Vol. 5, No. 3, May 2005, pp. 55–69 issn 1532-0545
05
0503
0055 informs ® doi 10.1287/ited.5.3.55 © 2005 INFORMS I N F O R M S Transactions on Education The Free ISP Model and Spinway, Inc. Erik Rolland, Daria Fedotova The A. Gary Anderson Graduate School of Management, University of California, Riverside, California 92521, Erik.Rolland@ucr.edu Raymond A. Patterson School of Business, The University of Alberta, 4-30E Business Building, Edmonton, Alberta, Canada, T6G-2R6, Ray.Patterson@ualberta.ca S ince 1996, several major free ISP services have been founded in the US and elsewhere, including Spinway, Inc. Spinway’s services were offered primarily through, and under the name of, large consumer companies such as Yahoo!, KMart, Costco, Spiegel, Barnes & Noble, Ace Hardware, and NBC among others. By early 2001, the free ISP market had been reduced severely, primarily due to the lack of revenues associated with such “free” services. Spinway, founded in 1998, was a major, and perhaps the largest, free ISP in the US with over 6 million customers at its peak. Spinway folded in December of 2000, and the remains of their services were taken over by Kmart’s BlueLight.com. In November of 2002, BlueLight.com was sold to United Online and the ISP services available through BlueLight.com are no longer free. This teaching case discusses the viability of free digital infrastructures, such as Internet Service Providers (ISP). We seek to highlight and learn from the general issues experienced at Spinway and at other ISPs and Internet companies of the time. 1. Introduction 2. The Free ISP: The Basic Model “I thought Spinway had a really smart business formula and a clear path to profitability,” boasted Steve Seabolt, the CEO of a fast-growing free Internet Service Provider (ISP) since July 2000. Over the first nine months of 2000, Spinway zoomed to 5 million registered users on its advertising network and targeted the 8-million mark by 2001. In November 2000, Seabolt named free ISP as the biggest marketing strategy of the company and predicted more ways to sell to the online audience in the future. “In partnership with Kmart’s online alter ego BlueLight.com, Spinway added more than 5.2 million subscribers in 11 months starting in December of 1999, averaged more than 12,000 subscriber sign-ups per day, and distributed ISP CD-ROMs in more than 1,400 Kmart stores nationwide” (Gibson 2000). The future, however, had a big surprise in store for the users and employees of Spinway: the company unexpectedly ceased operations on December 1, 2000. Kmart, the biggest partner of the deceased ISP, acquired its select assets. Despite the evidence, Steve Seabolt insisted that the free ISP model could still work in the future. To understand Spinway’s difficulty, it is helpful gain an understanding of the industry they operated in. The discussion will switch back and forth between issues relevant to Spinway, and those relevant to any generic ISP and their operating environment. To connect to the Internet via the telephone system, a user needs to go through an ISP. The ISP serves as an intermediary between the end user’s personal computer and the phone network, which is in turn connected to numerous Web servers. The ISP provides the user with a phone number to connect to, a username and password for the account, and (optionally) special software that manages the Internet connection. In the paid ISP business model, the user pays to the provider either a fixed subscription fee for a limited number of hours online per month, or a fixed monthly fee for unlimited Internet usage, or a variable monthly total based on online time and hourly rate. The ISP manages deals with the phone companies, so that Internet usage does not increase the end user’s phone bill. This is typically done through having numerous Point of Presence (POP) services, enabling the customers to reach the ISP via a local phone call. POPs are local or regional telecommunications facilities that the user connects to when getting online. The subscription model makes user payments an important source of revenue for the ISP, if not the main one. The revenue stream can be increased through online advertising, delivering special services like e-mail and other applications to the user. Free ISPs, as the name suggests, provide the user with Internet access for no fee. Their revenue 55 Rolland et al.: The Free ISP Model and Spinway, Inc. 56 INFORMS Transactions on Education 5(3), pp. 55–69, © 2005 INFORMS generation is largely dependent on advertising. To enable delivery of commercial content to the user, a free ISP must have specially designed software that usually incorporates and enables some form of advertising delivery to the user’s desktop when s/he is connected to the Internet. The ISP charges advertisers for delivering their ads to the online audience. Online advertising, like all advertising, was supposed to increase brand recognition and awareness (Philport and Arbittier 1997), but it was also thought to boost sales more directly, since clicking on the advertisement redirects the user to the advertiser’s website where s/he can make online purchases. A popular format for advertising in the early stages was the socalled banner ad. The banner ad worked as a slim “banner” position somewhere on the screen (often at the top or bottom), where ads were displayed to the user. A popular early measure of success for online marketing through banner ads was the socalled “click-through rate,” measured by the number of user clicks on the ad over a period of time. Subsequently, more sophisticated banner ads are displayed to selected audiences (for instance, only young mothers see commercials for baby powder), or at specific times (e.g. a TV channel can advertise a show for four hours before it starts), or when triggered by the user’s actions (an ad for a travel agency may appear when the user is searching for cheap airfares). Banner ad targeting was first expected to increase the clickthrough rate, increase the effectiveness of advertisements, and increase the advertising revenues for the ISP (Griffin et al. 1998, Briggs and Hollis 1997). However, later research cautioned against banner ads, and found the promises of this type of advertising to be questionable (Tuten et al. 2000). The main issue in creating a profitable and free ISP, is that the advertising revenue must be high enough to offset the telecommunication and customer service costs, less savings achieved in other areas. “Collection fees account for roughly 20 percent of all ISP costs, according to Dan Robinson, chief executive of Spinway” (Hu 2000). By offering the ISP services for free, there are no collection fee costs. Not everyone was convinced, however. “ISPs need to maintain a Figure 1 Spinway’s Timeline paying subscriber base to offset network management and other overhead costs, according to Joe Laszlo, an analyst at Jupiter Communications. Laszlo questioned whether getting a cut of advertising revenues would make up for non-paying users eating up bandwidth” (Hu 2000). 