and personalized services. For example, a benefit that users get from on-line publishers is that they have the possibility to customise a newspaper's contents to their own profile and preferences. In addition, the personalized service offered by Amazon.com on book recommendations, without imposing any extra work to the users, has greatly contributed to the high popularity that this site has gained (Nielsen, 1999). Equally, if not more, important to the type of response is also the speed of response to a user's interactive action or request. This may refer to the speed of responding to a user's e-mail request, returning search or query results, downloading the next web page requested, etc. · Character: The site's character refers to the impression created by the site's welcome page, the look and feel across pages, the proportion between text, animation and graphics, the site's structure, etc. This is an element that also has an impact on a site's quality. Multimedia is 6
gaining popularity on the Web with several technologies to support use of animation, video, and audio to supplement the traditional media of text and images. These new media provide more design options but also require design discipline. Unconstrained use of multimedia results in User Interfaces that confuse users and make it harder for them to understand the information. Care should be taken to provide a "high-touch" feeling in addition to the "high-tech" image of a WWW site (Nielsen, 1999). The weight that content has versus image and character depends on the purpose of a site, i.e. to entertain, sell, communicate, but also on the site's target audience, e.g. teenagers, businessmen, housewives etc. In addition, research has shown that users like the feeling of being part of a twoway communication with a site staffed by real humans and not just the recipients of a stream of bytes coming in over the net. 3.2 QFD "How's" The degree to which a site satisfies the above user requirements, i.e., the "what's" of QFD, depends on specific site characteristics. We have identified the following key characteristics universally relevant to web sites: · Performance: The term performance refers to properties such as the time it takes for the site to download, the response time experienced by users requesting a specific site service, the number, type and frequency of error messages popping-up, etc. As such, this attribute is associated both with a site's availability but also with a site's responsiveness. The technical characteristics that affect performance include: the server technology used for hosting the site, the server capacity, the site's underlying architecture, the size of the site's individual pages, the telecommunication lines capacity etc. Research on a wide variety of hypertext systems has shown that users need response times of less than one second when moving from one page to another if they are to navigate freely through an information space. Traditional human factors research also shows the need for response times faster than a second. The user's experienced response time is determined by the weakest link in the chain from server to browser (Nielsen, 1999), i.e.: ° The throughput of the server ° The server's connection to the Internet ° The Internet itself ° The user's connection to the Internet ° The rendering speed of the user's browser and computer Considering these fundamental facts in both human factors and computer networking, there is only one conclusion: Web pages have to be designed with speed in mind. To keep page sizes small, graphics should be kept to a minimum and multimedia effects should only be used when they truly add to the user's understanding of the information. · Name: As mentioned above, a web site's name and corresponding web address (URL) can greatly facilitate a user finding the site, even without using a search engine. There are several guidelines that have been published on naming and branding, including guidelines on web site addresses (Trout, 1997). · Welcome page: The welcome or home page of a site is the one creating the first impression to the user, the one that will grasp his attention and make him dig further into the site, or the one that will make him press the `back' button immediately. In addition, the way the home page is structured may greatly facilitate the user's navigation through the site. A firm rule for home page design is `more is less': the more buttons and options you put on the home page, the less users are capable of quickly finding the information they need. However, we must keep in mind that on the Web, the user fundamentally controls his or her navigation through the pages. Users can take paths that were never intended by the designer: for example, they can jump 7
straight into the guts of a site from a search engine without ever going through the home page. Users also control their own bookmark menu and can use it to create a customized interface to a site (Nielsen, 1999). · Structure: No matter how creative, informative and fast a site is, it is useless if the user cannot find what he is looking for. Under this perspective, it is important that a site has a clear and consistent structure, organized in the way users expect to find it, and not corresponding to the way marketers or designers have it in their mind. A usability testing of two different navigation schemes on the same site rendered an 80% success rate when people used the navigation scheme structured according to most users' mental model. The success rate dropped to 9% when people used the navigation scheme structured according to the company's internal thinking (Nielsen, 1999). The difference in usability is not due to differences in surface design but to differences in deep design: finding out how to best match a Web design to the users' needs and how to best structure the information architecture. In addition, explicit recognition of the site as a structuring mechanism is important for Web usability. Much information can be hierarchically organized, and an explicit representation of the hierarchy can be added to the top of the page to provide additional context and navigation options (Nielsen, 1999). · Content: The expression "content is king" has often been used in order to denote the importance of a Web site's content from the user's standpoint. Equally important is the process