One view contends that the Internet will become a major new retail format, replacing the traditional dominance of fixed location stores. However, little academic research exists to either disprove or support the claims of Internet penetration by retailers. Seeks to redress the balance by presenting a comprehensive and rigorous review of UK retailer Internet activities. A sampling frame of 1,099 UK retail multiples was used, and each Web site individually inspected to categorise the range of marketing functions and services offered.
The indings Indicated that, despite the hype, the majority of retail organisations surveyed have not yet registered a Web site address. Moreover, of those retail organisations that have developed a Web site, the vast majority are using it primarily as a communication tool to promote corporate or product information to Internet users, rather than to support direct sales. In conclusion, summarises the implications of these current levels of Internet activity for the future of retail marketing.
Retailer adoption of the Internet Cathy Hart, Neil Doherty Fiona Ellis-Chadwick European Journal of Marketing, Vol. 4 No. 8, 2000, pp. 954-974. # MCB University press, 0309-0566 Introduction Considerable attention has been focused on the Internet and Its commercial potential for retailers. However, two key areas of confusion emerge for retailers; first, what is it and what role can the Internet play in retail marketing? Some academics (Evans, 1996; Van Tassel and Weitz, 1997) assert that the Internet will provide a new retail format, usurping the traditional dominance of fixed location stores.
Alternatively, Hazel (1996) sees the Internet performing a supporting role for existing marketing activity. Whichever role Is adopted may ultimately determine retailing”. This raises the second area of confusion; speculation concerning the actual size, growth, or future potential of the ‘cyber retail market” has been rife, and existing research has failed to define the current structure of on-line retailer activity. Which retailers are on-line and are they using the Internet strategically or tactically as a marketing tool?
This information is critical for retailers developing Internet marketing strategies; and may help identify the sectors or variables that hold most potential for online retailing. Conversely, it could also expose any retailer weaknesses or threats to existing retail formats. This paper addresses the first issue by initially discussing the role of the Internet as portrayed in recent literature. The second issue is addressed via a comprehensive review and classification of I-JK retailer Internet activities.
A sampling frame of the top 1,099 1-JK retail organisations was used, and each Web site individually inspected to categorise the range of marketing functions and services offered. A discussion of the findings follows, which highlights the key implications for retail marketing strategy, efore the conclusions are presented. Contextual background Retail marketing and the internet If the 1990s were considered the era of database driven customer management, the early part of the Millennium could provide a more ‘ ‘integrated approach to retailing” (Mulhern, 1997) through interactive customer management.
Key IT technologies, such as EPOS, EFTPOS, loyalty cards and database marketing, have identified who was buying what, when, and how, enabling retailers ‘ ‘systematically to tie merchandising practices to customer buying behaviour” (Mulhern, 1997). However, this reactive, product-oriented approach epended on the skills of the merchandiser and buyer to interpret consumer needs effectively. Furthermore, direct interaction with the consumer was generally non- existent, slow or impersonal. Potentially, the Internet offers the means to target and interactively communicate with consumers individually, through to the final transaction of the product.
So far, the Internet has been used in three main ways to facilitate retail marketing. In the most basic respect it is a means of communicating information about the retail organisation, its products and services (Bruno, 1997). At the next level, it is used as a more proactive marketing tool, inviting consumers interactively to access the Web site to gain more product information to facilitate their buying decision-making process (Hazel, 1996), at the same time providing valuable consumer data for retailers to enable greater targeting.
US retailers view the Internet as a communication tool for attracting new customers, penetrating new markets, promoting the company’s brand and improving customer retention (Ernst and Young, 1999). A third level involves retailers physically selling products on-line hrough transactions with the consumers (Hoffman et al. , 1996), in other words providing an additional channel to an existing store based or mail order operation. Potentially, if the latter proved highly successful, it could replace fixed location stores, with a new electronic retail format (Van Tassel and Weitz, 1997).
The implications are considerable, ultimately the Internet could fundamentally alter the way that consumers shop and thus revolutionise the retail environment, transforming the local high street into a global cyber high street. Consumer demand for the Internet is retailers. Whether the consumer has access and how they use or perceive Internet shopping will affect its ultimate success (Shirky, 1997). Internet retailing offers a retail experience that is totally Retailer adoption of the Internet 955 European Journal of Marketing 34,8 956 different from fixed location retailing (Westland and Au, 1998).