3. The Rise of Spinway.com Spinway was founded in October 1998, incorporated in January 1999, and officially launched its service in December 1999 when it secured its first partnership with Kmart (see timeline of events presented in Figure 1). At its peak in the fall of 2000, Spinway covered 96% of US territory, as well as parts of Canada and Latin America. From the start, the company assumed the free ISP business model with advertising in the form of banner ads as the main source of revenue generation. Spinway paid for the network infrastructure, customer service, and other costs of providing access. To establish the advertising network, Spinway partnered with offline brand-leaders to co-brand their free ISP services, including Yahoo! Inc, Kmart’s BlueLight.com, Costco, Barnes & Noble, Ace Hardware’s OurHouse.com, NBCi, Spiegel, Hewlett Packard, and others (Figure 2). Spinway licensed its service to partner companies so that they could provide their users with free Internet access. Spinway supplied users with a software client that connected them to the Internet via selected local phone numbers, at no charge. When signing up for service, the user had to provide personal information that was later used for marketing purposes. In exchange for free Internet access, users were required to view various ads on a banner on their screen whenever they were online. The software was customized for Spinway’s partners to display their special offers and personalize the user’s Internet browser start pages to point to the partner’s website. Spinway developed full-motion videos that would play commercials while the user was being connected to the Internet, thus turning the “dead time” of the dialup into a revenue generator. The company claimed to offer “advertisers a new approach to online Rolland et al.: The Free ISP Model and Spinway, Inc. 57 INFORMS Transactions on Education 5(3), pp. 55–69, © 2005 INFORMS Figure 2 Spinway’s Partnerships advertising that uniquely targets consumers by demographics, psychographics, geography and keyword” (Appendix B, Figure B.2). Some free ISPs assumed that the user already had access to the Internet and distributed their software only through download from their website. Unlike others at this time, Spinway also offered its service on CD-ROMs at the cash registers at its partners’ brickand-mortar locations. This approach enabled Spinway and its partners to capture many customers who accessed the Internet for the first time and made Spinway their main service, rather than using it as a backup for their regular provider. During the first nine months of 2000, Spinway increased its user base to more than 5 million registered users, becoming the fastest growing ISP in history. Of this user base, 75% was acquired through the partnership with BlueLight.com. “Registered” users do not mean “regular”—fewer than half spend over 15 minutes online in a given month (Gardner 2000). Still, according to Spinway’s CEO Steve Seabolt, these percentages were in the same range as America Online’s. The demographics of Spinway’s customer base were surprisingly diverse and not limited to the stereotypical educated young affluent male (Figure 3). 4. Features Of Spinway Online reviews of free ISPs in 2000 consistently rated Spinway #1 or a close #2 in a tie with 1stUp.com (Woodall 2000). The reviewers during this time acclaimed “all sorts of cool specialized content


that other free ISPs just don’t have” (Hypermart 2000). This included such items as online communities and personalized start pages with news and information targeted to the customer. Spinway’s users positively Figure 3 Customer Base Demographics evaluated free e-mail accounts and Web hosting that came with the service (for more information and screen shots of Spinway’s product, please refer to Appendix B). Some reviewers maintained that Spinway was too inquisitive in gathering data about the users. On the other hand, this data was the basis for targeted advertising, and provided personalized content that the users seemingly enjoyed. Spinway declared the “true targetability” of their advertising: “target relevant advertisements to specific demographics regardless of which computer or browser a user is using” as well as “target based on time-of-day and geographic location” (Spinway corporate website media kit, 2000). Spinway never quite fulfilled the promise of advertising to specific user categories; for instance, all users regardless of ethnic origin viewed the banner ad for a black community bookstore. Action-triggered advertising was successfully implemented: specific ads ran when their associated keywords matched user’s query. For instance, when the user visited a jewelry website, s/he would typically see an ad for engagement rings; when the user searched for used automobiles, a banner ad for car sales would come up even before the user would see the results of the query. Spinway was selling banner ad time to the advertisers based on the ad duration, the desired exposure frequency, the number of keywords associated with the advertisement, and other targeting parameters. Financially, the banner ad revenue fell short of covering the cost of user account maintenance. A special treat for the advertisers was the rich media functionality of Spinway software. Spinway claimed to be the only company that offered “television quality, full-motion video, instantly, over any connection without impacting performance.” On request, Spinway digitally rendered regular TV commercials, and played them in user’s connection window during dial-up. While a user was waiting to connect to the Internet, 58 his attention was focused on the dial-up window, and this “dead time” was best used for advertising impression. The video was downloaded to the user’s computer while connection was inactive (the user was online, but not utilizing the bandwidth), stored on the hard drive, and played during subsequent connection attempts. Some users welcomed commercials during dial-up since they relieved the dullness of waiting; others complained about the music score of the videos mixing with the dialing noise of the modem in an unpleasant way. Advertisers did not immediately embrace the idea of Internet videos as a complement to TV commercials. In fact, Spinway was unable to sell video advertising time and had to give away videos for free in the hope that advertisers would eventually buy into the concept. This marketing move caused technical problems because the group of engineers working on videos did not know about upcoming give-aways, and was not ready to cope with the amount of advertising requests that resulted. Surveyed users complained about chronic disconnection problems which were, in fact, not a failure but an unannounced feature of Spinway service at the time of testing. Users were cut off after 50 minutes online and had to reconnect if they wanted to continue using the Internet. Automatic disconnection was supposed to cut down on user account maintenance costs, or so as Spinway figured, most people got online only to check email and browse the Web for a short period of time, and they often stayed connected when they are no longer actually using Internet. Auto-disconnection proved not to be such a great idea because users were not informed of the new policy, thought something was wrong with their side of the connection, and called customer service. The cost of customer service per user instantly exceeded the cost of maintaining an idle connection, and resulted in even more losses for Spinway. 