of "keeping it fresh" and ensuring the integrity and validity of the information. Other attributes associated with content are the depth and breadth of the available information. Depth is related to the structure of the site, e.g., the number of levels in a hierarchical information structure, as well as with the number of links to other relevant subjects and sites. Breadth has to do with the extent, to which the site can act as a "one-stop-shop", i.e., satisfy all the information needs of a user around a specific topic or area of interest. Another important element of content is the way it is expressed in writing. Most people absorb information from Web sites differently than they do from printed materials; on the Web, people tend to browse or scan information, rather than really "read" it the way you would read a book or a magazine article (Nielsen, 1999). In addition, the Web offers site designers many options for presenting diverse types of information, as well as various levels of detail. Thus, on a Web site, information must be presented in a manner that that exploits the medium's possibilities and respects its limits (Hewitt, 1998). · Usability: In the Web context, the term usability (Spool et al., 1998) covers aspects that influence a site's ease of use, navigability, consistency of functionality across pages (Nielsen, 1999) etc. This attribute is highly valued by users, as it ensures that they can actually use the "product". Given the great number of web sites that a user may visit even daily, the demand for effective usability is higher for WWW user interfaces than for normal user interfaces (Nielsen, 1999). In addition, a usable site is one where the user can easily find what he is looking for, nomatter how deep and broad the content is, and is supported by the existence of an efficient search mechanism, although we cannot simply rely on search as the main navigation feature. Navigational structure and overviews are necessary to avoid user confusion and should be provided both in the large (server structure and location) and in the small (structure for the individual pages with iconic markers for the various types of information) (Nielsen, 1999). The existence of a contents page and a site map can further improve navigation and search usability. · Aesthetics: One of the factors that influence a user's response to a site is the first impression that will be created upon visiting the site, usually the site's home page. This impression depends on the overall look-and-feel of the site, the colours, the graphics used, the analogy between text and images etc., or in one term, the site's aesthetics. There is a lot of discussion around the competitive "relationship" between graphics that improve aesthetics and site performance, bandwidth limitations and download times. Conservative use of graphics does not imply boring pages. Colour table cells and creative (but restrained) use of different fonts can add significant value. In particular, style sheets can be used to improve page design without 8
incurring a download penalty (Nielsen, 1999). In addition, simplicity is the essence of good Web design and can greatly improve a site's usability. · Services offered: The previously described site characteristics are not always enough to attract and keep users at a site. As Web development advances, users become more and more demanding regarding the attributes they would expect to find at a site. The additional features they look for include: - Interactivity, defined as the facility for individuals and organizations to communicate directly with one another regardless of time and space. The level of interactivity of a site would be critical in converting site visitors from interested contacts into interactive customers. - Personalization of content and services, i.e., dynamic adaptation of a site's content and structure to match the profile of each user (Pramataris et al., 2000). This can be based on user demographics or other information provided by the user, such as interests, past behaviour, etc. A good example of this feature is the Amazon electronic shop (www.amazon.com). - Customisation defined as the possibility for users to explicitly select between certain options, depending on their preferences and interests at the time. MyYahoo! is a good example of this possibility, as are the web sites of many on-line newspapers. - On-line customer service that refers to offering customer support apart from a site's basic informative role. It can include services such as providing the possibility for on-line form completion, on-line sales, direct question-answer session etc. The importance of most of the attributes discussed above is underlined by the results of the 10th WWW User Survey, conducted by the Graphic, Visualization & Usability Center (GVU) of Georgia Tech Research Corporation (GVU, 1998). According to this survey, the main problems mentioned by users of the Web include: Speed, slow advertisements, broken web links, graphics, registration, difficulty in finding new information, paying for content and script errors. 3.3 The web site House of Quality Figure 2 depicts our first attempt to create the basic House of Quality for a web site, illustrating the interrelationships between the customer requirements and the web site characteristics discussed in Sections 3.1 and 3.2. To fill out the various relational weights between QFD what's and how's, we have used the points raised in the discussion of customer requirements and design attributes of the previous two sub-sections. For example, a customer/user requires high site availability and thus, this requirement (Availability) is characterized by a high (strong) customer priority. Furthermore, the site's availability is directly linked to the site's performance, thus imposing a strong relationship between this customer requirement (Availability) and the corresponding design attribute (Performance). Finally, the design attributes Content and Services Offered are linked with negatively weak relationships to the Availability customer requirement, since the elaboration (additional effort) on these two design attributes negatively affects (to a certain extent) the site' availability. For the customer requirement Availability, no other strong, medium, weak or negative relationships result from the discussion of Sections 3.1 and 3.2.
