The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0956-4233.htm

IJSIM 13,5

432

E-CRM Web service attributes as determinants of customer satisfaction with retail Web sites Richard Feinberg and Rajesh Kadam

Purdue University, West Lafayette, Indiana, USA Keywords E-commerce, Internet, Customer satisfaction, Service quality Abstract Business is moving online, not as a matter of choice, but as a matter of necessity. The use of the Internet as a channel for commerce and information presents an opportunity for business to use the Internet as a tool for customer relationship management (CRM)/(e-CRM). Despite widespread agreement that CRM/e-CRM has direct and/or indirect impact on customer satisfaction, sales, profit, and loyalty, the significance of e-CRM and the various e-CRM features in influencing customer satisfaction has not been well researched. This study attempted to uncover relationships between e-CRM and customer satisfaction by determining the presence of e-CRM features on retail Web sites for which we have customer satisfaction data, and determining if the amount of e-CRM is related to customer satisfaction or which, if any, of the various features of e-CRM are related to customer satisfaction. It was found that retailers differ in the presence of the 42 different e-CRM features; that there is a positive relationship between the amount of e-CRM on a Web site and customer satisfaction with the Web site; and that not all e-CRM attributes are equal ± some are related to satisfaction and some are not. There was no relationship between the level of e-CRM on a retail Web site and sales and profit.

International Journal of Service Industry Management, Vol. 13 No. 5, 2002, pp. 432-451. # MCB UP Limited, 0956-4233 DOI 10.1108/09564230210447922

Introduction Customer relationship management (CRM) is a comprehensive business and marketing strategy that integrates technology, process, and all business activities around the customer (Anton, 1996; Anton and Hoeck, 2002). It is mostly defined in terms of the acquisition and retention of customers and the resulting profitability (Menconi, 2000; Nykamp, 2001). Effective CRM is assumed to lead to bottom-line benefits for the organization (Anton and Hoeck, 2002; Connelly and Yoger, 2001; Cusack, 1998; Rust and Zahorik, 1993; Swift, 2001; Tschohl, 2001). According to Bain and Company, profits increase by 2580 percent when customer retention rates increase by five points. The Internet has provided a platform to deliver CRM functions on the Web (e-CRM), thus as business moves to the Web, e-CRM will move to center stage. Despite widespread agreement that CRM and e-CRM has direct and indirect impact on customer satisfaction, sales, profit, and loyalty (Anton and Hoeck, 2002; Connelly and Yoger, 2001; Cusack, 1998; Rust and Zahorik, 1993; Sterne, 1996; Swift, 2001; Tschohl, 2001) the significance of e-CRM and the various e-CRM features in influencing customer satisfaction has not been well researched. This study attempted to uncover relationships between e-CRM and customer satisfaction by determining the presence of e-CRM features on retail

Web sites for which we have customer satisfaction data and determining if a Customer company's amount of e-CRM is related to customer satisfaction or which, if satisfaction with any of the various features of e-CRM are related to customer satisfaction. retail Web sites Determination of those e-CRM factors that are related to customer satisfaction will allow companies to focus on developing on their Web site only those aspects of e-CRM that are related to customer satisfaction. This could mean 433 reduced costs, since elements that are not relevant need not be included anymore in the e-CRM implementation. This also means (potentially) more customers, more sales, more profit, and more loyalty, as resources are focused on those aspects of e-CRM that work. CRM is a dominant point of discussion in business (in July 2002, Amazon.com listed 90 current books on CRM). Most/many companies have or will attempt a CRM implementation . . . at considerable cost (it is estimated that 45 percent of all companies have CRM applications and 37 percent have applications in implementation and planning stages (Leon, 2001)). According to a survey conducted by Infoworld, 77 percent of the business respondents considered CRM critical (Apicella, 2001). The CRM industry is believed to be a ten billion dollar industry growing at a 27 percent annual rate (iBiz STATs, 2001). There is little empirical evidence that CRM is related to profit, sales, or customer satisfaction (for some of the available evidence, see Rust and Zahorik, 1993; Swift, 2001). Vendors abound with sophisticated and expensive technologies delivering (or promising to deliver) customer relationships (e.g. Siebel (www.siebel.com); Covergys (www.convergys.com); Genesys (www.genesyslab.com); and literally hundreds of others). These systems have a lot of technical attributes/functions that have come to define CRM but, interestingly, they do not have an emphasis on customer satisfaction measurement that must be at the core of any CRM implementation. According to Forrester Research, only about 40 percent of CRM implementations are successful (see also Feinberg and Trotter (2001) who predict that there are even higher failure levels). Although it may be universally believed that e-CRM and CRM applications are crucial, it is, at the same time, clear that not everyone is successful in implementing CRM or even if CRM is related to customer satisfaction or sales/profit. These points sharply etch the need to better understand if CRM is related to customer satisfaction, sales/profit and, which, if any, aspects of e-CRM (for the specific purposes in this study of retail Web sites) are important to these variables. By understanding what part, or parts, of CRM/e-CRM create satisfaction, management of the e-CRM function becomes more efficient and more effective. E-CRM in the retail industry The retail industry appears to be a logical place to examine the relationship between e-CRM and customer satisfaction (GoÂmez, 2001). Retailers are at the front end of the supply chain and consequently are the ``interface'' from which consumers make their purchases. This suggests that the retail industry is likely