5. The Fall of Spinway, Inc.: Financials Spinway was a start-up that never made it to an IPO; being a private company, it did not disclose financial documents. Select figures are publicly available, some from anonymous sources, and other numbers can be deduced from financials of a similar, but public, free ISP NetZero (see Appendix A). Spinway secured the seed round of financing in early 1999, and closed the second round in October 1999. Danny Robinson, the CEO of Spinway until July 2000, would not disclose the company’s valuation or revenues, nor how much it had raised in the two initial financing rounds. The third round (Series C) was completed in May 2000, with $32 million raised. Deutsche Bank Alex.Brown and Spectrum Rolland et al.: The Free ISP Model and Spinway, Inc. INFORMS Transactions on Education 5(3), pp. 55–69, © 2005 INFORMS Equity Investors joined in Series C the second-round investors Softbank Venture Capital, Al Shugart International and Mentor Venture Partners. According to Robinson, the latest round was largely oversubscribed and the company raised $2 million more than originally anticipated (Braunschweig 2000). Hoover’s Online estimated Spinway’s annual sales as $10 million to $25 million. However, the grapevine at the company itself quoted much humbler revenue figures, in the low hundreds of thousands a month. Spinway’s burn-rate (the amount of cash that a startup company goes through every month or quarter before achieving profitability) was never disclosed, but can be estimated based on competitors’ rates. According to Gary Baker, vice president of public relations at ex-free ISP Juno, their burn-rate in the second quarter of 2000 was $40 million (Meier 2000). NetZero reported a $23 million per quarter burn-rate for the period ending September 2000. Despite the losses, these companies remained afloat due to significant cash reserves: $69 million at the end of September 2000 for Juno, $220 million in cash and investment for NetZero. Spinway’s fast growth in terms of number of subscribers was costly for the company. Growing user mass, especially geographically spread out, required higher bandwidth and more POPs. ISPs contract for the use of POPs from wholesale providers. The contract can be either a usage agreement, under which the POP provider charges the ISP for the aggregate number of hours its users are connected to the provider’s network, or a capacity agreement, under which the ISP pays for a fixed amount of the wholesaler’s telecommunications capacity even if it is underutilized. Usage agreements generally have a minimum purchase clause. Payments to telecommunications providers constitute the bulk of the “cost of revenue” for ISPs. “Some people were using free access to power their small businesses, and Juno said that such users were gobbling up nearly half of their telecommunication costs” (St. Pierre 2001). It is not known what type of contracts Spinway signed with its POP providers GTE, ICG Communications, and Level 3 Communications. Apparently, Spinway was not satisfied with its relationship with the original main POP provider ICG Communications, because this contract was negated in mid-2000. The rumors were that ICG did not deliver the bandwidth upgrades that it had charged Spinway for, and the ISP suffered substantial losses on this deal. Other ICG customers (Microsoft, NetZero, Earthlink) complained that their users were blocked by busy signals when trying to connect to the Internet through ICG phone numbers, or had connections cut off. The vice president of NetZero, John Fetveit, said that ICG Rolland et al.: The Free ISP Model and Spinway, Inc. INFORMS Transactions on Education 5(3), pp. 55–69, © 2005 INFORMS over-promised and then under-delivered on its service (Sherer 2000). By fall of 2000, ICG was entering Chapter 11, and Spinway switched to another big phone company, Genuity, who became the main POP provider until the December crash of the free ISPs. The tech-world gossip described the agreement as highly unfavorable for Spinway. Never the less, some industry executives remained optimistic, at least through early 2001: “NetZero execs seem confident they’ll survive the dot-com doldrums. They dismiss doubts by skeptical analysts that ad revenues will ever surpass telecommunications costs. ‘The advertising market is ugly now. But an ad-supported business can exist. One is called radio. The other is television,’ says NetZero CEO Mark Goldston” (St. Pierre 2001). Not everyone was sharing this optimism about the free ISP market: “


the fee-for-service giants are exacting their revenge. ISPs such as Earthlink, America Online, CompuServe, and Microsoft are stepping up efforts to win over frugal customers with alternative pricing plans aimed at folks who don’t spend a lot of time surfing the Web. AT&T is also courting customers who want to connect to the Net on the cheap. Its WorldNet service gives consumers 150 hours a month for $4.95. ‘We’ve always questioned the sustainability of the free model. The few dollars AT&T charges consumers allow us to cover costs,’ says AT&T spokesperson Janet Wiles.” (St. Pierre 2001) In the fall of 2000 Spinway felt the cash deficit and tried to push down the burn-rate. That was the time when automatic user disconnection after 50 minutes online was instituted without success. The company also attempted to cut back on operational expenses by such means as continuous brownouts in the office (economy on electricity cost), providing less food for employees in the corporate kitchen, and delaying long-promised pay raises. Spinway was working on securing the last round of financing before IPO in Fall 2000 and failed to acquire it. It is said that the company was looking for a $60 million investment half of which would be spent immediately to pay off liabilities. When shortterm financing did not happen, the ISP defaulted on its creditors and folded operations on December 1, 2000. ZipLink, an Internet infrastructure provider for Spinway since January 2000 (Luening and Hu 2000), announced in November 2000 that it was closing, and specifically cited the failure of their second largest customer Spinway. They disclosed that Spinway had defaulted on recent payments to the company (September’s bill was over $1 million). Interestingly, Spinway fought till the very end and as of November 17, 2000, the company was denying the default. Spinway founder Danny Robinson protested: “I don’t 59 know why they’re saying that. I’m disappointed that they’d say that. We’ve got enough cash on hand. We can pay them” (Krause 2000). The payment never happened, and ZipLink suspended operations. “In a press release announcing its closure, ZipLink said the decision ‘was influenced by a recent default on payments to ZipLink by its second-largest customer, Spinway.com, and general market uncertainty regarding the future of providers of free Internet access’ ” (Luening and Hu 2000). Soon after the ZipLink failure, on November 21st, Spinway laid off 30 employees. The rest were let go on December 1st when the company officially ceased operations. The laid-off employees blamed the management for the failure of the enterprise, saying that Spinway could have been the best free ISP around if not for foolish executive decisions. Most Spinway people did not believe that the business model was at fault, although some voiced the opinion that the whole start-up was a throw of the dice that just didn’t land the right side up. Employees accused the managers of lying