Ecommerce in East and Central Europe: Usability of Business Websites
Bob Travica, Rebecca
Olson
Indiana University, Bloomington, Indiana
(Note: This article is based on the study that was partially reported in B. Travica & R. Olson's paper in Preston, Cecliia M. (Ed.), Information Access in the Global Information Economy, Proceedings of the 61st ASIS Annual Meeting, Pittsburgh, PA, October 24-29, 1998, pp. 227-42.)
Abstract
Investigated were commercial Websites in eleven countries of East and Central Europe that provide a snapshot of the early stage of electronic commerce in this region. A framework based on the usability methodology was used, incorporating the method of proxy user. It was found that the ECE business Web is showing a capability of what may be called survival usability (e.g., bilingual sites, and satisfactory grouping of data and navigation at a basic level), while lacking for the most part more competitive usability (e.g., comprehensive data and effective use of graphics).An initiation of transactional capabilities has also been detected.
ACKNOWLEDGMENT
This study was partially supported by a grant from the Russian and East European Institute of Indiana University.
1. INTRODUCTION
The World Wide Web (Web) grows rapidly in terms of the number of sites, industry scope and geographical spread. The Web serves various functions, including support to electronic commerce (ecommerce) (cf. Choi et al, 1997). Firms represent themselves on the Web and implement certain business functions in the Web environment. For example, companies set up Web storefronts as an advertising channel and a tool for integrated marketing (ibid., p. 223). Various kind of data are provided in the Websites; for example, data on the given company, its product line, trading partners and on business developments. Various kinds of businesses increasingly establish their Web presence -- from florists to banks (ibid.). It appears that any company that "means business" must get into the Web.
New geographical spaces are being incorporated into the Web, including those in East and Central Europe (ECE) where the Internet started only in the early 1990s. In this region, the development of the business Web has coincided with the transition of the economy from the bureaucratic to the market model. It appears as if the digital revolution has replaced any other revolution in the minds of Eastern Europeans, particularly in Hungary and Czech Republic, where the Internet is exploding (Gregston, 1996). However, some countries lag behind (ibid., p. 55). Similarly to the West, the new enterprises in ECE are using the Web for self-presentation and advertising. This creates conditions for the embarking of this region into ecommerce. Furthermore, a quick glance over the Web sites in ECE shows that these sites can also provide information on broader, related phenomena, such as relevant legislature, the development of market economy, culture and politics in their respective countries. In effect, this wealth of data poses as a resource that can be used by the Western user for environmental scanning and other activities of business intelligence. This can, therefore, be another function of this Web, in addition to ecommerce.
In order to assess more precisely the usefulness of the ECE business Web for ecommerce and environmental scanning, however, one needs to look more closely at a number of usability dimensions concerning the quality of both the data and the overall information service provided. Given the premise that the ECE countries tend to become part of the global market economy and to deliver offerings to the most developed Western markets, one also needs to assess the understanding of both information and business maintained in ECE countries. All this suggests that a number of questions need to be asked concerning usability of commercial Websites ECE. For example, are textual data presented in a language that communicates well to the Western user? Are communication/transactional capabilities supported? Are the data provided grouped in a way that that facilitates easy navigation? How do the data rate on completeness and timeliness? What is the semantic quality of graphical artifacts? What esthetical aspects of the Web sites are present? What is the provision of hypertext links like? All these are important evaluation dimensions.
The present study was conducted in order to explore business-to-consumer Websites in 11 ECE countries -- Belarus, Bulgaria, Czech Republic, Estonia, Hungary, Latvia, Poland, Romania, Slovakia, Slovenia and Ukraine.
2. WEB EVALUATION RESEARCH
The research on evaluation of the Web is in its infancy. For example, the business-related Web was investigated by Cockburn & Wilson (1996). Basing their sample on the Yahoo! Directory, the authors provided a classification of Web-presented businesses and a description of some features of the Web sites. Travica & Cronin (1996) investigated the business-to-consumer Web in Russia. The authors used several evaluation dimensions suitable to the international business context; some dimensions are included in the present study. Similarly, Ho (1997) assessed ecommerce sites in the U.S., by categorizing them of the provision of information (on the company, products, etc.), advertising focus, and transactional capabilities. In the cognate field of information science, several studies have been conducted on the evaluation of Web sites of the U.S. government (see Moen & McClure, 1997; Wyman et al., 1997). These studies, for example, looked at data or information quality (e.g., completeness and accuracy), data format and appearance, ease of navigation, general and particular user satisfaction, and the match between user expectations and site provisions.
3. METHODOLOGY
3.1 Sample of Countries
The present study aimed at answering the question, What is the usability of the ECE business Web as part of global ecommerce? Eleven countries were included in the present study -- Belarus, Bulgaria, Czech Republic, Estonia, Hungary, Latvia, Poland d, Romania, Slovakia, Slovenia and Ukraine. A pilot exploration of the Web in all ECE countries preceded this selection, in order to get insight in the level of development of the corresponding national Webs. The main idea behind the selection that followed was to learn about the business Web in the countries that were likely to have the lead in the development of this Web. Countries in which transition toward market economy was fastest (e.g., Poland, Hungary, Czech Republic, Slovenia) were the most likely candidates – some other countries were selected on the basis of the impression of exhibiting a fast development of the business Web. (Note that Russia could certainly fit into the selection, but it was not included because it was investigated in the recent study of Travica & Cronin (1996)). Another selection criterion referred to similarities of pairs of countries in terms of the geographical proximity, and/or similarities in histories and sizes; for example, if Czech Republic is in the selection, then Slovakia must be in too; the same applies to Estonia and Latvia, Ukraine and Belarus, and Bulgaria and Romania. In this pairing, there would typically be one country that would “pull” the other (e.g., Bulgaria “pulled” Romania). Although selecting the sample of countries in this way is not perfect, we have no reason to doubt the representative character of it.