~ {
~
{
~
~
{{ {~{~
9
~ Strong = 9 { Medium = 3 Weak = 1 Negatively Weak = -1 z Neg. Medium = -3 ~ Neg. Strong = -9
Design Attributes
Customer Priority Performance Name / Address Welcome Page Structure Content Usability Nice look & feel / aesthetics Services offered
Customer Requirements
Availability Functionality
~~ ~
{~
~{
Integration of Information ~
Complete Services
{
{~
{
{
~
Responsiveness
{~
~{z{
Easy-to-find Site Character
~ ~
{{~
RISK ~
{{~~{{
Weights
Absolute 19 9 Relative 171 9
13 12 23 16 9 14 39 36 207 144 27 42
KEY ELEMENTS X
XX
Figure 2: Web-Site House of Quality­ 1st phase Following a similar reasoning, we proceed to the completion of all interrelationships between customer requirements and design attributes. It is important to note that we have used only research publications and prior research results to identify the levels-values of these interrelationships; however, in practical situations, one should resort to direct market research or expert opinions through structured methods such as questionnaires, Delphi, group meetings, etc. in order to derive accurate and meaningful relationships. As shown in Figure 2, the first step of the QFD process identified Performance, Content and Usability as the key site characteristics that must meet customer requirements. These key elements correspond to the highest relative weights (207, 144 and 171, respectively), which are a combination of the risk, or else importance for the customer, and the overall value of the symbols associated with each element. For example, the relative weight of Content is calculated as 9 multiplied by 23, where 9 stands for the strong risk this attribute has and 23 is the sum of all the symbols in that column (Negatively Weak + Strong + Medium + Strong + Medium=-1+9+3+9+3), i.e., a measure of the value that this attribute has for all the customer requirements. The implicit matrix of Figure 2 is the first of many matrices in the QFD process of translating customer needs into product requirements and system specifications. Following a three-phase QFD approach for the design and development of web sites, we adopt the following terminology for each specific phase: 1. Customer/ Visitor requirements: These are the requirements that a visitor or customer of a web site has towards the site. These are shown on the left side of the House of Quality.