IJSIM 13,5

434

to be ahead of other industries in understanding CRM. We know that 100 percent of the top retailers have retail Web sites with at least some e-CRM features (Feinberg et al., 2002, in press). Retailing is also at the forefront of e-commerce. According to the US Census Bureau, e-retail sales in 2000 amounted to $32.6 billion, up 19.3 percent from 2000. While online sales accounted for only 1.2 percent of all retail transactions in 2001 (Ewalt, 2002), it is clear that e-retailing will continue to grow and become a more significant portion of retail sales than it currently is. The use of retail Web sites by consumers for purchase and information gathering (unpublished research at the Center for Customer Driven Quality indicated that consumers are actually more likely to use the Internet as an information search tool and then go to stores to purchase, then make a purchase itself) suggests that these Web sites will be the essential e-CRM tool (not just an essential tool) for all retailers (Internet only, multi-channel, store only) in the future. This study improves on prior research in three ways after the fact that there is very little empirical work on e-CRM in the first place. First, we study the relationship between e-CRM features and customer satisfaction, using a sample of real retail Web sites and actual customer satisfaction data from consumers who have purchased from those sites. Past research makes assumptions about what is important to the consumer, asks consumers if that factor is important, and then when 1 or 3 (or whatever number) of those factors are statistically significant statements are made about the importance of the factor (e.g. Szymanski and Hise, 2000; Yoo and Donthu, 2001). Instead of using a small selected sample of fake or constructed retail Web sites the e-CRM attributes are coded from a large set of real retail Web sites. Having this ``real'' set of retailers allows us to find a set of them for which sales and profit data exist so that in addition to looking at the relationship between e-CRM and satisfaction we can look at e-CRM and profits/sales. Finally, in past studies the dependent variables were perceived satisfaction with a Web site. In contrast, the satisfaction data used in this study is obtained from Nielsen/NetRatings obtained from a national panel of online users of those sites. In this study, we take customer satisfaction data from a large set of ``real'' Internet retailers coded for a large set of e-CRM attributes and do the following: . Describe the state of e-CRM in retailing. What factors represent what retailers are doing on the Web? . Assess the degree to which e-CRM (in total) is related to satisfaction. . Assess the degree to which any one of the sets of e-CRM attributes are related to satisfaction. . Assess the degree to which e-CRM is related to sales and/or profit. Identification of a census of e-CRM attributes The present study uses 25 features first identified by Anton and Postmus (1999) as defining e-CRM with an additional 17 features defined by Feinberg

et al. (2002, in press) in their analysis and study of e-CRM in retailing. Table I Customer shows the complete list of features. satisfaction with The features are described below: retail Web sites . Site customization. The Internet brings enormous amounts of information to the desktop. While this is an advantage, it can also be an enormous disadvantage, since the user may not be able to readily access 435 the information he needs. Therefore, sites offer customization features allowing the user to filter the content they see. Thus, if the user, on the first visit, customizes the site to suit his tastes and preferences, he will see the customized content on subsequent visits. (Example: www.yahoo.com; www.my.yahoo.com) . Alternative channels. These are the different ways to contact the company. For instance: e-mail, fax, toll-free numbers, postal address, call-back button, voice over IP and bulletin board. Traditionally, only toll-free numbers and postal addresses could be used by customers to reach the company. However, over the past few years, e-mail has emerged as an important tool for company-customer communication. According to the Mailbox Report from United Messaging, the total

E-CRM features 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21.

Complaining ability Privacy policy Product Information online Product highlights Preview product Site map E-mail Purchase conditions Customer service area About company Local search Problem solving Cross sell/upsell/addon sale Online purchasing Check out Info first time users Membership Mailing list Prod. customization Your account info FAQ

% of companies with feature E-CRM features 99 99 98 97 97 96 96 96 96 94 94 90 90 90 90 89 82 81 79 79 75

22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42.

1-800 Track order status External links Member benefits Spare parts ordering On sale area Quick order ability Site customizing Postal address Order within 3 clicks Domain fault repair Find stores Gift certificate Fax Request catalog AFFINITY program Chat Bulletin board Site tour VoIP Call back button

% of companies with feature 75 74 66 66 66 65 64 62 54 49 48 35 33 29 28 21 12 10 6 2 0

Note: Table shows the number of retailers having feature on Web site Source: Nos 1 to 26 ± Anton and Postmus (1999); Nos 27 to 42 ± additional features identified in Feinberg et al. (2002, in press)

Table I. List of e-CRM features

IJSIM 13,5

436

.

.

.

.

.

.

.

.

number of e-mail boxes increased by more than 60 percent from the end of 1999 to the end of 2000. There were 891.1 million mailboxes worldwide at the end of 2000 (Jackson, 2001). It offers opportunities for improving customer service and reducing costs. Unlike telephone communication, which is synchronous, e-mail communication is asynchronous. This reduces costs, since customer service queries will now be handled in countries with lower labor costs, by taking advantage of the time-difference. Also, e-mail transmission happens on the Internet, which is essentially a public service and thus the company does not incur significant transmission costs. Local search engine. This feature allows the visitor to search on key words to quickly locate the required content on the Web site. This feature is helpful for those site visitors who are looking for a specific piece of information. Membership. The visitor can request a password. With this password he can continue surfing on password-protected Web pages within the Web site. This feature allows the company to collect personal information from users, when the user registers for the membership. It also allows them to track the customer's behavior at the site over time. This knowledge allows the business to assess which customers are worth retaining by looking at current and prospective customer profit and customer defect patterns (Swift, 2001). Mailing list. To receive more information, the visitor can add his/her e-mail address to a list to receive automated e-mails. Often, this is called a newsletter. This feature allows the company to build a database of e-mail addresses of potential users of the company's product or service. Site tour. The visitor can follow a tour through the Web site. This allows users to get familiar with the Web site contents. Site map. This is a hierarchical diagram of the pages on the Web site, also called a site overview, site index, or site map. This feature is helpful in understanding the general structure of the Web site. Introduction for first-time users. Visitors, who enter the site for the first time, can surf to an introduction page containing information about ``How to use the site most efficiently''. This feature could help attract prospects and facilitate a first-time purchase. Chat. This feature allows a visitor to chat real-time with others. Specifically, this could mean chatting other visitors on the site or with the customer service personnel. This feature, while not used widely yet, is likely to see a marked increase in use, as Internet penetration increases and bandwidth constraints decrease globally. Electronic bulletin board. Script-driven forums allow visitors to share information with others and can help shape a Web site to better serve

.