3.2 Country Samples
Using the Yahoo! directory's sub category of Business and Economy under the country to be evaluated, as well as directories located on servers in the countries themselves, all sites that were categorized as being a business were included in the Web population in each country. The business had to be a site, defined as "a home page and the pages hyperlinked to it that together have an identifiable owner" (Travica & Cronin, 1996:16). The site had to be located within the country of origin; sites created by residents of other countries but geared toward business in the sample country were rejected (i.e., a Polish-American firm in Chicago marketing to the domestic Polish market would not be counted if the domain was anything but "pl"). One problem of this method was that not every business uses a URL with the country's domain name. With a "net" or "com" in the URL, the site was assumed to be based in the US, with the exception of certain sites such as www.belarus.net, which was both categorized as a Belarus based company on the Yahoo! directory and only had contact information in Belarus.
After all business sites were thought to have been found within a country, a random sample of minimum 10% of the business Web population was drawn for each country. These samples were then used for evaluation. Their basic properties appear in Table 1. (Please see Appendix.)
3.3 Usability Dimensions
The countries sampled were investigated on a number of usability dimensions (see Table 2 in Appendix). In accord with the literature on usability of information system (e.g., Nielsen, 1993), the term usability is used here to mean a general usefulness of a Web site from the perspective of the customer-visitor. The Travica & Cronin (1996) study as well as other relevant research discussed above was used for designing the evaluation instrument which is depicted in Table 2. Some dimensions were developed exclusively for the purpose of the present study. Categorical measurement was used.
As Table 2 shows, the usability dimensions are grouped into two sets called survival and competitive usability. These labels are used to convey the basic idea behind these two groups of dimensions: survival usability refers to those dimensions that make a Web site capable of surviving in the global population of the Web world, whereas competitive usability refers to those characteristics that give a Web site advantages over comparable sites. It should be noted, however, that these dimensions are exploratory in character, and that they count only on an intuitive appeal. Whether dimensions are indeed additive, entirely mutually exclusive, and/or clearly separated into the two categories above could be concluded only on the basis of formal quantitative analysis. This in turn necessitates the precise operationalization of the dimensions, development of appropriate scales for measurement, and test of the scales on exploratory samples. At this stage of the Web research, however, engaging these elaborate techniques would be premature. The goal of this study is hence defined in a more limited way – to use dimensions that are simple enough and have an intuitive appeal, so that they can serve as a robust, pragmatic instrument for the evaluation of the Web ecommerce sites.
Evaluation dimensions are used in the following manner. First, several survival dimensions were applied on the entire sample. This step was performed by one proxy user (see below); on average, it took an hour to evaluate five sites. These dimensions were simple enough to allow for operationalization that could yield credible results with no formal test of measures conducted. Second, additional dimensions of survival usability as well as all the dimensions of competitive usability – both more complex than those used at step one -- have been applied just to select instances of Web sites in each country. This selection was based on impressionistic, rather than statistical criteria; the selection was done after exhaustive browsing through the country samples and by looking at the evaluation dimensions specified. This step was performed by another proxy users. Of these, one or two instances per country were selected as benchmark sites to be discussed in this paper. Since no scrutinized quantitative methods were used, benchmarking may be used directly for rank-ordering countries on usability of their business Webs.
3.4 Proxy User Method
Evaluation was conducted by two proxy users. One was an American female, graduate student, trained in using the Web and in evaluation procedures. The other was a male with East European background and permanently residing in the U.S., with an extensive experience of using and evaluating the Web geared toward ecommerce and other functions. Both proxy users used PCs with a configuration that approximated the computer station of the time used by an average ecommerce customer in the West – a 100-200 MHz processor, minimally 16 Mbytes of RAM, graphical user interface, and Netscape browser version 3.x. Proxy users were sensitized of the time demands characterizing the Western customer.
4. FINDINGS
4.1 Businesses with Web Presence
Computer-related services are the most frequent businesses present on the Web (Table 3 in Appendix). This category occupies 162 sites (30% of the entire sample), and includes software and hardware sales, software production, Web design and hosting, and computer instruction/education. The close second is telecommunications business, with 125 Web sites (23% of the sample). This category includes the providers of telecommunications channels, most frequently those using the telephone and the Internet. Other businesses are much less frequently presented, such as publishers, merchants , and manufacturers (mass production firms, process industry, such as oil and pharmaceuticals, and the small batch industry, such as bear breweries). Other businesses found include advertisers, banks, other financial institutions (traders with securities, brokers, and credit cards [just one site, owner -- American Express]), information providers (database vendors, and providers of various lists of businesses, such as manufacturers, retailers, and video stores), travel agencies, radio and TV stations, language services, real estate agencies, management consulting services, employment agencies, insurance agencies, accountants, legal services, graphics and architecture designers, inspection services, geological surveyors, organizers of exhibitions, music studios, adult entertainment, matchmaking, hotels/restaurants, seaports, construction firms, repairs, cleaning agencies, ticket distributors, modeling agencies, and cosmetics distributors. Table 3 lists the first ten kinds of business that have the highest frequency, and, in the lower pane, some of the more interesting (historically new and/or peculiar) businesses. It is apparent that the frequency of most of the businesses on the Web is very small in absolute terms as well as relative to the computer and telecommunication businesses.