10
2. Web-site characteristics: These are the specific site features that "satisfy" the customer requirements, corresponding to the topside of the presented House of Quality. 3. Design Attributes: These are design elements of the web site that are associated to the characteristics that have been defined during the previous phase. Each design attribute may be further analysed, in a hierarchical way, into more detailed attributes and this step may be repeated as many times as required in order to reach the desired level of detail. 4. Design Variables: The final phase links detailed design attributes to measurable features and parameters. For a web site these may correspond to specific variables that are monitored by the web server, such as number of site visitors, number of hits, etc., or to measurable design elements, such as number of web pages, size of web pages etc. This three-phase QFD approach adapted to the design and development of web sites is depicted in Figure 3. In the following section we will attempt to further refine the initial House of Quality of Figure 2 through this three-phase approach, by focusing our study on electronic retail web sites and considering only the design attributes Performance, Content and Usability that have far larger Relative Weights than the remaining design attributes. Customer/Visitor Requirements Web-Site Characteristics Design Attributes Design Variables Figure 3: Three-Phase QFD Approach to Web Site Development 4. Applying QFD to Electronic Retail Web-Sites The terms "electronic retail site", "e-shop", "online retail store" and the similar have emerged lately to describe web sites with e-commerce capabilities, i.e., sites where the visitor can get support throughout the buying process, from the need recognition phase, to product purchasing and postpurchase support (Vrechopoulos et al., 1999). With projections that the Internet will generate consumer and business-to-business sales in excess of $294 billion by the year 2002 (Deck, 1998), and with surveys reporting that 75% of respondents in the United States already purchase online from home (e-Marketer, 1999), online retailing raises many questions about how to market on the Net. Issues of technology usage become critical as businesses and retailers attempt to exploit the boom in electronic commerce and marketing. There are large differences between a physical store and its electronic counter-part. A help button on the home page of the Web shopping site replaces the sales clerk's friendly advice and service. The familiar layout of the physical store becomes a maze of pull-down menus, product indices, and search features. Now more than ever, the promise of electronic commerce and online shopping will depend to a great extent upon the interface and how people interact with the computer (Lohse & Spiller, 1998). Despite the possibilities offered by 11
technology, Spiller and Lohse (1998), in their study of 137 Internet retail stores, show that there is still long way to go in order to meet consumer expectations regarding online shopping. In this section we make an attempt to identify the specific design attributes of electronic retail sites that have the greatest importance for the end customer. These are associated with the general website characteristics that have the greatest weight, as per the first QFD phase, i.e., the Performance, Content, and Usability characteristics of a web site. Looking at the pertinent literature, we can find several references to the characteristics of "real" shopping with extensions to electronic shopping. Lindquist (1974) categorised physical store components into functional areas such as merchandise selection, price, store policies and store layout. His attribute list is a compilation from 26 researchers in this field. Lohse & Spiller (1999) translated this list into electronic shopping attributes, categorising them into four groups: Merchandise, service, promotion, and convenience. Merchandise variables measure product selection, assortment, quality, guarantees, and pricing. Service variables examine general service in the store and sales clerk service for merchandise return, credit policies, etc. Promotion variables record sales, advertising, and appetiser features that attract customers (e.g., a "What's new" section). Navigation variables include store layout and organisation features. In a virtual retail store, merchandise variable, and more specifically the product assortment, affect all three characteristics of content, performance and usability. Obviously, the content of a web store refers mainly to the product categories and the individual items this store offers for selling. On the other hand, the greater the number of products and the amount of information maintained per product, the lower the system performance. Apart from the overall system performance, the number of product hierarchies and the number of products per web page also affect the time delays perceived by the user while waiting for product pages to download. This last feature is also associated closely with the usability of the e-store and the ease of finding a product within the hierarchical structure of categories and subcategories. The service features of an e-store are mainly associated, in a positive relation, with the content of the web site. However, too many service features, especially when these appear on most of the site pages, may result in information overload and greater page-download delays for the user. Examples are links to "Frequently Asked Questions", "Contact us", "Gift services", "Sales Representative", etc. appearing on all product pages or extra get-to-know the customer questions. Promotion features have, in most cases, a negative relation with the characteristics of performance and usability. Banner advertisements and other promotional features result in bigger page sizes and thus lower response times. The aesthetic need for advertisements to have high quality graphics, animation and eye-catching features deteriorate this situation even further. These same features, especially on banner advertisements, tend to confuse the user as they attract the eye away from the core page content and result in poor site usability (Nielsen, 1999). Regarding the value offered to the web store customer in terms of content, this may be both negative and positive. While many eshoppers have a negative attitude towards banner advertisements, others are indifferent or want to get informed about new products and discounts. At the same time they all seem to appreciate the "special offers" and the "what's new" section in a retail site (Lohse and Spiller, 1998). Shopping recommendations, when well targeted and meaningful, may also be perceived as a service by the eshopper (e.g., www.amazon.com). Navigation variables greatly affect the usability of the web store. Vrechopoulos et al. (2000) define three different web store types depending on the navigation links between different product categories, and associate these store types with sales performance and customer satisfaction measures. Navigation is also associated with the hierarchical structure of product categories and the amount of information appearing at each level. For example, if the price, size and brand of a product is shown at the products-listing page and the customer can purchase an item at this level, this will save him/her an additional `click', which is translated into greater usability. However, at 12
the same time the products-listing page becomes quite big, takes longer to download, while part of its content is only visible after the user scrolls down. We thus define a weak negative relation between navigation variables and performance and content. The latter negative relationship is also supported by the fact that the greater the amount of information (in our case products) at a site, the more difficult the navigation becomes. The combination of all the above result in the 2nd phase House of Quality depicted in Figure 4. For example, the characteristic Content is strongly related to the design attribute Merchandise Features since the content of an e-retail store is defined by the breadth and depth of its offerings to the endcustomer. On the other hand, Content has relatively medium correlation with the design attributes Service Features and Promotion Features because although both these attributes affect a retail store's Content, they are not the key factors in defining the essence of this characteristic. Finally, the design attribute Navigation Features is negatively related to the characteristic Content since the larger and the more information rich a retail site, the more complex its structure and the more difficult the navigation through the site by the end-customer.