.

.

.

.

.

.

.

.

.

.

the customer needs. A visitor can post a message or can respond to a Customer posted message on a special Web page. This interaction, over time, satisfaction with creates a community of users around the company's service or product. retail Web sites Many companies have successfully used bulletin boards. For example, Dell has a highly successful bulletin board where the customers post and respond to queries. In addition, Dell service personnel continuously 437 keep track of the bulletin board, allowing them to keep abreast of the problems that customers are encountering with their products. Online purchasing. Visitors are able to purchase services or products online. This feature is probably the most critical part of the Web site. Product information online. Visitors can read product information on the Web site. This feature is critical since Web surfers gather product information on the Web and subsequently buy the product from a brickand-mortar store. Customization possibilities. Visitors can customize their service or product online before ordering. For instance, a visitor may want to assemble online a PC of a configuration that is not listed on the Web site. Or a jeans pant buyer can design a pant that fits his or her body shape exactly. Purchase conditions. The purchase and contractual conditions can be viewed online. Purchase conditions contain shipping policies, return policies, warranty, guarantee and other company commitments. Preview product. The (customized) product can be viewed before purchasing. The product can be viewed in a motion picture or a demo. External links. The visitor can easily and seamlessly link to complementary products from other companies. FAQs. Frequent asked questions and their answers are available for reading. This feature acts as self-help for customers looking for answers to their queries. Like other information on the Web site, this could potentially help in reducing contact center traffic. Problem solving. Customers can solve problems with products or services themselves with online self-help routines. This feature is not very prevalent since customers show resistance to using this feature. Complaining ability. Complaints and problems can be detailed online. The Web site has a specific area for customers to log in their complaints and get action. Spare parts. It is possible to order spare parts and complementary products online. In addition to repeat customer service, this feature ensures repeat traffic to the site. Affinity program. Affiliations with philanthropic agencies or organizations.

IJSIM 13,5

438

.

. .

.

.

. .

.

.

. .

.

Product highlights. The benefits of a particular products/services are highlighted. This feature allows the company to highlight products or services that may be relevant in a particular context. The context could be a particular festival or a season, among others. This feature also prompts repeat traffic. Request for catalog. Allows the user to request a catalog. Quick order ability. This feature basically allows the user to check out the product within three-clicks. Amazon.com pioneered this feature. Because of intellectual property issues, other sites use variants of Amazon's 3-click process. Ease of check out. This is a subjective rating of the ability to check out on an ``ease'' scale of 1 ˆ easy to 4 ˆ difficult. This parameter is important, since significant numbers of purchases are abandoned at the time of checkout. A Jupiter consumer survey found that 66 percent of consumers have abandoned a purchase while in the process of ordering a product online. This compares poorly with the 2-3 percent abandonment rates for brick-and-mortar retailers. Part of the reason for the abandonment is site performance and layout: 10 percent abandoned the process because the site was too slow, while 6 percent abandoned because the site crashed during the process and 7 percent said that the order forms were too difficult to complete (Blank and Daniels, 2001). Ability to track order status. This feature allows the user to find out which stage of the shipping process his/her order is in. Customers are more likely to feel satisfied if they know of the status of their order than if they do not. According to a Forrester Survey, 58 percent of customers use customer service to find out when they can expect delivery of their order. Gift certificate purchase. Store locator. This feature is helpful since customers gather product information on the Web and subsequently buy the product from a brickand-mortar store. On-sale area. This is a highlighted place on home page highlighting sale item(s). Member benefits. Description of benefits of shopping or of being a site member. Order. This feature allows the user to place an order within three clicks. Speed of download page. The Web site was considered fast if the Web pages downloaded in less than 15 seconds. Account information. If the user had registered on the site, the feature allowed him to view his personal information that he has given to the site. This could include his credit card information, if he has authorized the Web site to retain this information for repeat purchases.

.

.

.

Customer service page. These pages give details on contacting customer Customer service. satisfaction with Company history/profile. These Web pages give the company retail Web sites background. In the case of retail shopping sites, there will be a link to the corporate Web site. Posted privacy policy. Privacy concerns are coming to the fore with 439 increased Internet penetration. Today, most sites have a privacy policy and post it on their Web site. This not only assures the customer that his information is protected, but will also protect the company from lawsuits from perceived privacy violations.

Hypothesis development These hypotheses were tested in two parts. In the first part, Study 1, the presence (or absence) of the 42 possible e-CRM service features on the retail Web sites used as the sample in this study was assessed. The most and least available (popular) e-CRM features might represent those features that the industry believes to be important. Based on the findings from Feinberg et al. (2002, in press), it is predicted that: H1. Retailers differ in the presence (absence) of the 42 e-CRM features. Feinberg et al. (2002, in press) assessed the state of e-CRM in retailing by analyzing the sites of the top 100 mass merchants, specialty stores, and Internet-only retailers. In this study, the sample of retailers assessed is broader and wider, consisting of both larger and smaller retailers across a broader range of types of retailers. It is expected that the availability of e-CRM features will vary across retailers. As was found in Feinberg et al. (2002, in press), the most popular features are expected to be e-mail, 1-800 number, privacy policy, postal address, membership, product information online, preview product, online purchasing, store locator, purchase conditions, product highlights, customer service area, download speed and company profile on site. In the second part, Study 2, e-CRM features were related to customer satisfaction. The amount of e-CRM (amount being defined as the number of e-CRM attributes present on a Web site ± from 0 to 42 features) should be related to customer satisfaction. Therefore, we predict that: H2. The amount of e-CRM (an index score created by adding the number of e-CRM factors present ± score can range from 0-42) is related to customer satisfaction. Based on findings from Feinberg et al. (2002, in press), we also predict that: H3. Only a small set of e-CRM features is related to customer satisfaction. In Feinberg et al. (2002, in press), only chat feature, spare parts availability, gift certificate purchase, mailing address, search engine, links and company profile were related to customer satisfaction.