The prevalence of computer businesses also holds at the country level, with the exception of Slovenia; see Table 3). In addition, the proportion of telecommunications and computer businesses holds at a proximate level (1:1.10-1.30) for more than a half of the sample; exceptions are Romania (1:4), Hungary (1:2.1), and Slovenia (1:0.9); Slovakia, Belarus and Bulgaria have no representative sites in the telecommunications category.
4.2 Survival Usability -- Language and Communication
Table 4 depicts the data on language dimensions studied. It is assumed that a condition for becoming part of global ecommerce is the use of legible English, which at the moment is the lingua franca of global ecommerce. The use of English can be exclusive or complementary in bi/multilingual sites. One finding is that approximately one in five Web sites is bilingual (if a site uses both a domestic and a foreign language but so that the latter is not consequently implemented at all pages, such a site is not considered bilingual). The distribution of the bilingual sites, however, varies greatly from one country to another. There are no bilingual sites in Romania and Slovakia; Latvia, Ukraine and Slovenia have less then 20% percent of these sites; Hungary, Poland and Czech Republic are in the 20%-range, Belarus in the 30%-range, while the top of the list is headed by Estonia and Bulgaria.
It is interesting to note that some countries have either no or a small number of bilingual sites, but do have sites which are exclusively written in foreign languages (usually English, sometimes German, or Russian; there was also one instance of Chine se). As Table 4 indicates, Romania has no bilingual sites, but has 8 sites in English (40% of the Romania sample). Latvia has the greatest number of sites in a foreign language, while Slovakia has neither bilingual nor foreign language-only sites.
Another dimension of survival usability is automated communication (the link that launches an email application) with the Web site master or with representatives of the business that owns the Web site. Communication can be seen as conducive to establishing elementary transactional capabilities. Most of the entire sample does exhibit this capability. With the exception of Latvia, this finding also holds for each country.
4.3 Total Usability by Country
This section continues the discussion on findings, by expanding the scope of survival usability and by bringing competitive usability into the picture. For each country, one or two representative benchmark sites are discussed. The notion of representatives used here is based on impressionistic rather than statistical criteria, as was discussed in the methodology section.
4.3.1 Hungary
The Hungarian ecommerce-dedicated Web leads in size in the sample studied. Some of sites in this Web exhibit the difficult to reach balance between information and esthetical dimensions. One example is Webstar Csoport (www.dravanet.hu/webstar), a Web site designer, that sports interesting design and a fair amount of data on its staff and products. Another feature that is less frequent in other countries studied is the communication facilities, such as chat corners. However, one of the links that was supposed to lead to a bulletin board (www.enet.hu) turned out to be “dead,” while the link to another chat facility (www.eps.hu) froze the screen of the proxy user. Another example of good intentions that failed is the music applet provided at a site of an international insurance company (www.generali.hu); the music downloaded is of such a degraded quality that the site would do much better without providing this feature.
There are conspicuous examples of trade-offs between information and esthetical dimensions. For example, the Drava Radio site at www.dravanet.hu/dravaradio appears to be graphically simple but attractive, while, at the flip side, it could be difficult for visitors to understand immediately what sort of enterprise the site represents. There is one problem in Hungary, however, that appears less frequently in other countries – semantic ambiguity. For example, a site of a cotton and wool manufacturer (www.eleneder.hu/rico) brings the following slogan in its pages: “Our shareholders are Hungarians in 100%”; another example refers to a driving school (www.datanet.hu/majsai) that uses an icon of a g-string-clad woman on a motorcycle as a navigation button. While the first example could indicate a specific understanding of business in which political connotations play a role, the second may reflect cultural peculiarities that can have unintended connotations in the international marketplace context.
The benchmark selected for this discussion belongs to company W3 Computer (www.w3.hu); see Table 5. Even though it is small, the site offers data on the company in a legible and esthetically acceptable format. The strong side of it is the fact that it exists in four different versions of data formats -- from text only (the low-end), though animation and frames (the high-end). It also has a touch of humor in text provided, and makes an attempt at being especially attractive in graphical solutions (e.g., it has a watch with the angled dial over which a seconds hand is moving – the watch, however, is not quite correct).
4.3.2 Poland
Poland, the largest country in the sample of countries studied, houses the second largest business Web. A larger size and diversity characterize this Web. The kinds of businesses represented in the Web that are less frequent in the entire sample include oil distillation, private computer schools, smoked salmon production, music stores, road construction, medical merchandise, a car dealership that offers online ordering of cars, and a sea port. The data on company profiles and services are regularly provided and easily accessible in this Web. On a weaker side, a significant number of sites, more so than in most of other countries studied, was under development at the time of data collection; this especially applies to English pages. Poland also demonstrates remarkably the trade-offs between data/information content, navigation and esthetics of the Web. Specifically, a noticeable number of sites have a “clean” design that provides easy access to company profiles and product data, esthetical attraction n is missing in many of these sites.