~ Strong = 9 { Medium = 3 Weak = 1 Negatively Weak = -1 z Neg. Medium = -3 ~ Neg. Strong = -9
z { Retail Site Design Attributes
Customer Priorit y Merchandise Features Service Features Promotion Features Navigation Features
General Web Site Characteristics
Performance
~z
~
Content Usability
~~
{
{
~
~
~
RISK ~
{
~
Weights
Absolute 5 Relative 45
1
-15
7
3
-15
63
KEY ELEMENTS X
X
Figure 4: Web site House of Quality ­ 2nd phase for an electronic retail store The symbols associating retail site design attributes with web site characteristics in the centre of the House of Quality are justified by the discussion above. Customer priority has been set to high for all site characteristics, as these have been identified as the key elements during the previous QFD phase. The third type of input in the above House of Quality, the values of "RISK", are based on the findings of Lohse & Spiller (1999), whose study identified the following regarding the four categories of design attributes we examine here: · Merchandise: - Additional products in the store attract more traffic (17%). · Service: - Featuring a FAQ section in the store is associated with more traffic (54%). - Providing a feedback section for customers is associated with lower traffic and higher sales (1%). · Promotion:
13
- Appetiser information has no significant effect on traffic or sales. - The number of featured products along departmental navigation path has no significant effect. · Navigation: - Improving the browsing and navigation capabilities of stores and especially product lists can generate significantly higher traffic and sales (61%). - The number of levels between home page and end product pages has no significant effect on visits and sales. - Consistent menu bars have no significant effect in the models. For those variables above that are significant, the numbers in parentheses represent the variance in the store traffic or monthly sales explained by the corresponding variable. As Lohse and Spiller comment in their study, we should not expect a causality relation between the existence of a FAQ section in a store and sales or traffic. It is more probable that a FAQ section is added to cope with high traffic generated by the customer feedback section. We thus conclude that, based on the above study, merchandise and navigation features are the most important in driving web-store sales and traffic and are thus associated with higher risk. The calculations of the absolute and relative weights of the House of Quality of Figure 4 follow the method described for the same weights of Figure 3. Building further on the findings of Lohse & Spiller (1999) and the design variables they define, we can derive the 3rd phase House of Quality for an electronic retail store of Figure 5. On the left of the House of Figure 5, we place the key design attributes that we identified during the second phase of the QFD methodology that is Merchandise Features and Navigation Features. On the top of the House we present some of the specific design variables associated with them, as defined in the Lohse & Spiller (1999) study, i.e.: a. Number of Products in the Store, which is a measure of the breadth of the store's assortment. b. Number of Product Hierarchies, which is a measure of the store's organisation and depth structure. c. Lines of Product Information, which is a measure of the content provided to assist the buying process by the end customer. d. Different Shopping Modes that relates to the flexibility and adaptability of the retail store to the customer's habits and possible needs. e. Number of Links that is a measure of the navigational effectiveness of the retail store. f. Number of Products on End Product Page, which is a measure of the organisational hierarchy of the product offerings and relates to the "easiness" for the customer to identify what he/she would like to buy without resorting to several clicks and having to go through multiple overloaded pages with information concerning groups of products. g. Number of Scroll Listings that defines the compactness of information presentation and relates to the navigational effectiveness and usability of the site. h. Product Listing (Name, Description, Picture), which is a measure of the quality and "smartness" of the presentation of the products. Further discussion on the above variables can be found in the original work of Lohse and Spiller (1999). These design variables that are presented in Figure 5 on the "roof" of the House of Quality are now very concrete and measurable elements that have either a negative or positive, weak or strong relation with the key design attributes previously identified. In particular, the design attribute Merchandise is strongly related with the design variable Number of Products in the Store, since by definition this is the variable that characterises the product assortment. A strong relationship is also established between the design attribute Navigation and the site design variables Number of Product Hierarchies, Number of Links and Product Listing, as each of the latter 14
variables strongly affects the path followed by the customer within the site in his/hers effort to fill the shopping cart and complete the buying process.