IJSIM 13,5

440

H4. If we assume total e-CRM is related to satisfaction and satisfaction is related to sales and profit then total e-CRM should be related to sales and profit. In Study 1, retail sites were content analyzed on the 42 factors that were identified as defining e-CRM. The sample of retailers for Study 1 was the set of 271 retailers tracked by Nielsen/NetRatings (the data was made available for this study by Nielsen/NetRatings) (see Table II). These retailers represent a broad array of retail types and sizes (from Wal-Mart, the largest retailer in the world, with annual revenues of 217 billion, to Bookpool LLC Wordsworth Books, with annual Internet sales of $750,000). These Web sites were chosen because they represent the census of retail Web sites tracked by Nielson/ NetRatings panel data on customer satisfaction. Two independent raters examined the Web sites of the 271 retailers for the presence and absence of the 42 e-CRM attributes. The two independent raters were trained to recognize the factors identified as e-CRM factors on retail sites not in the sample until they achieved >95 percent agreement independently. Disagreements after training were resolved with a discussion between raters. The second study was designed to see if any of the e-CRM features identified in Study 1 determine consumer satisfaction. To accomplish this, customer satisfaction ratings were obtained from Nielsen/NetRatings. The sample for Study 2 was the set of 271 retailers tracked by Nielsen/Netratings on a continuous basis. Nielsen/NetRatings (www.nielsen-netratings.com) is the largest research firm focusing on Internet audience measurement and analysis. Internet user behavior (click behavior) is monitored by a real-time meter installed on individuals' computers worldwide, both at home and work. Nielsen/NetRatings is affiliated to Nielsen Media Research, a leading televisionaudience measurement company, and ACNielsen, the global leader in providing market research and analysis to the consumer products and services industries. Nielsen/NetRatings collected the Internet usage data and follow-up survey work from a representative consumer panel of Internet users who purchased from the different Web sites. The satisfaction number represents mean level of overall satisfaction with a particular site among respondents who seriously considered or made a purchase from that site (1 ˆ ``completely dissatisfied'', 10 ˆ ``completely satisfied''). The Nielsen/NetRatings satisfaction data for the sample of retailers is available in Table II. Results Presence of e-CRM features As expected, (H1) retailers vary in the presence of the various e-CRM features (see Table I). The most common e-CRM features (>90 percent of retailers having them) were: complaining ability, privacy policy, product information online, product highlights, preview product, site map, e-mail, purchase conditions, customer service area, about company, local search, problem solving, cross-sell/up-sell/add-on-sale, online purchasing and check out. The

Company 123greetings.com 1-800 Contacts, Inc. 1-800-Flowers 1bookstreet, Inc. 3Com Corporation ABC distributing, inc. Abercrombie & Fitch ADB Systems International, Inc. (bid.com) Adobe Air Tran Airways Alaska Airlines (Alaska Air Group) Alloy, Inc. AltaVista Company (CMGI) Amazon.com Amer Group Plc. (Wilson Sporting Goods Co.) America online America West Airlines American Airlines American Eagle Outfitters American Express American Greetings Amtrak Ann Taylor Retail, Inc. Apple ARTISTdirect, Inc. Ashford ATomShockwave Corp. (AtomFilms.com) Auction.com Inc. (Nimbus) Audio Book Club (MediaBay) Avon BarnesandNoble.com Beaniemania Bertelsmann AG (Bookspan) Best Buy Best Fares USA, Inc. Beyond.com Corporation Bigwords.com Blockbuster Blue Mountain (American Greetings.com Inc.) Bluefly, Inc. BMG Entertainment BONZI.COM Software, Inc. Book-of-the-Month Club, Inc. Bookpool LLC Bookspan ± Quality Paperback Books Borders Bose Corporation Broderbund LLC

Mean satisfaction score, overall (1-10 scale) 7.9 7.54 7.93 7.91 7.21 8.1 7.86 7.03 7.99 7.97 7.76 8.63 8 8.52 8.01 7.33 7.89 7.54 8.49 7.81 8.26 7.52 7.93 8.35 8.5 8.14 8.78 6.98 7.37 7.79 8.19 7.06 7.84 7.6 7.54 7.03 8.33 8.3 8.67 7.65 7.79 7.11 7.94 7.88 7.76 8.1 7.59 7.77

E-CRM score 21 31 31 32 25 34 20 22 27 29 31 35 22 38 16 21 31 29 29 34 24 29 38 33 27 32 25 25 30 28 35 11 30 33 30 15 9 31 24 31 27 32 31 29 27 39 27 35 (continued)

Customer satisfaction with retail Web sites 441

Table II. E-CRM score and satisfaction with retail Web sites tracked by Neilsen/Netratings

IJSIM 13,5

442

Table II.

Company Brookstone Company, Inc. Burpee Seeds Buy.com Cabela's Inc. Campmor, Inc. CD Now CD Universe CDW Chadwicks (Brylane) Cheap Tickets Christianbook.com Circuit City City Auction Clairol, Inc. (Bristol Myers Squibb) Classic Fragrances, Limited Clinique (Estee Lauder) CNET CNN/Sports Illustrated (GolfOnline.com) (CNN News Group) Coldwater Creek CollectingNation.com, Inc. Columbia House Comp USA Compaq Conde Nast Inc. (Epicurious.com) Continental Airlines Cooking.com Costco Wholesale Corporation Crate & Barrel (Euromarket, Inc.) Crutchfield CVS Corporation Cyberian Outpost, Inc. Damark.com (Provell) Delia's Corp. Dell Computer Corporation Delta Disney Online Doubleday Book Club (Random House) Doubleday Direct, Inc. (Crossings.com) (Barnes and Noble) Drugstore.com, Inc. Drugstore.com, Inc. (Beauty.com) Eastbay eBay Inc. EBWorld.com, Inc. ecampus.com Eckerd Corporation eCost.com (PC Mall Inc.)