The counterpart to “clean” designs are those with graphically represented information, that often feature frames as well as animation. These could be seen as being more “exciting,” but they, on the other hand, suffer from a slower response time and navigation difficulties (e.g., the problem of differentiating “clickable” from “non-clickable” graphics, as demonstrated by a software distributor’s site at www.hsn.com.pl). English usage varies from being very good (e.g., www.centrozap.com.pl; see below) to unacceptable (e.g., the computer firm at www.demos.pl). Not unlike other countries in this study, there are examples of beginner errors, such as the “click here”-links, too long pages containing complex information formats, the lack of home page-buttons that force the visitor to rely on navigation functionality supported by the Web browser, and so on. It is to be noted that one of the sampled sites offers cars for sale online (www.it.com.pl).
Import-export Company Centrozap (www.centrozap.com.pl; see Table 5) possesses a Web site that is representative of the more usable sites. The main strength of this site is a well-categorized main-menu, which is placed in a left-hand frame; the menu facilitates easy navigation, even though the referenced files appearing in the right-hand, larger frame do not correspond to the menu items in the one-to-one relationship. Also, data coverage on company activities is exhaustive, and appropriate graphical representations are used for financial data. The site is lacking several features, and especially links to other related Web sites and multimedia formats. Another select site is Commpol SA, a computer hardware distributor (ww.commpol.com). This site excels in its home page (company name, address and images [“non-clickable”] of brand marks merchandised are in a larger frame, while the menu items that indeed look like buttons are vertically aligned in the right-hand frame), effective use of graphics and text ( hypertext links are implemented in text, while colorful images, for example, of brand names accompany the text), easy navigation (the main menu is repeated in the bottom of every page), the links to interesting sites provided, and search engines offered. The site, however, does not provide text in foreign languages, except a page on the company profile written in English.
4.3.3 Czech Republic
The Czech Republic ranks third on the size of its business Web. The Czech business Web exhibits a diversity of businesses, including those that are infrequently present in the entire sample (e.g., beer brewing, shipping, lithography, car dealers, construction and surveyor services, porcelain manufacturing, etc.; it also includes one of two adult entertainment sites found in the entire sample-- the other is in Slovenia). Several sites with developed ordering capabilities and ads on sales of computer software and hardware were detected.
From the perspective of usability, the Czech sites vary greatly in terms of quality – from very simple and sometimes unattractive, through more complex, professionally crafted sites. Some appear to be very usable on certain dimensions, even though not so in overall scoring (e.g., ORSA Real Estate Agency at www.arsystem.cz/orsa, which exhibits a neat categorization of data/information and a form-supported communication with visitors). The yielding to a "temptation of graphics “can be identified with a noticeable number of sites, which is associated with slower response times. Another problem that can be encountered refers to company data that sometimes is not readily accessible, and so the visitor has to learn about the sort of business a site owner is in from product descriptions. Yet another potential inconvenience for the Western user is the location of the button for accessing English text, which usually is in the bottom of home pages. Moreover, another navigational peculiarity is that home page buttons in the English version of sites sometimes appear in Czech only.
The benchmark site is Intersoft (www.intersoft.cz), a software vendor which appears to be run by computer science students from University of Brno. As Table 5 shows, this site has lost points on slower response times (due to an exaggerated amount of graphics), the quality of English, and the lack of hypertext links. Strong points are given for the provision of both iconic and textual menus, and a graphical symmetry and appealing selection of colors. An example of competitive usability is represented by the site Video on Line (www.vol.cz), a provider of telecommunication, Web design and hosting, and advertising services. This site has earned points on billingualism, quality English language, data coverage, esthetical dimensions and navigation. Also, it is among the rare sites in the entire sample that provide more developed transactional capabilities -- some sort of automated payment (the payment can be done by credit cards, but so that the payer must first call the company by the telephone). Moreover, the site provides an abundance of links, some of which lead to virtual stores in other countries, while others rather cater to the potential curiosity of the customer, such as a link to the site of the famous sport car manufacturer Ferrari. On the other hand, this site needs an improvement in terms of timeliness of data and difficulties with accessing data on company profile.
4.3.4 Estonia
A salient characteristic of the Estonian business Web refers to a variety of information-related businesses, apart from those under the headings of computer and telecommunications. These include online banking, information brokerage, and stock exchanges. An example is Forex Online Bank (www.forex.ee), that declares itself to be a “virtual bank”, and represents an operation of Forex Bank which was established in 1992. Forex Online Bank offers remote account management (opening, querying, payment withdrawal), ATM cards, quotations for securities, conversion rates for non-convertible currencies of the countries that descended from the former Soviet Union, etc. This site ranks very high on transactional capabilities in the entire sample. However, problems relate to the server response time, semantics of graphics used and navigation are weaker aspects of this site (Table 5).
Another interesting example of an Internet-Web-based information business is represented by Ekspress Hotline (www.hotline.ee) – an information broker providing data on firms and the economy, entertainment, lottery, and transportation time tables. The company started out by using only the telephone for communication with customers, while company operators served as intermediaries to Internet-based information sources (access is direct now). Estonian Web sites, in a number of instances, provide auxiliary data on company histories and the development of market economy. This sort of data could be useful to the foreign customer in terms of learning about Estonian economy and perhaps about credibility of domestic firms. A foreign environmental scanner can draw a similar value.