~ Strong = 9 { Medium = 3 Weak = 1 Negatively Weak = -1 z Neg. Medium = -3 ~ Neg. Strong = -9
{
{{ ~
{{
{
{
~
~
~
~~
Retail Site Design Variables
Customer Priorit y Number of products in the store Number of product hierarchies Lines of product information Different shopping modes Number of links Number of products on end product page Number of scroll listings Product listing (name, description, picture)
Retail Site Design Attributes
Merchandise Navigation
{~{ ~{~
{
{
{
{~{{~
RISK
~
~~{~
Weights
Absolute 12 12 2 6 10 6 4 12
Relative 108 12 2 6 90 54 4 108
KEY ELEMENTS X
X
X
Figure 5: Web site House of Quality ­ 3rd phase for an electronic retail store
Using a similar approach and findings as we have in the previous phase of the QFD methodology, we distinguish the number of products in the store, the number of links in the site and the type of product listing as the key elements that influence merchandise and navigation features of an electronic retail store. These measurable variables should be constantly (on-line) monitored throughout the life cycle of the retail store in order to provide meaningful feedback to the store's decision-makers that are responsible for the store's offerings to the end-customer as well as for the gross-sales and the overall performance of the Internet shopping outlet. The input from customers concerning these variables is implicit, since it can be only retrieved by processing the number of customer visits, the number of products browsed and bought, the "location" of these products in the site, and the path followed by the customer in order to reach these products. These metrics are easily captured on-line and are stored at appropriate server files (site monitoring tools). Having completed the three-stage QFD methodology proposed in this paper for the case of an electronic retail store, we are now capable of defining the structured approach for embarking on such designs. Customers require Availability, Functionality and Integration of Information. The key elements in the list of site characteristics that should be addressed to handle the satisfaction of these customer requirements are Performance, Content and Usability. The design attributes affecting the key site characteristics for electronic retail stores are Merchandise features and Navigation features. These design attributes are, in turn, strongly coupled with the design variables Number of products in the store, Number of links in the site and Type of product listing. As a result, careful fine-tuning of these design variables and appropriate balance of their values and their relative contributions to the site's presence, structure, aesthetics and content, is necessary to address the key customer requirements. It is important to note that following the reverse QFD process, i.e., from the design variables to the
15
customer requirements, we can end up associating the critical measures related to the key elements in the list of design variables with the initial customer requirements regarding the services and characteristics of a web site and, more specifically, of an electronic retail site. This reverse QFD process can be the basis of a continuous improvement methodology, applied to web store dynamic redesign and customer profile customisation, following an automated update of the design variables based on the constant monitoring and control of the level of satisfaction of the customer requirements. 