Mean satisfaction score, overall (1-10 scale)

E-CRM score

7.6 8.17 8.21 8.15 8.37 8.43 8.04 7.85 8.04 7.2 8.4 7.43 7.03 8.06 7.66 8.41 8.06

36 22 33 36 38 32 33 31 31 28 33 29 24 21 25 31 16

7.48 8.52 8.18 7.68 7.16 7.47 8.16 7.97 8.21 7.65 8 7.87 8.15 8.35 7.31 8.25 7.75 7.73 7.98 8.32

11 34 26 31 33 25 21 28 31 28 37 33 34 26 30 31 35 29 23 26

8.44 8.52 7.82 7.98 8.34 7.81 7.55 7.06 6.92

29 29 35 28 33 33 28 34 31 (continued)

Company Eddie Bauer (Now part of Speigel) Egghead.com Egreetings Network, Inc. (americangreetings.com) Electronic Arts ESPN Estyle, Inc. E-Toys Expedia Travel FAO Schwarz Federated (Bloomingdale's) Federated Department Stores ± Macy's Fingerhut (Federated dept. Stores) Fingerhut (PCFlowers.com) (Federated) Foot Locker Fossil, Inc. FOX Interactive Television, LLC (20th Century Fox Film Corporation) FragranceNet.com, Inc. Frederick's of Hollywood FTD.com Gap (Gap Inc.) Gap, Inc. ± Banana Republic Gateway Gateway ± NECX Direct GiftCertificates.com Global Sports Interactive (Fogdog.com) Gloss.com, LLC Godiva (Campbell Soup Company) GORP Greyhound Lines, Inc. Gurney Seed and Nursery Co. (MySeasons.com) Half.com by eBay Hallmark Handspring, Inc. Hanover Direct, Inc. (Domestications.com) Hertz Hewlett-Packard Company Hifi.com (Cambridge Soundworks, Inc.) Home and Garden Television Home Depot Homestore.com, Inc. Hot New Products, Inc. (FitnessZone.com) Hotel Reservations Network, Inc. (Hotel Discounts.com) (Hotels.com) HSN Interactive LLC (USA Interactive) ibeauty.com IDEAL International, Inc. (DealDeal.com)

Mean satisfaction score, overall (1-10 scale)

E-CRM score

8.18 7.81

35 37

8.54 8.25 7.58 7.92 8.2 8.09 6.66 7.44 7.32 8.01 7.28 8.46 8.1

29 27 33 27 32 31 33 26 32 31 32 32 32

8.32 8.14 7.97 7.68 8.19 7.72 7.68 7.79 8.51 7.54 6.73 8.38 7.88 7.76 8.2 8.26 8.31 7.94 7.68 7.73 7.59 7.94 8.1 7.63 7.5 7.18

28 29 31 30 33 33 33 31 33 30 29 32 32 30 31 31 26 29 32 24 26 36 31 31 30 32

7.49 8.12 7.87 8.07

32 32 26 25 (continued)

Customer satisfaction with retail Web sites 443

Table II.

IJSIM 13,5

444

Table II.

Company IdeaMall, Inc. (MacMall.com) iGo Corporation IKEA Insight Enterprises, Inc. International Business Machines Corporation Intuit iVillage Inc. (HomeArts.com) J & R Electronics, Inc. J.Crew J.Jill JCPenney JustBalls, Inc. K-B Toys Kmart Corporation Lancome (L'oreal) Lands' End Lane Bryant (Brylane Inc.) LEGO Company Levi Strauss and Co. LL Bean Lowe's Lowestfare.com Macromedia, Inc. Marriott Martha Stewart Living Mary Kay Inc. Mattel, Inc. (Barbie.com) Maybelline, Inc. McAfee Merck & Co., Inc. ± Merck-Medco Micro Warehouse, Inc. MicronPC, LLC Microsoft Moet Hennessy Louis Vuitton ± Sephora More.com MP3.com Neiman Marcus Net Central Inc. ± Books-A-Million NetFlix Netscape Newport News NFL Enterprises NHL Enterprises, L.P. NIKE, Inc. Nordstrom Northwest Airlines Office Depot Office Depot-Computers4sure

Mean satisfaction score, overall (1-10 scale) 8.16 8.33 7.36 7.96 7.26 7.98 8.43 7.91 8.04 7.94 8.02 9.24 8.18 7.77 8.12 8.31 7.65 8.26 7.95 8.29 7.67 7.47 7.48 7.86 8.04 8.52 7.56 7.62 7.94 8.06 7.67 7.37 7.78 8.28 7.32 7.64 8.31 8.35 8.07 7.29 8.36 8.1 7.15 7.83 8.05 7.51 7.75 7.67

E-CRM score 33 31 19 26 31 24 19 32 33 37 30 32 32 33 32 31 19 34 20 37 30 31 31 27 27 27 7 22 27 34 30 21 34 36 26 33 34 30 31 28 25 30 32 31 37 22 31 34 (continued)