Comparison and price shopping across a greater number of sites will be easier and could be achieved within minutes. Indeed, previous experience in the US (Ernst and Young, 1999) has shown that more consumers use the Internet for research in the early part of the uying decision-making process, than buying direct on the Web (subsequently purchasing the product by store or order on the telephone or fax). However, the Internet also places more demands on end users ?± their cognitive ability (Thurig et al. , 1995), comprehension, cognitive overhead (Conklin, 1987) and their information and searching and sorting abilities (Alba et al. 1997) may also deter some consumer segments from switching to cyber shopping. An obvious parallel may be drawn between fixed and virtual stores, and various attempts have been made to apply tore image theory (Spiller and Loshe, 1997) and retail patronage behaviour (Palmer, 1997). However, certain aspects of retail marketing theory do not transfer easily from a physical store to a virtual one. Image and patronage dimensions that were once relevant to fixed location stores may no longer be applicable to the virtual retail store.
Thus a new set of image dimensions for the virtual store may need to be determined via future consumer research. Consumer predisposition toward the Internet as a new retail format may also constrain future growth. Within the I-JK, similar to other ountries, growth of retail formats is traditionally gradual, due to establishing the physical infrastructure, providing the consumer time to adapt to the new methods of delivery. While Internet retailing could be interpreted to be at the innovation stage of the retail life cycle (Davidson et al. 1976), the rate of retailer penetration of the Internet as a new format may, in turn, affect the rate of adoption by consumers, thus signalling the demise of other, older formats. As a consequence, those retailers with a strong, established brand presence, physical distribution relationships, and capital nvestment in traditional formats may be less inclined toward expansion into a non- store, electronic format (Ghosh, 1998). Additionally, new entrants to the retail sector may face fewer barriers to entry, and are thus poised to adopt the new format; witness the success of Amazon, the on-line bookstore.
By the same token, it has been suggested that smaller retail organisations are a category most likely to adopt the Internet due to their greater flexibility (Auger and Gallaugher, 1997), limited resources (O’Keefe et al. , 1998) and lack of scale, encouraging collective marketing via mall business networks (Poon and Jevons, 1997). While it appears that size of organisation may be a critical factor influencing retailer adoption of the Internet, the US experience indicated otherwise; a survey of the top US retailers suggested that smaller companies were less likely to be on-line (Morganosky, 1997).
The type of the Internet. Convenience, shopping and speciality goods may offer varying attractions to the on-line consumer. Evidence from the US experience of Internet retailing indicated a preference for electronic and related products (Morganosky, 1997). In addition to PC peripherals, banking services, books and magazines accounted for 47 per cent of Internet sales in the USA (Pavitt, 1997), whereas grocery shopping was considered less compatible with Internet shopping. In contrast, retail structure within the I-JK tends to be dominated by the food sector, and within that sector, the large food multiples.
High concentration and competition, combined with static food sales and possible saturation of the market, suggest that the major I-JK food multiples may perceive the Internet to provide additional growth via a new marketing channel. It might also be expected that I-JK non-store or mail order companies are likely early adopters seeking to extend their ‘ paper formats into the electronic world of cyber-commerce” Cones and Biasiotto, 1999). Currently, the home shopping market struggles to maintain about 5 per cent of all European retail sales.
Unless on-line retailers can overcome consumer resistence to nonstore based formats, the retail marketing potential of the Internet may be limited; compared with traditional mail order, its key advantage will be the ability to target niche markets directly (Reynolds, 1998). On-line shopping in the I-JK is currently estimated at 0. 2 per cent of I-JK retail trade, but sales are optimistically predicted to multiply by 1 5 times over the next three years as more older consumers accept the electronic format (Verdict, 1999).
Previously, some I-JK retailers have been criticised for not exploiting the marketing opportunities of the new medium, ‘there is no point in selling on-line if all you’re going to do is sell three bottles of wine and update the site once a year” (The Grocer, 1998). Can this criticism apply to the retail industry in general or are different sectors more developed? Given the flexibility of the Internet, at any one time retailers could be introducing Web sites, and existing sites may be updated to develop a new offering.
The dynamics of this medium thus cloud the true picture. The current and future potential of the Internet Since businesses have been permitted to use the Internet, speculation about its potential as a new marketing channel have been rife. Field (1996) reported Internet commerce to be worth IJS$518 million in 1996 and forecasted to reach IJS$6. 6 billion by 2000. According to Pyle (1996) ‘ ‘its global connectivity opens up new avenues for business n a manner that traditional commerce conduits cannot match”.
Yet concerns about lack of security (Bhimani, 1996), payment methods (Panurach, 1996), access restrictions (Hoffman and Novak, 1998) and various technological restrictions (Bell and Gemmell, 1996) have limited the Internet’s succession over traditional marketing tools. While few dispute the increase in the number of computers connected to the Internet, its actual size is somewhat indistinct, having been measured in terms of registered domains since 1986 when there were 213 recorded domains and by 1998 there were 29,650,000 (Network Wizzards, 1998).