to them, concealing information about the true state of affairs at the company until the very last day, and betraying the corporate code “People are our most important asset.” The concern over the future of Spinway is illustrated in the following excerpt from an article that appeared between the time it was announced that Spinway was failing to pay its current obligations and just prior to Spinway’s collapse. “The failings at Spinway are not terribly surprising because just over a week ago, 1stUp.com, another provider of free Internet access, was hit with the news that its parent company was cutting funding in the next 60 days. That company is expected to fold in a matter of weeks, leaving the companies it works with, including AltaVista and Excite, out in the cold and without a partner to offer Internet access to their subscribers. Should Spinway fold, it’s not clear who could pick up all its subscribers. With 1stUp out of business, it’s hard to think of a company that could take Spinway’s place. The only other large free ISP, NetZero (NZRO), has a very different business model and likely would not pay to pick up those subscribers. Another large ISP might be able to pick up the slack, but because the model seems to have proven itself untenable, it hardly seems likely anyone would want to come rushing to the rescue.” Krause and Helft (2000) BlueLight.com, the main partner of the company and the online representative of Kmart, acquired intellectual property and other select assets of Spinway and retained key employees to continue free service over the Christmas holidays (Luening and Hu 2000). According to the CEO of BlueLight.com, Mark Goldstein, “the cost of acquiring the core assets of Rolland et al.: The Free ISP Model and Spinway, Inc. 60 INFORMS Transactions on Education 5(3), pp. 55–69, © 2005 INFORMS Spinway was very modest” (BlueLight 2000). BlueLight did not take over the liabilities of the failed ISP. However, they did maintain ISP services for Barnes and Noble, Costco, and Spiegel (Luening and Hu 2000, Fischer Lent 2000). “All other customers chose to redirect to another provider. Those customers include Capital One Financial Corp., Yahoo Inc. and OurHouse Inc.,” said BlueLight spokesman Dave Karraker (Verton 2000). 6. Discussion “ ‘The general Spinway business model to co-brand with brick-and-mortar companies was a terrific Figure A.1 NetZero Financial Statement business model,’ said Lydia Leong, a principal analyst at Stamford, Conn.-based Gartner Group Inc.” (Verton 2000). “It brought customers to the table who might not have had Internet access. However, it was dependent upon advertising, and dot-com companies have spent a lot less money on ads in the last few months” (Verton 2000). “Other free service providers that relied on ad dollars also haven’t survived. Those include San Francisco-based 1stUp.com Corp., which delivered 3 million of AltaVista Co.’s 5.5 million paying and nonpaying customers. Gary Baker, a spokesman for New York-based Juno Online Services Inc., one of the few remaining providers that offer free Internet Rolland et al.: The Free ISP Model and Spinway, Inc. 61 INFORMS Transactions on Education 5(3), pp. 55–69, © 2005 INFORMS service, said Spinway’s business wasn’t diversified enough” (Verton 2000). Steve Seabolt, the Spinway CEO, looked at the failure from a different point of view. This is evident by the following quote: “ ‘We were like a fully loaded 747 taking off, climbing over the Pacific,’ says Steve Seabolt, former CEO of Spinway, which had 6 million customers before it went under. ‘Only there was not enough gas and we had no way to refuel in midflight. We were blowing through fuel so fast there was no time to stop, pause or do a course correction.’ (Krause 2001)” “Dylan Brooks, an analyst for ISP-tracking firm Jupiter Research, does not think that the free ISP business model will go away entirely” (Schiffman 2000). Brooks stated that “a shakeout in the space may have a positive effect on the industry, by reducing the supply of free-ISP banner ads, which could help drive prices up. It’s a healthy cleansing that could make the handful of survivors a little healthier” (Schiffman 2000). Perhaps many free ISPs would have survived if the advertising market hadn’t folded with the economic slowdown. Conservative analysts say there’s still cause for concern and “the whole consumer end of the ISP market has a giant question mark


and aside from AOL, it’s not even clear that paid ISP’s business model is working out” (Schiffman 2000). Fred Morgan, an analyst for an investment firm Jefferies, and who follows for-pay ISPs, is adamant: “The free model has failed” (Smetannikov 2000). But has the “free model” failed? We see many examples of “free technology” giveaways that still persist: Microsoft gives away free e-mail and storage space through Hotmail, and free instant messaging services through MSN Messenger, as well as free Web site hosting through MSN Communities; Opera gives away a free browser containing a banner ad window (the paid version is offered without the banner ad window). The fundamental question is why the free ISPs failed, yet the “free model” still lives on? Figure B.1 7. Postlog Figure B.2 BlueLight quickly reengineered the business model of the acquired ISP to focus its free Internet on Kmart shoppers. It determined that the typical consumer who uses its ISP spends online an average of 15 hours per month, and only a small percentage of subscribers use up over 25 hours a month, thus accounting for a significant share of telecommunications cost. To trim down costs, BlueLight limited free Internet use to 25 hours a month per user. “We want to continue to offer the best Internet service possible for the people who shop at BlueLight.com, and instituting hour limitations on high-impact users will allow us to do just that,” said Mark Goldstein in the December 22nd Spinway Company Information Prior to Takeover press release. “The cost of acquiring the core assets of Spinway was very modest for BlueLight and will have little impact on our current march towards profitability,” said Mr. Goldstein. “We remain dedicated to our subscribers. BlueLight’s Internet service is an invaluable marketing tool for the company and we feel it is in our best interest, and in the best interest of all our subscribers during these tough times for pureplay dot com companies, to make sure it continues” (Gibson 2000). On December 11th, 2000, BlueLight terminated the relationship with Spinway’s dial-up provider Genuity (Wagner 2000). Genuity turned off the service, losing a major account with $8.7 million credit on Spinway. BlueLight’s CEO stated that “while the Genuity Grinch was about to steal Christmas from millions of online holiday shoppers, we acted quickly with PSINet and WorldNet to maintain our nationwide Internet service” (Wagner 2000). Genuity filed a lawsuit against BlueLight. Using new infrastructure providers and software maintenance by ex-Spinway employees on retainer packages, BlueLight successfully made it through the holiday season, with a sales increase of 1,060% compared to the same period in 1999. After a period of hesitation as to whether to continue free Internet service or not, BlueLight decided on a tiered model with some free and some paid access. Spinway’s Advertising Information Rolland et al.: The Free ISP