Company Office Max Old Navy (Gap Inc.) OneHanesPlace OneTravel.com, Inc. Overstock.com (Gear.com) Palm Inc. Panasonic (Matshushita Electric) Patagonia, Inc. PC Connection, Inc. PC Flowers & Gifts.com Inc. (Gifts.com) PC Mall, Inc. Pegasus Systems, Inc. ± Travel Web (Pegasus solutions, Inc.) Pier 1 Imports Planetrx.com, Inc. Powell's Books Priceline.com Proctor & Gamble ± Cover Girl Proflowers Quixtar (Alticor Inc.) QVC Radio Shack Ralph Lauren ± Polo Random House, Inc. RedEnvelope, Inc. Reflect.com (P&G) REI Revlon Consumer Products Corporation Richard Simmons Rosenbluth Intl. (BizTravel.com) Sears Sierra SkyMall, Inc. Smarter Kids (Excelligence Learning Corp.) Soccer 4 All Sony Southwest Airlines Sparks.com Spartan Stores (VirtualFlowers.com) Spencer Gifts Spiegel SportsLine.com SportsLine.com, Inc. (MVP.com) Staples Symantec TAEBO Target Corporation Textbooks.com

Mean satisfaction score, overall (1-10 scale)

E-CRM score

7.47 8.35 8.09 8.44 8.11 7.35 7.23 7.83 8.17 6.23 7.78

34 34 31 27 30 29 26 33 32 33 31

8.63 8.27 7.45 8.42 7.38 7.77 8.6 9.21 8.64 7.62 8.02 7.02 7.72 7.72 7.94 7.33 7.74 7.41 7.53 7.7 7.45 7.93 8.01 7.58 8.62 8.6 7.42 8.54 8.28 8.37 7.98 7.57 7.98 8.76 7.97 7.8

27 32 5 35 29 25 26 30 34 28 31 24 33 33 36 23 29 19 37 16 34 31 20 31 24 31 25 30 27 29 32 33 23 20 32 35 (continued)

Customer satisfaction with retail Web sites 445

Table II.

IJSIM 13,5

446

Table II.

Company The Limited, Inc. ± Express The Literary Guild The Sharper Image The Sports Authority The Vitamin Shoppe TigerDirect, Inc. (Systemax Inc.) Tower Records (MTS) Toys R Us Travel Online Travelocity Travelzoo Inc. Trilegiant Corporation (NetMarket.com) Trip.com (travelbuys.com) Tucows Inc. Ty Inc. U.S. Retail Flowers, Inc. (GeraldStevens.com) uBid United Airlines Universal Studios US Airways Varsity Group Inc. Viacom (iwon) Victoria's Secret (Intimate Brands Inc.) Videoflicks.com, Inc. Vitacost.com Walgreen Co. WalMart Williams Sonoma, Inc. ± Pottery Barn Williams-Sonoma, Inc. WinZip Computing, Inc. Wizards of the Coast, Inc. (Hasbro Inc.) WordsWorth Books X10 Yahoo! Inc. Zappos.com ZD Net Zones Inc. ± Mac Zone

Mean satisfaction score, overall (1-10 scale)

E-CRM score

7.69 8.07 7.72 7.56 7.79 8.01 8.03 7.78 7.69 7.94 8.06 6.56 7.81 7.78 7.91 7.99 7.7 7.76 7.76 7.66 7.98 8.27 7.98 7.41 7.85 8.05 7.93 7.96 8.21 7.66 9.17 7.31 7.62 7.91 8.33 7.99 7.77

18 27 33 34 35 33 31 39 27 30 28 36 31 23 29 28 24 30 23 25 32 32 34 27 32 32 32 34 33 19 33 29 35 30 34 22 33

least popular features (<15 percent of the sites) were chat, bulletin board, sitetour, VoIP and call-back button. Relationship between e-CRM features and satisfaction The presence of an e-CRM attribute does not speak of its importance or usefulness. Study 2 was designed to find those e-CRM attributes that may be related to customer satisfaction. H2 was about the relationship between e-CRM and customer satisfaction. There was a statistically significant positive relationship between the number

of e-CRM attributes present (an index score created by simply adding up the Customer number of e-CRM features on a retail site (0-42) and customer satisfaction, satisfaction with r…n ˆ 271† ˆ 0:14, p < 0:05†. Thus, as e-CRM increases on a site, the greater retail Web sites the customer satisfaction will be. The possible relationships between any particular e-CRM attribute and customer satisfaction was analyzed using Chi-square analyses on a merged 447 data set of retail e-CRM scores and Nielsen/NetRatings satisfaction scores. Customer satisfaction ratings was Nielsen/NetRatings panel's self-reported satisfaction with shopping on the Web site. The set of 271 retailers were divided into retailers with or without a particular feature (e.g. had ``e-mail'' versus ``did not have e-mail''). Each of these retailers was defined as being high or low in customer satisfaction or site traffic based on a median split of that category. Thus, to analyze the effect of e-mail on the Web site we would have a 2  2 contingency table of retailers who have or do not have e-mail and who are low and high in customer satisfaction (and site traffic). Chi-squares were completed on the e-CRM features and the customer satisfaction and customer traffic scores for each of the e-CRM attributes. H3, which stated that there are some e-CRM attributes related to customer satisfaction with the site and some that are not. We had hypothesized (based on a previous study of retail sites ± Feinberg et al. (2002, in press)) that chat feature, spare parts availability, gift certificate purchase, mailing address, search engine, links and company profile would be related to customer satisfaction. However, in this study, the significant relationships were between having mailing list, 2 …1† ˆ 3:96, p ˆ 0:05, quick order ability …2 …1† ˆ 7:48, p < 0:05†, gift certificate purchase …2 …1† ˆ 5:59, p < 0:05†, and affinity program …2 …1† ˆ 11:1, df ˆ 1, p < 0:05†, your account information …2 …1† ˆ 3:76, p < 0:05†, and customer satisfaction. Only the finding that availability of gift certificate purchase is related to customer satisfaction matched the finding from the Feinberg et al. (2002, in press) study. H4 concerned the relationship between e-CRM and profits or revenues. Pearson correlations indicated that total e-CRM was not related to profit or revenue (p > 0:05). Although it is commonly believed that e-CRM leads to customer satisfaction, which leads to sales and profit, the data here indicate that while e-CRM is related to satisfaction it is not related to sales and profit. Discussion and implications It is generally accepted that e-CRM/CRM is related to customer satisfaction, profit and sales. This research suggests that e-CRM is related to customer satisfaction with a retail site but not to sales and profit. The results also indicate that not all e-CRM is the same. If e-CRM is related to satisfaction, the relationship is not strong. Thus failure of CRM implementation may not be because the implementation is a failure but because there just is not much that can happen. With expectations high, very low results, while natural, may be disappointing and defined as failure. The