The number of end-users or otential online customers is even less defined, as the techniques for measuring the on-line market are currently considered to be ‘an inexact” art form (Nua Ltd. , 1997) and can only be indicative of a growing market of end-users. Businesses have also Retailer adoption of the Internet 957 European Journal of Marketing 34,8 958 companies are planning to offer Internet-based services ‘in the near future” (Matthews, 1997).
Until more reliable data are forthcoming, only vague predictions are available to speculate about the Internet’s commercial potential, or electronic commerce as it is generically termed (Whinston, 1997). A number of academic studies have focused on the commercial adoption of the Internet (O’Keefe et al. , 1998; Hoffman et al. , 1996; Cappel and Myerscough, 1996; Cockburn and Wilson, 1996; Auger and Gallaugher, 1997; Spiller and Loshe, 1997; Griffith and Krampf, 1998; Jones and Biasiotto, 1999) but provide limited contribution to the structure or scale of adoption.
Moreover, previous research has tended to focus only on active Web sites or those sites having a high global/brand presence. Indeed, the majority of research has concentrated on the highest areas of activity in the US, for xample the top 100 US retailers, (Morganosky, 1997), yet little evidence exists to suggest whether European (KPMG, 1996) or I-JK retailers as a sector will adopt the Internet to the same extent. The studies may have identified that there are variations in how companies use the Internet for marketing, but it is not possible to suggest implications of the Internet for retail marketing by examining only on-line companies.
Established companies may not automatically adopt electronic commerce and thus their approach to the new technology will impact on its future penetration. There is s much if not more to be learned from successful companies’ absence from the Internet, as there is to be learned from those currently involved. Research aims The primary aims of the research were to develop an understanding of how the Internet is used by retailers to interact with consumers and to explore the extent of its adoption within the I-JK retail sector.
The secondary aims were to develop a reliable means of surveying the on-line vendor/consumer dialectic and to establish a clear point of reference for current and future research by quantifying levels of retailer Internet adoption from a representative sample at a given time in history. Previous surveys involving Web site reviews often used convenience samples, typically only targeting high profile or highly active on-line retailers. Thus, an on-line survey, targeting all the UK’s leading retailers, was initiated to assess objectively current levels of Internet activity.
More specifically, the survey had the following distinct research objectives: . To explore the general extent of Internet adoption by UK retailers and to investigate whether this is influenced by the type of retail activity undertaken.. To examine the types of on-line marketing activities and services being offered by major retailers from information services to on-line sales.. To investigate whether the extent of Internet adoption by retailers is dependent upon the size of retail operation, in terms of the number of retail outlets.
The next section of this paper presents the methods by which these objectives were explored and follows with a discussion of the findings. Research method An obvious organisations is a direct review of their Web sites, identifying the range of functions and services currently being offered. The rest of this section discusses the design and implementation of the on-line survey. Design of a pro forma data collection document To facilitate the accurate and consistent capture of information pertaining to the range of functions and services being offered through retailers’ Web sites, a pro forma Web site assessment document was created.
The content of this document, in terms of Web site functionality, was established by thoroughly reviewing the Web sites often retailers with a high profile Internet presence. A classification of Internet functions was ultimately developed, which was subdivided nto the following three primary categories: (1) Registration. The presence of a registered ‘uniform resource location” (URL) was interpreted as an indication of on- line activity.
However, registration of a URL does not necessitate instant development of the Web site, thus ‘ ‘ registration” implied the first stage of Internet development. (2) Information provision. Many Web sites use the Internet as a communication channel to provide existing and potential customers with various types of information. ‘ ‘Information provision” was thus categorized as the next stage of Internet Web site development. It should be noted that the key objective was to explore the type of information being presented rather than the depth or quality of information. 3) Interactivity. The pre-testing exercise had confirmed that the Internet is also used interactively for two-way communication and transactions, with the main method of on-line communication being e-mail. Some of the sites reviewed also provided on-line ordering and payment systems, although these varied greatly with respect to geographical limitations, merchandise ranges and levels of security. Interactive services, therefore indicated a third stage of Internet Web site evelopment to be explored through the on-line survey.
In addition to classifying a wide range of generic Internet functions and services, the pro forma document was also used to collate information about each retailer’s area of activity (grocery, electrical, clothing, etc. ). To this end, a retail activity categorisation scheme was created by adapting the classifications used by the Business Monitor (1996) and the Corporate Intelligence on Retailing Survey (1995). However, to reflect the increasingly global nature of Internet retailing, the American Standard Industrial Retailer adoption of the Internet 959
European Journal of Marketing 34,8 960 Classification (Levy and Weitz, 1995) was also consulted. Product type and assortment, size of operation, the nature of retail activity and ownership were also considered in arriving at the final categorisation, as illustrated in Table l. Finally, the pro forma data capture document was also used to record retailer size in terms of the number of retail outlets, in order to explore whether levels of Internet activity varied according to the scale of the organisation. The resultant document was then pre-tested on another 20 retail Web sites to test for content validity.