Model and Spinway, Inc. 62 Figure B.3 INFORMS Transactions on Education 5(3), pp. 55–69, © 2005 INFORMS Spinway’s Consumer Information In March 2000, BlueLight offered Basic Service that consists of 12 free hours online a month, and Premium Service with 100 hours online monthly for $9.95. Basic Service was no longer advertised; a new user could only purchase the Premium package, and in event of cancellation, the service reverted to the basic 12 free hours a month. In November of 2002, United Online, which also operates the Juno and NetZero Internet consumer brands, purchased the ISP assets of BlueLight.com for $8.4 million U.S. cash (United Online 2002). As of May of 2003, BlueLight Internet Services now charges $9.95 per month with the first month free, plus $1.95/minute (or $14.95/call in certain areas) for service. United Online still offers an introductory free ISP service (up to 10 hours per month) through NetZero and Juno. United Online is repositioning its offerings away from competing solely on price (although not totally) to positioning themselves as a lower price, limited-content alternative provider to AOL. Figure B.5 Spinway After Takeover by Kmart Case Appendix A. NETZERO Financials The basis for comparison between Spinway and NetZero is their analogous business models, number of subscribers, and fate of the free ISP service. NetZero had 1.45 million customers as of December 1999 (International Telecommunication Union 2000), and 4.5 million at the end of August that same year (Lake 2000). In June of 2000, only 2 million of NetZero’s users actually logged on (Lake 2000), thus illustrating that a lot of users of free ISP services were relatively inactive. Comparatively, during the first nine months of 2000 Spinway grew from about 1 million to about 6 million customers. The financial statement for NetZero is presented in Figure A.1 below. Case Appendix B. SPINWAY SCREEN SHOTS This section shows screen images from Spinway’s own web sites. The web sites are no longer available, and are presented here for the reader to get a flavor of the Spinway product. References Figure B.4 Spinway’s Major Features BlueLight Press Release. 2000. BlueLight.com to acquire select assets of Spinway, Inc. December 4, http://www.spinway.com/ (no longer accessible). Braunschweig, C. 2000. Private equity week: Spinway topples toward IPO. Al Shugart International: Press Room http://www. (no longer alshugart.com/press_room/news_artical.html accessible). Briggs, R., N. Hollis. 1997. Advertising on the Web: Is there response before click-through? J. Advertising Res. 37(2) 33–45. Fischer Lent, A. 2000. BlueLight adopts orphaned spinway ISP customers: ISP service continues, but it may not remain free of charge under new management. PCWorld.com, Monday, December 04, http://www.pcworld.com/news/article/0,aid, 35942,00.asp. Gardner, E. 2000. Growing through partners. Internet World Magazine (November 15), http://www.internetworld.com/ 111500/11.15.00fastforward1.jsp (no longer accessible). Gibson, B. 2000. BlueLight.com to acquire infrastructure from Spinway. MacCentral Online. December 4, http://maccentral. macworld.com/news/0012/04.bluelight.shtml. Rolland et al.: The Free ISP Model and Spinway, Inc. INFORMS Transactions on Education 5(3), pp. 55–69, © 2005 INFORMS Griffin, K., R. Whitehead, R. Gatlin. 1998. Banner advertising on the World Wide Web: An analysis. Internat. J. Management 15(2) 262–265. Hu, J. 2000. Spinway sees itself as answer to ISPs’ payment problems. CNET News.com (February 4), http://news.com. com/2100-1023-236551.html%20. Hypermart. 2000. Free ISPs: Spinway. The Get Paid Encyclopedia, http://icc2000.hypermart.net/getpaid/isps/spinway.html. International Telecommunication Union–ITU. 2000. Top 20 Online Service Providers, 1999, http://www.itu.int/ITU-D/ict/ statistics/at_glance/Top20ISP.html. Krause, J. 2001. The price of being free. The Industry Standard (Feb. 26), http://www.thestandard.com/article/0,1902,22216, 00.html?body_page=1. Krause, J. 2000. Something’s rotten in the state of telecom. Security Informer (November 17), http://www.security-informer.com/ english/crd_companies_292858.html (no longer accessible). Krause, J., M. Helft. 2000. Spinway needs cash—and bluelight needs spinway. The Industry Standard (Nov 22), http://www. thestandard.com/article/display/0,1151,20356,00.html%20. Lake, D. 2000. It’s not easy being free. The Industry Standard (Aug. 28), http://www.thestandard.com/article/0,1902,17988, 00.html. Luening, E., J. Hu. 2000. BlueLight to snap up Spinway assets. CNET News.com (December 4), http://news.com.com/21001023-249323.html. Mariano, G., E. Hansen. 2000. CNET News.com (August 2), http:// rss.com.com/2100-1023-243950.html. Meier, L. 2000. You get what you pay for. Red Herring (August 17), http://www.redherring.com/investor/2000/0817/invjuno081700.html. 63 Philport, J. C., J. Arbittier. 1997. Advertising: Brand communications styles in established media and the Internet. J. Advertising Res. 37(2) 68–76. Rayport, J., B. Jaworski. 2004. Introduction to E-Commerce, 2nd ed. McGraw-Hill/Irwin. Schiffman, B. 2000. The beginning of the end for free ISPs? Forbes (November 22), http://www.forbes.com/communication/ 2000/11/22/1122freeisp.html. Sherer, P. M., G. McWilliams. 2000. A brash provider of internet services became unplugged; Chapter 11 looms. The Wall Street Journal (November 13). Smetannikov, M. 2000. Free ride nears end. Interactive Week. http://www.ventanalakes.org/bboard/ (December 11), freeinet.htm. St. Pierre, N. 2001. End of the line for free ISPs?. Business Week Online (January 12), http://www.businessweek.com/ bwdaily/dnflash/jan2001/nf20010112_571.htm. Tuten, T. L., M. Bosnjak, W. Bandilla. 2000. Banner-advertised Web surveys. Marketing Res. Winter 1999/Spring 2000, 11(4) 16–21. United Online press release. 2002. United online completes acquisition of BlueLight.com ISP Assets. November 5, http://www. irconnect.com/untd/pages/news_releases.mhtml?d=33476. Verton, D. 2000. BlueLight.com rescues failing ISP Spinway. Computerworld. December 8, http://www.computerworld.com/ managementtopics/xsp/story/0,10801,54949,00.html. Wagner, J. 2000. A dial-up mess in the E-aisles. Internetnews.com. December 11, http://www.internetnews.com/xSP/article. php/8_531361. Woodall, B. 2000. Rating the free ISPs: Bluelight.com/Spinway.com. The Bulls Eye.Com, May 19, http://www.thebullseye.com/ articles/tech/woodall05192000-5.html. Rolland et al.: The Free ISP Model and Spinway, Inc. 64 INFORMS Transactions on Education 5(3), pp. 55–69, © 2005 INFORMS Case Discussion The Free ISP Model and Spinway, Inc. T his paper summarizes our experience using a case study of a failed dot.com company, Spinway Inc., in advanced undergraduate level E-Commerce courses. The Spinway case introduces the students to a case where knowledge about many aspects of e-commerce business modeling is required. The case has been used effectively to improve the students’ skills in issues related to business modeling, ranging from topics related to uncovering market opportunity to financial models. The case material is based upon interaction with Spinway before they ceased operations in December of 2000. The case describes the issues that Spinway faced. Spinway’s real financial statements are not included, but comparisons to other similar companies at the time (NetZero) are