The representation of main menus in graphical format appears to be a standard in the Estonian Web. The visual side of the Estonian Web is striking with respect of numerous photographs depicting company staff (sometimes, from the chairman all the way down to the driver). This could make implications for time expenditure expectations and sense-making of the Western ecommerce customer.
4.3.5 Latvia
The most remarkable characteristic of the Latvian business
Web is the lack of company profiles; most of the data provided is about
products and services offered. This is in a stark contrast with Estonia, where
Web sites provide exhaustive data on company structure, history, staff and
finances. Another problem concerns the clarity of data: sometimes it is not clear
whether a site represents one or more businesses. An example of this is the
mutual referencing of a night club site and a restaurant site which offers a
casino and limousine service; the latter refers to the former as “a place to go
after.” An additional problem of the Latvian business Web concerns the use of
English, which is often at a grammatically unacceptable level.
An interesting linking of business, Web and the Internet in a broader sense is represented in the Internet cafe called Internet Kafejnica Bilteks (www.binet.lv/home/cafe/english/). The site represents a cybercafe, which, in addition to coffee, tea and soft drinks, offers Internet stations for perusal of its clientele. The offerings include, in addition to coffee and tea, access to the Internet that runs from 15 minutes to an hour, training on navigation, and printing services (fees for training are twice the access fees). For self-promotion, Kafejnica uses digital snapshots of its terminal space, which are updated in five-minute intervals. As Table 5 shows, the Kafejnica Web site scores on a number of usability dimensions. It is interesting to note that a quick search on cybercafes in small cities in the U.S. (supposedly, the leader in ecommerce developments), conducted for mock-comparison purposes, showed that cybercafes are rather an exception than rule.
5.3.6 Ukraine
Computer, telecommunications and Web companies dominate Ukraine’s business Web, which is typical for younger Internet environments. There are examples of instances of interesting design solutions. However, the spirit of computer “hacking” can sometimes result in expressions that could be more welcome in the San Francisco Bay Area than at Wall Street; for example, a Web business Ukraine Online advertises itself with the slogan “This server made with 100% recycled electrons” (see www.gu.kiev.ua). On the other hand, there appears to be competition among Web and telecommunications providers in terms of graphics, which in some instances approaches gigantic proportions, resulting in long server response times. This can be accompanied by Java applets for animated banners or icons, which furthermore increase the response time; the time problem could be coupled with functionality difficulties, if the Java code contains a bug, or would not run on the client machine, in which case the browser freezes, or some other function fails (e.g., the anointed banner is not decoded properly); for an example of both these and the graphics problems see www.ww.net/mars.
Monolit is a company in the computer and telecommunications business (www.cs.kiev.ua; see Table 5) which also is its own Web site designer. It scores well on data representation and completeness, basic communication support, and esthetics (e.g., one menu is implemented as a plaza with “clickable” buildings that represent particular CD-ROM services Monolit provides).
The Monolit Web site, on the other hand, losses points on navigation (go back-buttons are missing, there are menu options that are not accessible to general public, which may be confusing), and server response time (overwhelming graphics appears to be responsible for this slowness). The graphical representation of data leaves an impression of extravaganza when the visitor encounters plain text that is implemented as a picture. While a demonstration of the technical proficiency might be able to explain the owner-designer’s motivation, this can incur unreasonable time costs on the visitor side. Some symbols presented at the Monolit site could be indicative of the transition toward market economy in Ukraine. For example, the home page invites the visitor to “Get some $$$ now” by bringing new clients to Monolit. The suggestion of this economic quid pro quo logic is accompanied by an icon of a stereotypical capitalist (the cylinder, and round, smiling face).
4.3.7 Slovenia
The Slovenian business Web exhibits diversity comparable to that in the Czech Web. Among the businesses that are typically infrequent in the entire sample is a manufacturer of cemetery equipment, Chinese restaurant that takes online orders, brewery , food distributor, and an adult entertainment (pornographic) site. A good balance between data/information and esthetical aspects can be found in this Web; another stronger point refers to navigation aspects.
Semantic ambiguities can be found at some Slovenian sites. For example, a manufacturer of cemetery equipment (www.menina.si) prints the following motto on it s home page: “Ready to a journey into unknown.” While this motto may carry undertones of black humor, the following has rather political ones (once grammatical problems are put aside) – “Mobitel was founded in the year of the independent state of Slovenia” (www.mobitel.si). It remains unclear when this happened, and what is the significance of this fact for the Mobitel business.
An instance of a blend of survival and some competitive usability is the mobile telephony provider Mobitel (www.mobitel.si; see Table 5). The site provides sufficient coverage of data on company and products, while exhibiting navigational problems in r elation with the use of frames, slower server response time, errors in English grammar, lack of ordering capability and semantic ambiguities (one is mentioned above, while another refers to the semantic role that pages on bird species have within the site ).
4.3.8 Romania
The Romanian business Web is small, mostly written in English, and infrequently visited. It contains instances of survival usability at best. Common problems frequently encountered concern very slow server response time (often influenced by graphic s-demanding screen backgrounds and other items), low quality English, dated timestamps, many links referencing files under construction. A particularly conspicuous problem is that it can be very difficult to understand what sort of business is a particular web site is about, because it is not specified directly in the home page and/or company data is not readily available (see for example verena.roknet.ro/pemido/). Another peculiarity is the self-advertising of some ISPs in the top of customer home pages. In addition, some sites also present a copy of the work certificate of the company represented.