Targeting of research instrument To fulfil the primary objective of the research, it was necessary to population. The Corporate Intelligence on Retailing survey (1995) was initially used as the data source, as it claimed to include ‘ ‘all of the largest retail organisations, as well as those, which are significant because they are perceived as innovative, or hold a powerful position within a niche Using this document it was possible to identify and review the Web sites of 330 of the UK’s leading retail organisations.
The Domain Name System (DNS) has evolved as a way of ensuring no two Internet sites have the same name (Albitz and Liu, 1997). For the United Kingdom the top-level domain name is ‘ . uk”, a node one the top-level US domain ”. com”) and a sub domain is likely to be a company name. The data were collected using the following three stage procedure: (1) Presence of a registered URL. The on-line survey commenced by determining whether each of the 1,099 target companies had a registered URL, using the on-line search tool ‘ which has been specifically designed to find domain names. (2) Web site activity validation. Each registered URL was checked for site activity.
If there was an error message indicating an incorrect address then the spelling was rigorously examined and if it still failed to connect to a Web site the URL was assumed to be inactive. Where some sites had pointers to Sites coming soon, these sites were also recorded as inactive. If the URL pointed to a Web page which downloaded with clear indication that it was a retail company site ?± i. e. company logo, name or address detail ?± then the site was recorded as active however extensive or limited the information content. (3) Web site review procedure: Each active Web site was reviewed and the content recorded using the pro forma document as a review guide.
Retailer adoption of the Internet 961 European Journal of Marketing 34,8 962 The above method was used to ensure that the collection of data was conducted in a igorous and consistent manner. The data capture exercise was conducted during a six-month period ending December 1997. Research results The following presentation of the research results is explicitly related to the three research objectives proposed earlier in this paper. The relationship between retail sector and internet adoption Table II presents an analysis of Internet adoption, based on the three primary classes of Internet activity: registration; information provision; and the availability of interactive services. Generally, the on-line survey showed that Internet enetration in the I-JK retail sector has been limited.
Only 31 per cent of the retailers surveyed had a registered URL and consequently the majority of organisations (69 per cent) had no Internet presence. Furthermore, only ten per cent of retailers had progressed to develop live Web sites providing either informational or interactive services. The discrepancy between registered URLs (31 per cent) and active Web sites (10 per cent) echoed previous US findings of a 27 per cent discrepancy. This was partly attributed to retailers ‘developing, updating and/or redoing the site” and the rge to secure a domain name without necessarily developing on-line activity (Morganosky 1997). The retailers’ reticence in developing Web sites, could also be attributed to a lack of Internet strategy, and/or available resources for its implementation.
Other sectors in which the level of registrations was high included electrical goods, toys and DIY, all of which had registration rates of about 5 per cent. Retailer categories with less than a 30 per cent registration level included convenience stores, footwear, mixed retail stores, news, and health and beauty. (2) Informational services. The largest concentration of retailers providing informational services was the electrical sector with 33 per cent. Grocery retailers had 31 per cent and toys 24 per cent. All of the other sectors had less than 20 per cent with the lowest recorded percentage being specialist food retailers with Just 1. 5 percent having Web sites offering informational services. (3) Interactive services.
The number of interactive sites was marginally fewer than the number of sites providing information, and the distribution of these services across the 17 retail sectors was very similar to the results for informational services. Having provided an overview of the level of Internet adoption within the 17 retail sectors it as also possible to evaluate whether there is a statistically significant relationship between retail sector and adoption. A chi-squared test for goodness of fit was conducted, and the results are presented in Table Ill (the numbers in upright type in the table are actual values, and those shown in italics are the expected values for each cell). The data in Table Ill indicates that there are a number of expected values of less than 5, which can be problematic when using a chi-squared test.
However, as the number of such instances is within the parameters suggested for large cross- abulations (Chatfield, 1978, p. 150), this method of analysis is valid in these circumstances. The results of the chi-squared analysis indicate that there is a highly significant relationship (at the 0. 001 level) between retailer category and the level of Internet adoption. More specifically, an inspection of the actual and expected values confirms that those organisations operating in the grocery, electrical and mail order sectors are most likely to have registered a URL and developed a transactional site, whilst those operating in the speciality foods and health and beauty sectors are least ikely.