possible for this purpose. The goal of the case is not in studying the possible points of failure for Spinway, but rather it is creating a complete and sustainable business model. “Those who fail to learn the lessons of history are doomed to repeat them.” George Santayana, American philosopher (1863–1952) 1. Synopsis of the Case Study This case illustrates salient issues related to E-commerce business models. The case discusses the life of Spinway, which in 1999 was reputed to have been the largest free Internet Service Provider (ISP) in the world with almost 6 million customers. Spinway services were offered through, and under the name of, large consumer companies such as KMart, Costco, Barnes & Noble, Ace Hardware, Spiegel, and NBC among others. By 2001, the free ISP market encountered financial difficulties due to the lack of advertising revenues associated with these services. Spinway, founded in 1998, declared bankruptcy in December of 2000. Spinway’s services were taken over by KMart’s BlueLight.com. In November of 2002, BlueLight.com was sold to United Online, which is still in operation. This teaching case discusses the viability of free digital infrastructures. The case aids the students in understanding ISPs, and it also helps the students in understanding and considering the diverse issues involved in building successful business models. This case is ideally suited for use in E-Commerce and E-Business courses, but could potentially be used in a capstone course, a strategy course, or any course that discusses business models. The case is equally applicable in advanced undergraduate courses, master’s courses, or executive education. We have included 2 sets of supplementary case questions for such classes. 2. Teaching Objectives The main objective of this case is to teach the students how to think about business models in context of E-commerce. This is achieved through a series of skill-based exercises addressing many of the points comprising the major challenges of successful E-commerce. In addition, there is opportunity to learn about technology infrastructures and about transitioning of high-tech products over time. For example, with the increase in availability of broadband, dialup ISP’s have transitioned over the past 5 years to become technology providers for occasional access outside of the home and office. 2.1. Detailed learning objectives Students improve their skills in the following areas, including an understanding of: • The business opportunity • The value system, value chain, and value proposition • Other important components of business models, including boundary issues (e.g., alliances) • Product positioning • Financial models (including those based on advertising revenues) • Creating a sustainable competitive advantage • Market segmentation • Market and firm structures The skills can be reinforced by using the case questions at the end of this case. The lessons learned from Spinway are equally applicable to any business model that prices a product or service below cost in exchange for market share, especially share of a market that does not produce a clear revenue stream. As with any case, the issue of timeliness may be a point of contention from the students’ perspective. As suggested by an anonymous referee “


business students have tended to treat the dot.com ‘bubble’ and subsequent ‘bursting’ as one-time events that are now over—time to get on with real business. Yet the pace of technological innovation has not slowed, nor has ‘real’ ecommerce (online trade that is grounded in sound business models), regardless of what has happened in the financial markets.” Indeed, Rolland et al.: The Free ISP Model and Spinway, Inc. INFORMS Transactions on Education 5(3), pp. 55–69, © 2005 INFORMS we recommend that a brief discussion of the case’s timeliness be held at the onset of discussing the case. 2.2. Incidental Benefits Another benefit of this case with the group exercises that are provided is that students are required to learn how to intelligently search both library systems and the internet in order to find necessary information. In our experience, the quality of the students’ answers is highly dependent on their search efforts. Utilizing group exercises, students are encouraged to learn to communicate effectively in groups. For example, with the use of Blackboard or WebCT, students will often resort to online meetings because of difficulties of coordinating face-to-face meetings. Students may also have to deal with disparate effort levels exhibited by various team members. 3. Case Analysis The case analysis consists of several parts: 1. Case reading and background. Students should be required to read the case prior to the start of the course. The first set of questions entitled “CASE QUESTIONS–A: The Information Systems and Technology Perspective” are typically assigned immediately. These questions encourage the students to learn what an ISP is, as the case does not explain the technical details of the ISP industry. Once the students have a thorough understanding of Spinway and the ISP industry, the real business model analysis can commence. 2. E-Commerce business model issues. The purpose is to introduce the students to business models, and the issues that are necessary for success as these models relate to E-Commerce. 3. Questions for Analysis. For each issue related to the business model, we have in the section “CASE QUESTIONS–B: The Business Model Perspective” created a set of sample questions. These questions reflect issues that we believe currently are important to understanding and designing useful E-Business models. 4. Possible Extensions of the Case Study While free ISP’s still exist, the type of connections and the customer needs for such connections are changing. Wireless is the most obvious change. A ubiquitous access model, where customers roam internationally, still require ISP connections, and broadband ubiquitous access is on the verge of realization in some regions. 65 5. Teaching Suggestions and Benefits We suggest that this case be used as a supplement to the class book. The case can be used as a course foundation case which is referred to throughout the regular course/book material. Further, we propose that all key concepts introduced in the textbook, such as segmentation, financial models, etc., be practiced using the case. It is our experience that positioning the homework in a historical context is useful. That is, require the students to develop answers to the case questions for two time-periods: 1999 and current (2004 or later). Also, we have found it useful to have students work on the case exercises in groups. This enables better instruction and feedback, as they can present their finding in short, in-class, presentations. It is useful to make the number of groups as small as possible while keeping the group sizes to 5 or less. 5.1. Usage We have used this case successfully in the classroom at the University of California, Riverside, for 2 undergraduate (mostly seniors or year 4 students) E-Commerce courses. Three MBA students were also enrolled. The case was used as a key piece of the teaching materials, in that all aspects of E-Commerce business models were related to the case in the form of student exercises. The question set “CASE QUESTIONS–A: The Information Systems and Technology Perspective” were used at the beginning of the course to motivate the students to read the case and to gain an understanding of the ISP business. The second set of questions, “CASE QUESTIONS–B: The Business Model Perspective,” were used throughout the course to give the students an opportunity to “practice” the material presented in class. The class culminated with the full business model analysis for Spinway. In all classes, we have used the full business model as an example to spur creativity in the students by asking them to propose a full-fledged business model specification for an E-Commerce idea of their own choosing. 