5. Conclusions and Further Research Clearly, the user interface is an essential link between the customer and the retail store in Webbased shopping environments. Although it is very important to make a good store design up-front, in order to attract and retain online customers, it is even more important to track the effectiveness and usability of this design and constantly monitor the degree to which it satisfies the basic customer requirements. More specifically, by understanding the effect that the virtual store interface elements have on the consumer buying behaviour, virtual retailers will be able to design their stores more effectively, satisfy their customers through innovative and personalised ways, and gain competitive advantage (Vrechopoulos et al., 2000). The QFD process, as presented above, gives electronic retailers and web-site designers a practical and handy tool to identify important design attributes and monitor the extent to which they contribute to the needs of the retail on-line store customers. QFD offers both a structured approach and a visual representation to easily distinguish the key elements that represent the high-risk and high-opportunity areas in a web site's design, as well as the relative importance and conflicting behaviour between alternative design attributes. In that sense, it guides web-site design decisions and supports the development of web sites, bringing in the field of information systems design methodologies the expertise gained in the Manufacturing Industry. However, contrary to the static nature of manufactured products, information systems and web sites in particular are a lot more flexible and dynamic in nature. This fact greatly increases the potential that QFD has if used in that context, as it allows for the online measurement of product (site) effectiveness and the dynamic adaptation of product (site) characteristics in order to best meet customer needs. The work presented in this paper is a first attempt to demonstrate the usage and potential of QFD methodology for the design, development and maintenance of web sites. There are still many aspects of this approach that need to be clarified and tested through the extended application of the process in practical cases. Among all, the area that probably exhibits the highest interest for future research is the backward association of measurable design attributes with the initial customer needs through an online feedback loop. Research is required in order to identify those web-site elements that can be dynamically adapted and incorporate the knowledge gained through such a feedback mechanism. The development of the appropriate software solutions and tools will ultimately make it possible to address dynamic web sites automatically-adapted to customer needs via a QFD methodology applied twice in a backwards-forward loop. References 1. Akao, Y., ed. (1990). Quality Function Deployment, Productivity Press, Cambridge MA. 2. Clausing, D. & S. Pugh (1991) Enhanced Quality Function Deployment. Design and Productivity International Conference, Honolulu HI, 6-8 Feb. 3. Dean, E. B. (1992) Quality Function Deployment for Large Systems. Proceedings of the 1992 International Engineering Management Conference, Eatontown, NJ, 25-28 October. 4. e-land (1999) Just How Big is the Web? The Web Characterization Project. Online Computer Library Center, Inc. (OCLC). http://www.emarketer.com/estats/092099_size.html 16
5. e-Marketer (1999) News from the world's largest Web Survey. eMarketer Newsletter 49 (Dec. 7 1999). Available at: http://www.emarketer.com/estats/113099_idc.html 6. Erikkson, I. (1993). Quality Function Deployment: A Tool to Improve Software Quality. Information and Software Technology, 35(9), 491-498. 7. Forrester Research (1997) Web Outsourcing to Reach $10 Billion by 2000. News release, 1997. http://www.forrester.com 8. GVU, 1998. WWW User Survey, conducted by the Graphic, Visualization & Usability Center (GVU) of Georgia Tech Research Corporation, from October 10, 1998 through December 15, 1998. Available at: http://www.gvu.gatech.edu/user_surveys/survey-1998-10/ 9. Haag, S., M. K. Raja, and L. L. Schkade (1996) Quality Function Deployment Usage in Software Development. Communications of the ACM, 39(1), 41-49. 10. Hauser, J.R. & Clausing D. (1988) The House of Quality. Harvard Business Review, 3, MayJune, 63-73. 