IJSIM 13,5

448

negative/low relationships between satisfaction and sales and profit, if accurate, show that the relationship between e-CRM and sales and profit is, at best, complex. And it really should be. Profitability and sales are more than simply a direct result of CRM. Expenses, inventory, SG&A and a host of other variables determine profitability. At the expense of being simplistic . . . revenue is about the stuff. Like the Clinton campaign where the theme was ``It's the economy stupid'', these results suggest that it is the ``stuff'' stupid. Consumers consume stuff not e-CRM. The talk of e-CRM/CRM seems to hide the importance of the merchandise. If it is not compelling you get no sales/revenue. If e-CRM is not compelling you still may get sales and revenue. Not all aspects of e-CRM are equally important in determining customer satisfaction. This implies that some of the CRM implementation failures (estimated at >75 percent) may be due to the implementation of features that executives believe affect customer satisfaction (that is what their experts and consultants told/promised them) but in reality do not have any effect on it at all. Thus, when customer satisfaction or sales figures do not reach an acceptable ROI for the investment in e-CRM/CRM the implementation is considered a failure. It may be that the implementation is on features that should not have an impact. The features that were found significantly related to customer satisfaction in this study are not consistent with those found significant in the past research. Can it really be that having a mailing list, quick order ability, gift certificate availability, affinity program, and your account information determines customer satisfaction with a retail Web site? It is unlikely. So what does this mean? First, it means that more research needs to be done. It means that right now CRM vendors who tell retailers (businesses) that their attributes/ features are the thing to do are probably wrong. Each company needs to research what aspects of e-CRM determine their customers' satisfaction. It may also mean that there is ± despite the findings of variability of e-CRM attributes across sites ± parity in the important ones . . . just about everyone has e-mail and complaining functions, so any positive impact of having that attribute is equal across all sites. There are a number of interesting aspects of the findings of differential availability of e-CRM features that exist across retail sights. For example, a 1-800 toll free number is available on only 75 percent of the sites. Telephonic access is considered the entry point for all businesses as they create exceptional customer relationships. That 25 percent of Web sites do not show 800 access (if the company has it and does not publish, it is probably even a greater mistake, since people right now prefer telephone to Web access). On the other hand, running or outsourcing an 800 number is expensive. If, as this research suggests, an 800 number has no impact on customer satisfaction with a site then maybe these companies are smart . . . saving money. E-mail is available on 96 percent of the retail sites of this study. This reflects the growing importance of e-mail as the preferred touch-point for customer complaints received at the modern contact center. The popularity of e-mail is a good thing. The cost per contact between e-mail and live person (phone or in

person) is significantly different. It is much cheaper to have your customer Customer contact e-mail based (much cheaper still if the contact were Web based). satisfaction with It is well established that complaints can have a significantly positive retail Web sites impact on a business. There is considerable evidence today that dealing effectively with complaints can have a dramatic impact on consumer evaluations of retail experiences, as well as enhance the customer's likelihood 449 of repurchase and limit the spread of damaging negative word-of-mouth (Blodgett et al., 1997). Complaints are pieces of market intelligence and give businesses the ability to improve functions (they find out what they are doing wrong) and to recapture otherwise lost customers. Complaining ability is available on 99 percent of the Web sites surveyed. This shows not only that retailers understand the importance of complaints but they also see the Web as an important touch-point for soliciting feedback. There are some e-CRM attributes that clearly are not seen as important, despite thinking that might reflect that they are. It seems to be generally believed that community of consumers creates loyalty. The bulletin board is a customer-to-customer communication feature. If customers can communicate they can create other/better uses for a company's products or services and thereby create additional satisfaction. Bulletin boards also fix problems that consumers would otherwise have to handle directly with the company. Yet, despite it being a positive e-CRM feature, it is available on only 10 percent of the sites, respectively. Voice over IP (VoIP ± the ability of consumers to talk directly with a representative while on the Web site with no additional telephone line) is available on only 2 percent of the sites in the sample and a while call-back button is not available on any of the sites at all. VoIP is not available simply because the broadband technology that makes it possible is yet to be widespread. However, call-back, which allows you to speak to a representative after disconnecting with the simple click of a button, may be its equivalent and it is available now to every retail site. Yet it is not present. Does this reflect ignorance of the technology or ignorance of its potential importance in sales and service? The increased importance and sensitivity of privacy is reflected in that 99 percent of the sites had this posted. The strong privacy concerns among consumers and also pressure from federal regulators to adhere to norms protecting the privacy of the site visitors has led Web sites to focus on privacy issues. Another feature that has significant potential and seems to be understood by retailers is product customization. It is available on a relatively large percentage of the Web sites (79 percent). Customization is associated with higher sales and margins. This study shows that some e-CRM features may have a relationship with customer satisfaction. Statistically significant relationships were found between mailing list, quick order ability, gift certificate and affinity program and customer satisfaction. However, at face value, these attributes seem peripheral in nature and are not central to the functioning of an e-commerce