5.2. Student Feedback In order to assess student opinions about the case, we designed a 10 item questionnaire that was administered during the last class meeting for the courses. Gender demographics in the E-Commerce undergraduate courses were 33 males, 23 females, and 2 unreported. The average age was 22 (standard deviation of 1.9), with a minimum age of 20 and a maximum age of 30. 35 students expect an A, 7 an A−, 5 a B+, 7 a B, 1 a B−, and 3 students did not declare an expected grade. 39 students were business administration majors, 12 were engineering majors, one economics major, one biology major, three MBA students, and two did not declare their major. Rolland et al.: The Free ISP Model and Spinway, Inc. 66 INFORMS Transactions on Education 5(3), pp. 55–69, © 2005 INFORMS Table 1 Summary of Results Average Median Standard deviation Minimum Maximum All students N = 58 Q1 The case was well-prepared Q2 The case was interesting Q3 The case was timely Q4 The case was relevant Q5 The case questions were appropriate Q6 The case assignment was a useful experience for you Q7 You learned something valuable from this case Q8 The required analysis was challenging Q9 The required analysis was too difficult Q10 Overall, this case was excellent 4.30 4.16 3.57 4.26 4.09 4.29 4.36 3.72 2.69 4.05 4 4 4 4 4 4 4 4 3 4 0.65 0.67 0.84 0.76 0.80 0.84 0.67 0.83 0.94 0.54 2 2 1 2 2 1 2 2 1 3 5 5 5 5 5 5 5 5 5 5 Business Administration Majors N = 39 Q1 The case was well-prepared Q2 The case was interesting Q3 The case was timely Q4 The case was relevant Q5 The case questions were appropriate Q6 The case assignment was a useful experience for you Q7 You learned something valuable from this case Q8 The required analysis was challenging Q9 The required analysis was too difficult Q10 Overall, this case was excellent 4.45 4.08 3.49 4.31 4.18 4.41 4.36 3.77 2.82 4.13 4 4 4 4 4 5 4 4 3 4 0.55 0.70 0.85 0.73 0.72 0.72 0.63 0.81 0.97 0.52 3 2 1 2 2 3 3 2 1 3 5 5 5 5 5 5 5 5 5 5 Engineering majors N = 12 Q1 The case was well-prepared Q2 The case was interesting Q3 The case was timely Q4 The case was relevant Q5 The case questions were appropriate Q6 The case assignment was a useful experience for you Q7 You learned something valuable from this case Q8 The required analysis was challenging Q9 The required analysis was too difficult Q10 Overall, this case was excellent 4.00 4.25 3.58 4.25 3.92 4.25 4.58 3.92 2.50 3.92 4 4 4 4 4 4 5 4 2 4 0.85 0.62 0.67 0.62 0.90 0.75 0.51 0.79 0.90 0.51 2 3 2 3 2 3 4 3 1 3 5 5 4 5 5 5 5 5 4 5 Male N = 33 Q1 The case was well-prepared Q2 The case was interesting Q3 The case was timely Q4 The case was relevant Q5 The case questions were appropriate Q6 The case assignment was a useful experience for you Q7 You learned something valuable from this case Q8 The required analysis was challenging Q9 The required analysis was too difficult Q10 Overall, this case was excellent 4.27 4.12 3.36 4.18 4.00 4.30 4.30 3.64 2.61 3.94 4 4 4 4 4 5 4 4 3 4 0.72 0.70 0.86 0.77 0.79 0.92 0.73 0.74 0.79 0.56 2 2 1 2 2 1 2 2 1 3 5 5 5 5 5 5 5 5 4 5 Female N = 23 Q1 The case was well-prepared Q2 The case was interesting Q3 The case was timely Q4 The case was relevant Q5 The case questions were appropriate Q6 The case assignment was a useful experience for you Q7 You learned something valuable from this case Q8 The required analysis was challenging Q9 The required analysis was too difficult Q10 Overall, this case was excellent 4.32 4.17 3.83 4.35 4.17 4.22 4.43 3.87 2.91 4.26 4 4 4 5 4 4 4 4 3 4 0.57 0.65 0.72 0.78 0.83 0.74 0.59 0.97 1.08 0.45 3 3 2 3 2 3 3 2 1 4 5 5 5 5 5 5 5 5 5 5 Rolland et al.: The Free ISP Model and Spinway, Inc. 67 INFORMS Transactions on Education 5(3), pp. 55–69, © 2005 INFORMS For the survey’s 10 questions, the answers were given on a 5-point Likert scale (1 = Strongly Disagree, 2 = Disagree, 3 = Neutral, 4 = Agree, 5 = Strongly Agree). Students rated the case as well-prepared (Q1: average score was 4.30), interesting (Q2: average score of 4.16), and relevant (Q4: average score of 4.26). The case was a useful experience (Q6: 4.29) where they learned something valuable (Q7: 4.36). Further, they rated the case questions as appropriate (Q5: 4.09). Student ratings for timeliness of the case (Q3) averaged 3.57. The required analysis was appropriately challenging (Q8: 3.72), yet not too difficult (Q9: 2.69). Overall, the students agreed that the case was excellent (Q10: 4.01). Table 1 holds the results for each of the 10 questions. Results are provided for all students, including averages, medians, standard deviation, and min/max. Also, the answers are reported for two of the subgroups (business administration majors, and engineering majors), as well as for male and female students. As some of these subgroups had small sample sizes, we did not perform any further formal analysis of the difference in the answer scores across majors or gender. 5.3. Conclusions Based on our in-class experience with this case, we recommend use of this case as a team project spanning an appropriate section of the course. We believe that this case is an excellent supplement for advanced undergraduate, master’s, and executive education courses in E-Commerce, E-Business, capstone, strategy, and other courses that discuss business models. Acknowledgments We would like to thank Mr. Curt Varner, one of the founders of Spinway, for his comments and discussion. We also thank several other anonymous ex-employees of Spinway for comments and insight. We also acknowledge Ms. Daria Fedotova as co-author of the case upon which this article is based. Rolland et al.: The Free ISP Model and Spinway, Inc. 68 INFORMS Transactions on Education 5(3), pp. 55–69, © 2005 INFORMS Teaching Notes and Case Questions: The Free ISP Model and Spinway, Inc. 1. Basic Issues This case is useful in a multitude of contexts within a business school. First, the case can be used to relate technology and information systems issues such as telecommunications infrastructure. Second, the case bridges the fields of strategy and e-commerce, and should be of use to classes in both these disciplines. Finally, the case can be used in a capstone course, where students must demonstrate their command of integrated management (and technology) issues. This case doesn’t have enough insider information to fully evaluate strategic, management, or technology mistakes that Spinway might have made, and it is not intended to. However, it has sufficient material for discussion of the underlying model. Further, the case relates to many associated technology, marketing, finance & accounting, and strategy issues including: telecommunications infrastructure, co-branding, longterm financing and cost structures, Porter’s 5-force model, the value chain, and others. 