Semantic ambiguities are not rare. For example, the link launching email program is in a number of sites associated with the label inviting the visitor “to fax” the company (e.g., www.kappa.ro/yoursite/eprecizia). Moreover, serious navigation problems can be encountered at some servers, resembling a true maze of sites that are hard to differentiate one from another (e.g., www.kappa.ro/furniture).
Cybershop (www.cybershop.ro) is an instance of more usable sites. Its stronger points are provision of sufficient product data, communication support, including automated ordering, and basic esthetical features. On the weaker side, the site is not consistently bilingual, has a somewhat ambiguous menu block which implicates difficulties in navigation, implementation of a peculiar vertical black stripe across the middle of the screen which reduces legibility of text, the lack of hypertext links to other sites and a long server responses time.
4.3.9 Slovakia
Slovakian business Web contains some examples of simple and effective designs. On the other hand, examples of overemphasized graphics (e.g., www.gamoba.sk) and the focus on esthetical aspects (e.g., www.reprosova.sk) can often be found, which offset data-information and time aspects. Small visitor numbers are also remarkable as well as relying on maps, which might not communicate well to visitor with no knowledge of local geography.
The site of Maxum advertising agency (internet.sk/maxum) is among the more usable ones, exhibiting almost all of survival usability except data timelines, and competitive usability limited to quality English and esthetics. The main menu categorizes the data provided in an effective and creative manner (items include: Who we are, What can you expect from us, and Who likes us).
4.3.10 Belarus
Belarus has the second smallest business Web in the selection of ECE countries studied. Most remarkable characteristics rather signify a lack of certain usability. For instance, the company profile is incomplete and difficult to locate. Another problem refers to semantic ambiguities. For example, one Web master advertises itself as “the really multimedia server,” while another emphasizes its “independence” (see below). Overemphasized graphics are common.
The benchmark site in this selection is WWW Belarus (www.belarus.net; Table 5), an ISP and Web master. It appears to be exhaustive in data/information coverage, supports communication, and offers functional links and other screen items as well as textual and graphical information formats that blend into an esthetically acceptable design. However, company information is difficult to locate, because it is accessible, counter-intuitively, behind the Service item in the main menu, as if the goal is to hide the fact that WWW Belarus’ Web operation is also a business (more discussion on this possibility provided further below). Possibly a related comment featured in the site, contributing to the confusion of the foreign visitor: “Sorry, we have problems. Enemies of democracy in Belarus destroyed our server. But we work hard!”
WWW Belarus also exhibits navigational problems (there is competition for menu items between hypertext links in the central part of the screen and items in the left-hand frame; also, the former set of items consists of icons, images and some text, and it is not always clear which graphical elements implement hypertext links). Other problems concern the amount of graphics, a multiplicity of font sizes and colors in the same screen, and inefficient use of space. For example, there is a welcome page, but the only data/information it provides is a map of Europe, with an area under lenses that supposedly is Belarus, and no contact information is provided, nor are any communication features such as guest book. To make this information inefficiency even worse and to incur more time expenditures on the part of the user, this page loads slowly, and confuses navigation because it is, counter-intuitively, implemented as a hypertext link to the page with the main menu.
The WWW Belarus site is also representative of semantic ambiguities that can be found in the Belarus business Web. For example, the visitor can find many iterations of the claim that the site is “independent” -- “This is really independent Internet resource!,” “Independent version,” etc. These warnings might throw the Western visitor in confusion as much as the accusation of the "anti-democratic server killers" above.
4.3.11 Bulgaria
The business Web in Bulgaria is characterized by an emphasis on English that is of variable quality, a quicker server response time, and less satisfying data coverage and format. One peculiarity is the presence of photographs, which in some cases includes the picture of the owner (see www.eunet.bg/tonisto).
The site of Alian Ltd. Travel Agency (www.alian.bg) passes the test of survival usability, except on ease of navigation, and is competitive in quality English, ordering capability, and data completeness (provided are data on company profile, geographical area, entertainment, tourist tips, pictorial representation of tourist objects, etc.). Its competitiveness is reduced by a longer server response time caused by a great number of photographs, whose impact may vary with different foreign markets (e.g., photos insinuating that guests may be nude at premises of some tourist facilities).
Table 6 summarizes findings on the salient usability drawbacks found in the countries investigated. Indeed, these drawbacks can be found in any of the countries, but more so in some than others. Neither this nor Table 5 is intended for a scrutinized rank-ordering the countries; one practical purpose is to help inferring possible design enhancements. The criteria used build on the usability dimensions investigated, and is some cases expand these dimensions (e.g., all the criteria at the top of the table refer to the data completeness dimension). The most often drawback is semantic ambiguity identified in five (textual or graphical data that convey meaning which may not be readily attainable to the Western customer), followed by the problem of overemphasized graphics salient in three countries.