IJSIM 13,5

450

site. The relationships found may be simply chance occurrences. The fact that the significant relationships in this study differed from the previous work suggests that different samples of retailers may yield differing relationships. If this is true, the nature of the sample becomes very important in interpreting the conclusions of research and the ratings of experts. There are a couple of methodological weaknesses of this study that need to be recognized as the results are assessed. Customer satisfaction measurements of Web sites were taken from secondary data. While the Nielsen/NetRatings data is good for the industry, it may not be good for academic research. We cannot guarantee the accuracy or representativeness of the data. Essentially we need to ask if the customer satisfaction scores really represent satisfaction. It may be that the data are neither valid nor reliable. If this is true, then any associations that we may make are inaccurate. Although we believe we have identified 42 aspects of e-CRM, it may be that we do not understand the defining features of e-CRM, implying that the 42 e-CRM may not really define CRM. We may have missed some attributes that will turn out to be really important. In addition, e-CRM is not static but a dynamic state. Thus, the results at any point in time must be considered in light of the changing nature of the industry, since the results may be very different if the same study is done a few years later. This study focuses on the link between e-CRM/CRM and customer satisfaction because it is believed that customer satisfaction mediates purchase. However, this may be of limited value since customer satisfaction may not necessarily lead to purchase. We know that in many cases satisfaction does not lead to purchase and dissatisfaction does not necessarily lead to defection (Feinberg et al., 1990). Thus, increased customer satisfaction may not necessarily improve the bottom line nor customer dissatisfaction necessarily harm the bottom line. Finally, as the results of this study are considered, we must reflect on why the relationship between e-CRM and satisfaction is so low, why only some of the relatively insignificant features of e-CRM were related to customer satisfaction and why e-CRM was not related to sales and profit. It may be because e-CRM may simply not be an important issue in consumer choice. CRM may differentiate retailers, all things being equal, but most of the time CRM has little to marginal impact on purchase and satisfaction. The real test of the importance of CRM/e-CRM would be in some trade-off study looking at the marginal/significant impact of e-CRM as product choice becomes less optimal. We would suggest that it really is all about the product/service. If the ``stuff'' is not right no amount of e-CRM/CRM is going to make a difference. If the product/service is equal across retailers maybe e-CRM/CRM would make a difference? But would this difference be greater than a difference in price, location, and convenience? Simplistic discussions and pronouncements of the importance of e-CRM/CRM may simply cover up how little difference it makes. Vendors and gurus in this area make no money if all that is important is the ``stuff''.

References Anton, J. (1996), Customer Relationship Management, Prentice-Hall, New York, NY. Anton, J. and Hoeck, M. (2002), e-Business Customer Service, The Anton Press, Santa Monica, CA. Anton, J. and Postmus, R. (1999), ``The CRM performance index for Web based business'', unpublished paper, available at: www.benchmarkportal.com Apicella, M. (2001), ``Solid CRM is difficult, but not impossible'', InfoWorld, 26 April, pp. 55-6. Blank, J. and Daniels, D. (2001), ``Analyzing customer drop-off rates: understanding why customers abandon shopping carts'', Jupiter Communications, July, p. 1. Blodgett, J., Hill, D. and Tax, S. (1997), ``The effects of distributive, procedural, and interactional justice on post complaint behavior'', Journal of Retailing, Vol. 32, pp. 185-210. Connelly, P.J. and Yoger, T. (2001), ``Can CRM win and retain loyal repeat customers?'', InfoWorld, 17 April, pp. 58-9. Cusack, M. (1998), Online Customer Care, ASQ Quality Press, Milwaukee, WI. Ewalt, D. (2002), ``E-retailing boomed in 2001'', Information Week, 25 February, p. 32. Feinberg, R. and Trotter, M. (2001), ``Immaculate deception: the unintended negative effects of the CRM revolution: maybe we would be better off without customer relations management'', in Renner, D.H. (Ed.), Defying the Limits 2, Montgomery Research, San Franscisco, CA, available at: www.cfs.purdue.edu/conscirt/quality.html Feinberg, R., Kadam, R., Hokama, L. and Kim, I. (2002), ``The state of electronic customer relationship management in retailing'', International Journal of Retail & Distribution Management, in press. Feinberg, R., Widdows, R., Wyncott, M. and Trappey, C. (1990), ``Myth and reality in customer service: good and bad service sometimes leads to repurchase'', Journal of Customer Satisfaction, Dissatisfaction and Complaining Behavior, Vol. 3, pp. 112-15. Gomez Best of the Web Guide (2001), Prima Publishing, Rosevill, CA. IBiz STATS (2001), ``The customer is still king'', PC Magazine, July, p. 7. Jackson (2001), ``The e-mail marketing report'', May, p. 50. Leon, M. (2001), ``CRM'', InfoWorld, 17 July, pp. 34-7. Menconi, P. (2000), ``CRM 101: building a great customer relationship management strategy'', in Renner, D.H. (Ed.), Defying the Limits: Reaching New Heights in Customer Relationship Management, Montgomery Research, San Francisco, CA, pp. 31-3. Nykamp, M. (2001), The Customer Differential: The Complete Guide to Implementing Customer Relationship Management, American Management Association, Chicago, IL. Rust, R. and Zahorik, A.J. (1993), ``Customer satisfaction, customer retention and market share'', Journal of Retailing, Vol. 69, Summer, pp. 193-215. Sterne, J. (1996), Customer service on the Internet: Building Relationships, Increasing Loyalty, and Staying Competitive, John Wiley & Sons, New York, NY. Swift, R.S. (2001), Accelerating Customer Relationships Using CRM and Relationship Technologies, Prentice Hall PTR, Englewood Cliffs, NJ. Szymanski, D. and Hise, R. (2000), ``E-satisfaction: an initial examination'', Journal of Retailing, Vol. 76, pp. 309-22. Tschohl, J. (2001), E-Service, Best Sellers Publishing, Minneapolis, MN. Yoo, B. and Donthu, N. (2001), ``Developing a scale to measure the perceived quality of an Internet shopping site (SITEQUAL)'', Quarterly Journal of Electronic Commerce, Vol. 2, pp. 31-45.

Customer satisfaction with retail Web sites 451