2. Teaching Focus We recommend presenting this case to students as a conceptual study rather than a story of one company. The main focus could be the viability of a free service business model. We question the viability of offering technology services with high associated variable cost per incremental user, substantially below cost, as a mechanism for creating market presence, growth, and future potential revenues. The free ISP business model offers technology services that generate little or no revenues from the customer, generates an uncertain advertising revenue stream, and has high variable cost as a mechanism for creating market presence and direct customer contact. Just imagine the added annual revenues from Spinway’s 6 million users paying only $5 per month. It is advised that students research financials and business reports on NetZero (or another publicly traded free ISP) since its business model is (or was) very similar to that of Spinway, while information is publicly available. An abbreviated balance sheet and income statement for year ended June 2000 is attached to the case. NetZero’s 10-K contains business model description and information on customers, competitors, vendors, and technology that may be helpful to students unfamiliar with the ISP business. The first set of case questions (Set A, attached) are a useful aid in helping the students understand the technology behind ISPs. The second set of case questions (Set B, attached) allow the students to practice specific skills related to building business models. We strongly recommend that these skill-building exercises are used to allow the students to take away practical as well as theoretical knowledge. 3. Technology Failure and Red Herrings The case has some “red herrings” in it. Any referral to the user experience with Spinway is only slightly relevant to the discussion. Since Spinway managed to attract as many as 6 million subscribers, user satisfaction with the service must have been on an acceptable level. Problems of the company may stem from an infeasible business model and possibly mismanagement, but probably not from lack of features in the software. Also, we do not know if users would continue using Spinway as their ISP if it, as a paid service, met all of their requirements. If so, what would be their threshold price? It is difficult to evaluate how much Spinway met, or did not meet, users’ expectations without this data. Therefore, students may want to largely ignore the technical side of the story and concentrate on the free ISP model itself. Interestingly, we believe that the technical side is the area students will concentrate on if given the chance, just as Spinway did. Contrary to popular belief, we believe there is very little proof that the technology itself failed to deliver. It did not fail substantially for advertisers, nor did it for customers. In our viewpoint, the failure was their financial structure/model rather than a technical failure, as alluded to by the former Spinway CEO, Steve Seabolt, a few months after the bankruptcy (Krause 2001). Heavy cash burn rates due to high variable costs and low revenues, while failing to obtain sufficient long-term financing, resulted in bankruptcy. Technological improvements would unlikely have altered this fate. Evidence supporting our hypothesis is the continued success of online advertising engines like Commission Junction where advertisers spend money online, particularly on banner ads, as well as the continuation of NetZero/Juno and other free ISP’s who now cover variable cost through low monthly subscription rates. Rolland et al.: The Free ISP Model and Spinway, Inc. INFORMS Transactions on Education 5(3), pp. 55–69, © 2005 INFORMS 4. The Free Business Model To address the “free model” issues, many relevant questions arise. How should you price your product in an emerging market? How much money can you afford to loose in order to establish your market presence? When and how do you convert moneylosing customers into profitable customers? How does the cost structure of the product, in terms of variable and fixed costs, affect the firm’s pricing options? How does long-term financing affect the firm’s pricing options? In Spinway’s case, did the technology fail to meet the needs of the advertisers, or did it fail to meet the customer needs? Both of these beliefs are common about ISPs. Alternatively to the technological failure explanation, did Spinway simply fail to meet their own financial needs? Case Questions—A: The Information Systems and Technology Perspective 1. Describe the telecommunications infrastructure for an ISP? Specify differences between dial-up needs versus DSL/Cable modem connection needs. 2. Assuming that the majority of the market is covered by DSL and cable modem connections, is there a need for a dial-up ISP? If so, what is the role of the dial-up ISPs now? 3. Is there a need for dial-up ISPs in the long run? 4. Discuss the expenses an ISP incurs and how they can be managed. 5. Do you believe that free Internet access should be of the same quality as paid dial-up service? 6. Can the quality of service be controlled or guaranteed by the ISP? If so, how? Case Questions—B: The Business Model Perspective The following questions are well suited for collaborative group work. These questions require diverse thinking, and also serious research into the salient issues. The students need to use library and web resources extensively to aid in finding good answers. For most of these, requiring the students to put together PowerPoint presentation works well. Using a page limit of about 18 slides is recommended. #1: Segmentation 1. What should the market segmentation have looked liked in 1999 in order for Spinway to be successful? 69 2. What should the market segmentation look like today in order to be successful in a dial-up ISP business (dial-up in 2004 could, for example, be viewed as “occasional connections” outside the home and the office)? #2: Value System 1. What is the industry value chain for Spinway in a. 1999? b. today? 2. What should Spinway’s value system look like in: a. 1999? b. today? #3: Product Positioning 1. Determine the product positioning for Spinway in terms of its composition (physical, service, information) a. What was the positioning in 1999? b. What should it be today? #4: The Business Model Create a summary of Spinway’s business model the way you envision that it must have been. You must address the bullets and sub-bullets below if applicable: • Value Cluster:
Competitive Advantage
Value System
Segmentation • Marketspace offering:
Product Specification
Possible product transformation • Resource System:
Channeling
Boundary Issues
Market and Firm Restructuring • Financial:
Capital vs. Expense
Revenue Streams
Cost
Give your assessment as to the sustainability of the business model (does it lead to a sustainable competitive advantage?) You need to be creative in how you specify the model, and you may want to use figures and graphs to the extent possible. For example, a figure for the segmentation would be good. You need to specify an actionable and meaningful segmentation. Also, since you have little information about the resource system and the financials, you may want to search the library (system) extensively to find information related to these issues.