5. DISCUSSION
Computer businesses prevail in the Web-based business-to-consumer ecommerce in ECE. This is in agreement with the findings of the Cockburn & Wilson (1996) study of the commercial Web in which they studied sites accessible via a Yahoo! directory. A direct comparison of the size of the telecommunication category, which is the second most frequent in the ECE Web, with its counterpart in the Cockburn & Wilson study is not possible, because of different contents of this category. It is, however, interesting that ECE sites representing what Cockburn & Wilson (1996) call business activities (real estate, advertising, marketing, employment, conference organizing) make just about 6% of the sample as opposed to 14.2% in the Cockburn & Wilson (1996) study. This can be explained by the fact that market economy in ECE was hibernated for four or more decades, and therefore these typical market-related services are just taking off. The prevalence of computer over telecommunication sites holds for all countries investigated but Slovenia. This finding, however, differs from that in Travica & Cronin’s (1996) study of Russian business Web, where telecommunication providers were represented more than twice as much as computer firms. Increasing accessibility of the Web to various businesses, including computer-based ones, may be responsible for this difference, rather than local Russian conditions. An implication of the computer businesses prevalence is that hi-tech businesses lead the development of ecommerce in ECE.
The fact that about 28% of Websites uses English either in combination with a domestic language or exclusively is an indication that ECE’s ecommerce is being developed with the foreign customer in mind. However, not all the countries appear to be in the same lane. The fastest are Bulgaria and Estonia, then follow Latvia, Hungary, Czech republic, and others. An inward orientation may be characterizing Slovakia with no bilingual or foreign language site. Romania appears to be in the same category, since just 8% of sites use a foreign language.
One specific problem in ECE’s ecommerce comes from semantic ambiguities. The Belarus site discussed earlier is notorious in this sense, with its claim of being “a really independent Internet-resource.” Trust issues that loom large in the electronic context, ranging from branding through payments (cf. Choi et al., 1997), may be affected by this sort of ambiguity. Another problem concerns data completeness. This is true particularly for Latvia and Belarus, where company profiles are rather difficult to locate in Web sites. In the background there might be specific socio-cultural environments in these countries that are in the process of the transition ensuing the dissolution of the Soviet Union. The vacuum left after the dissolution of the bureaucratic economic and social order in the former Soviet countries appears to be filled by social forces that border with organized crime (see Handleman, 1994; Timofeyev, 1992). Therefore, there may exist an urge of hiding from racketeers that prompt ecommerce players to “camouflage” their Websites, that is, hide their business purposes. This hypothesis may be supported by the finding that Latvia has the largest number of Websites in foreign languages, which could provide another shelter from mobsters. However, Belarus has no site exclusively written in foreign languages, although it is in the cluster with a high percentage of bilingual sites. If this hypothesis on “camouflaging business” is true, then integrating these countries into international ecommerce flows will develop with difficulties.
Company profiles are also lacking in Romania, but the cause here may be in a specific cultural understanding of business and information (see below). Estonia and Poland serve as a counter-balance to the meager data on company profiles mentioned above. Also, Estonia on its part provides auxiliary data on the history and development of market economy in this country, as well as on history of certain firms.
Sometimes, it is not easy to differentiate between businesses represented on the same site, and so the visitor may not be certain which business data apply to which company. This is manifest in the Romanian and Latvian Web. In Romania, for example, ISPs advertise themselves right at the top of client home pages, and sometimes menus lead to what appears to be a different company's site. The occurrence of strikingly obsolete data (i.e., year or more time lapse from last site update as indicated by timestamps) can also be found in Romania. This could be another piece of evidence of a specific understanding of the concepts of information and business in this country.
A more widespread shortcoming is that of semantic ambiguity, which is particularly present in Hungary, Romania, Slovenia, Ukraine and Belarus. Instances of these are statements on “truly independent Internet”, “100% domestic stockholders”, relating dates of business to political events, invitations to the visitors to “fax” instead email Web masters, exhibition of pictures of business documents (certificates), etc. These ambiguities are probably related to various cultural and social factors in respective countries. These factors can be expected to be resilient, and as such they imperil global ecommerce developments in ECE.
Instances of overemphasized graphics are common in ECE’s commercial Web, and particularly in the Czech Republic, Ukraine and Belarus. Photographs (for analytical purposes these are separated from other graphical data) are more abundant in Bulgaria and Estonia. It appears as though the Webmasters were attempting to demonstrate their technical savvy before customers by building abundant graphics and photographs into their Web designs. Web mastering businesses are the most salient example of this tendency. Overemphasized graphics/photos contribute little or nothing to informativeness of a site; an example is photographs of company people ranking from a driver to a chairman in an Estonian site. Cultural reasons might also play a role here: the Web may be understood as a display space that should be complemented with a sort of the family photo album; or, it may be that developing relationships of trust (getting to know someone better) necessitates conveying an image of the people behind the product (a similar phenomenon was detected among Western scientists in Hiltz’s (1984) study of computer mediated communication). Whatever the reason, excessive graphics bear costs.
One cost of overemphasized graphics is a larger expenditure of visitor time. A larger costs is that the visitor never becomes a customer. Time expenditures can also be inflated by confusing navigation which may be caused by the way data is grouped, poorly-linked frames, menus that are misplaced or are not communicative enough, and certain navigation peculiarities. The last item was identified in Slovakia, where many sites use geographical maps of the country for navigational purposes as if the visitors should know Slovakian geography. Another peculiarity was found in the Czech Republic, where the English version of sites can contain navigation buttons with inscriptions in Czech. Other time eaters are multiple frames (some sites had up to eight on a page) and non-functional screen items (Hungary and Ukraine). The importance of the speed of moving around Web pages and retrieving necessary data cannot be overstated. Given the accelerating competition in the emerging global electronic marketplace, the visitor time costs/savings are likely to become a basis of ecommerce sites differentiation.