11. Hewitt, P. (1998) Evolution of Corporate Web Sites. A CONTENTIOUS guest editorial 6/1/98. Available at: http://www.contentious.com/articles/1-3/guest1-3.html 12. Ho, J.K. (1997) Evaluating the World Wide Web: a global study of commercial sites. Journal of Computer Mediated Communication, 3(1), June. 13. Internet Advertising Bureau (1999) Internet Ad Revenue Report. Available at: http://www.emarketer.com/estats/082399_iab.html 14. Jackson, H.K.J. & Frigon, N.L. (1994) Management 2000: The Practical Guide to World Class Competition, Van Nostrand Reinhold, NY. 15. Kalakota, R., and Whinston, A. (1996) Frontiers of Electronic Commerce. Addison-Wesley. 16. King, B. (1989). Better Designs in Half the Time: Implementing Quality Function Deployment in America, GOAL/QPC, Methuen MA. 17. Lamia, W. M. (1995). Integrating QFD with Object Oriented Software Design Methodologies. Transactions from the Seventh Symposium on Quality Function Deployment,, Novi MI, 11-13 June, pp. 417-434. http://www.nauticom.net/www/qfdi/7thSymposium.Papers/QFD_OOD.pdf 18. Lindquist, J. D. (1974). Meaning of image. Journal of Retailing, 50(4), 29-38. 19. Lohse L. G. & Spiller P. (1999) Internet retail store design: How the user interface influences traffic and sales. Journal of Computer Mediated Communications, 5(2). 20. Lohse, G.L., and Spiller, P. (1998) Electronic Shopping. Communications of the ACM, 41(7), 81-87. 21. Mizuno, S. and Y. Akao, ed. (1994) QFD: The Customer-Driven Approach to Quality Planning and Development, Asian Productivity Organization, Tokyo, Japan, available from Quality Resources, One Water Street, White Plains NY. 22. Nielsen, J. (1999) Useit.com. Alertbox, Papers and Essays. Accessed September 1999. Available at: http://www.useit.com 23. NUA (1999). How Many On-line? http://www.nua.ie/surveys/how_many_online/index.html 24. Pramataris, K.C., Vrechopoulos, A.P., Papamichail, G., Mylonopoulos, N. & Poulymenakou, A. (2000) Personalized Services and Promotions in Internet Retailing. Proceedings of eBusiness and eWork 2000 Conference, Madrid, Spain, Oct 2000. 25. Roh, J.H., Romano, C. & Ratner, J. (1997) Roles, organization, and support: building usability into the design process. Proceedings of 3rd Conference on Human Factors & the Web, Designing for the Web: Practices & Reflections, Denver, Colorado, USA, June 12, 1997. 26. Deck, S. (1998) Study sees growth in online shopping. Computerworld, May 21. Available at: http://www.computerworld.com/home/online9697.nsf/all/980521study1F956 27. Spiller, P. and Lohse, G.I. (1998) A classification of Internet retail stores. International Journal of Electronic Commerce, 2(2). 28. Spool, J.M., Pool, J. Scanlon, T., Snyder, C., DeAngelo, T., and Schroeder, W. (1998) Web Site Usability: A Designer's Guide. 1st edition (November 15, 1998). Morgan Kaufmann Publishers. ISBN: 155860569X 29. Thackery, R., G. Van Treeck (1990) Applying Quality Function Deployment for Software Product Development. Journal of Engineering Design, 1(4), 389-410. 30. Trout, J. (1997) The New Positioning. McGraw-Hill, January 1997. ISBN: 0070653283 17
31. Vora, P. (1997) Human factors methodology for designing Web sites. Human Factors and Web Development: Erlbaum Publishers. 32. Vrechopoulos, A.P., O'Keefe, R. & Doukidis, G.I. (2000). "Virtual Store Atmosphere" in Internet Retailing. Proceedings of the 13th International Bled Electronic Commerce Conference, Bled, Slovenia, June 19-21, 2000. 33. Vrechopoulos, A.P., Pramataris, K.C. & Doukidis, G.I. (1999) Utilizing information processing for enhancing value: towards a model for supporting business and consumers within an Internet retailing environment. Proceedings of 12th International Bled Electronic Commerce Conference, Bled, Slovenia, June 7-9, 1999. 18

G Ioannou

File: a-quality-function-deployment-approach-to-web-site-development.pdf
Title: 2
Author: G Ioannou
Author: AUEB
Published: Wed Jun 1 14:14:45 2005
Pages: 18
File size: 0.15 Mb


On Bail, 33 pages, 0.2 Mb

, pages, 0 Mb

A Novel, 62 pages, 0.34 Mb
Copyright © 2018 doc.uments.com