It appears as if all the Web designers in a particular ECE country attend the same classes in Web design. Design solutions, including navigation errors, repeat with a certain regularity within the same national context (e.g., Slovakian sites emphasize animation, Hungarian sites utilize audio, while Czech sites incorporate flamboyant graphics). This design uniformity could impact on impression management of customers, diminishing their capabilities of memorizing and perceiving differences across firms.
Email communication with visitors is supported in majority of ECE’s ecommerce sites (83% on the average). Hungary attempts to provide additional communication channels, such as electronic bulletin boards; however, all facilities of this sort that were randomly selected for testing turned out to be non-functional. This feature represents nascent transactional capabilities. True transactional capabilities in terms of ordering and other services exist in some sites in almost countries except Ukraine and Belarus. The leaders is Czech Republic, followed by Hungary and Poland. In Czech republic, some sites have true commercial flavor, featuring, for example, sales of old software and computer products. In Poland, one site offers cars for online sale. Transactional capabilities are also supported by online business forming the emerging information sector in Estonia. Even in the countries that are just starting venturing into the Web universe, there exist some sites with ordering capabilities (e.g., Romania). The state of transactional capabilities may well be an indication of a potential for developing ecommerce in ECE countries.
With regard to esthetics, the ECE Web is very uneven, with the design ranging from very simple, Spartan designs, through overly ambitious, baroque designs featuring frames, graphics and animated banners. There are instances of very esthetically pleasing sites that exhibit a touch of expert graphical design (e.g., in Hungary). The trade-off in such sites, however, is losses in information clarity and navigation -- it can be unclear as to what kind of business the site represents, what the menu items mean, what part of the screen “hides” a hypertext link, etc. These sites can, again, have diminished usability in the eyes of the Western customer used to speedy service.
External conectedness (the provision of relevant hypertext links) is satisfactory in the 60% of the sample which is comprised of telecommunication and computer firms. In contrast, the reminder of the sample is rather poor in hypertext links. This is peculiar, given the simplicity of building hypertext links into a Website. An explanation could be that there is the lack of understanding what constitutes the value for the visitor. Often, the visitor needs to learn about a family of products, compare and contrast against competitive products, and so on. Easing these activities may generate value for the visitor, make him appreciate its provider, and perhaps make a purchase.
In conclusion, this study has taken a snap-shot of early days of business-to-customer ecommerce in ECE. The business Websites, on the average, demonstrate a healthy capability of survival usability, which is another way of saying that ECE has markedly made its entry into global ecommerce. In addition to advertising and supporting online sales, the business value of the ECE Web lies in its provision of information that is useful from the perspective of environmental scanning. This conclusion applies to some countries more than to others. Specifically, Czech Republic, Poland and Hungary are on the lead. Most of other countries are in the middle bracket, while a minority is lagging farther behind. The ECE entry into global ecommerce is conspicuous particularly given the short period of both using the Internet and developing market economy. If these initial achievements get scaled up toward competitive usability, ECE’s ecommerce may be able to close the gap dividing it from the Western counterpart in the foreseeable future. Further research needs to reassess the ecommerce issues addressed in this study, including the gap question. In addition, it would be interesting to see to what extent the businesses engaged in ecommerce are different in the organization and management sense from more traditional businesses, and if ecommerce businesses make any impact on organizational change in ECE economies.
REFERENCES
Choi, Soon-Yong, Dale O. Stahl, & Andrew B. Whinston (1997), The economics of electronic commerce. Indianapolis, IN: Macmillan Technical Publishing.
Cockburn, C., & Wilson, T.D. (1996), “Business use of the World-Wide Web,” International Journal of Information Management 16, 2, 83-102.
Gregston, Brent (1996), “The European Picture: The Net is Conquering the Old World,” Internet World, December, 52-55.
Handelman, Stephen (1994). Comrad criminal. London: Michael Joseph.
Hiltz, S.R. (1984). Online communities: A case study of the office of the future. Norwood, CT: Ablex.
Ho, James, "Evaluating the World Wide Web: A Global Study of Commercial Sites", JCMC, 1997, 3(1), Available: http://www.ascusc.org/jcmc/vol3/issue1/ho.html
Moen, W.E., & McClure, C.R. (1997). An evaluation of the federal government's implementation of the government information locator service (GLIS): Final report. Submitted to the General Services Administration Office of Information Technology Integration. Washington, D.C.
Timofeyev, L. (1992). Russia's secret rulers. New York: Alfred A. Knopf.
Travica, B., & Cronin, B. (1996). The business Web in Russia: Usability for the Western user. In S. y Hardin (Ed.), Proceedings of the 59th ASIS Annual Meeting (pp. 16-23). Medford, NJ: Information Today.
Wyman, S.K., Beachboard, J.C., & McClure, C.R. (1997). User and system-based quality criteria for evaluating information resources and services available from federal websites: Final report. Dublin, OH: OCLC Online Computer Library Center, Office of Research.