SINESSSCHOOL 9-602-037 REV: MARCH 13. 2003 ANDREW MCAFEE Webvan On January 12, 2001, the Webvan Group was notified that it had 90 days to bring its stock over a dollar for 10 consecutive trading days to avoid being delisted from the Nasdaq: it had failed to meet a minimum bid price of $l for 30 consecutive days. In April, the company sought $25 million to stay afloat. At the end of trading on July 6, 2001, Webvan s stock stood at 6 cents a share. The following Monday, Webvan, the second best-ever-financed on-line retailer, announced that it was filing for Chapter 11 bankruptcy protection and would lay off most of its 2,000 employees. On the morning of July 9, many workers learned of their fates as they arrived at work only to find the gates locked This fall from grace in the eyes of investors was in some ways puzzling. At one point, the ompany had been the best capitalized of the online grocers, with a war chest of approximately $800 million raised from private investors and the public via an initial public offering(IPO).Many customers who had used Webvan's delivery service loved it; and for three straight quarters in 2000, it had been voted the best online grocer of 12 in a survey. In some markets, the company had exceeded the average consumer order size that it had projected for itself at this point in time and in the last quarter of 2000, it had posted a gross margin of 27% that compared favorably with the 27% to 30% seen by large conventional grocers. It had also continued to innovate initiating a successful program to take advantage of under utilized daytime delivery capaci However, Webvan had also projected in its 1999 prospectus that the expensive distribution centers DCs)it used were likely to be operating at breakeven capacity within five quarters of being launched. At the end of fourth quarter of 2000, six quarters had passed since Webvan's flagship DC in Oakland, CA, began operation and the company had not yet hit this widely publicized target. In addition, the company had faced several unanticipated difficulties with its operations and with convincing enough consumers to purchase groceries often enough. In September 2000, Georg Shaheen, the companys CEO, announced that Webvan would indefinitely delay expansion into the Washington, D.C., Baltimore, and Bergen Country, N] markets as had originally been planned although the company had already spent $45 million on the DCs to serve these areas. At the start of 2001, the company had $212 million in cash and a perilously high cash"burn rate. It chose to cut back marketing expenses stating that it would instead focus on getting current customers to order more lently. Customers were using the companys services only 1.7 times a month, and were often having difficulty getting a close-in"delivery window. ed this case under the supervision rew mcAfee. This case was HBS cases are developed solely as the basis for class discussion. Cases are not intended to serve as endorsements, sources of primary data, or illustrations of effective or ineffective management Copyright o 2001 President and Fellows of Harvard College. To order luce materials, call 1-800-5-45-7685, rite Harvard Business School Publishing, Boston, MA part of this publication may be reproduced, stored in a retrieval system, used in a spreadsheet, or form or by any means-electronic, mechanical photocopying, recording, or otherwise--without the permission of Harvard Business School
9-602-037 REV: MARCH 13, 2003 Research Associate Mona Ashiya prepared this case under the supervision of Professor Andrew McAfee. This case was developed from published sources. HBS cases are developed solely as the basis for class discussion. Cases are not intended to serve as endorsements, sources of primary data, or illustrations of effective or ineffective management. Copyright © 2001 President and Fellows of Harvard College. To order copies or request permission to reproduce materials, call 1-800-545-7685, write Harvard Business School Publishing, Boston, MA 02163, or go to http://www.hbsp.harvard.edu. No part of this publication may be reproduced, stored in a retrieval system, used in a spreadsheet, or transmitted in any form or by any means—electronic, mechanical, photocopying, recording, or otherwise—without the permission of Harvard Business School. ANDREW MCAFEE MONA ASHIYA Webvan On January 12, 2001, the Webvan Group was notified that it had 90 days to bring its stock over a dollar for 10 consecutive trading days to avoid being delisted1 from the Nasdaq: it had failed to meet a minimum bid price of $1 for 30 consecutive days.2 In April, the company sought $25 million to stay afloat.3 At the end of trading on July 6, 2001, Webvan’s stock stood at 6 cents a share.4 The following Monday, Webvan, the second best-ever-financed on-line retailer, announced that it was filing for Chapter 11 bankruptcy protection and would lay off most of its 2,000 employees.5 On the morning of July 9, many workers learned of their fates as they arrived at work only to find the gates locked.6 This fall from grace in the eyes of investors was in some ways puzzling. At one point, the company had been the best capitalized of the online grocers, with a war chest of approximately $800 million raised from private investors and the public via an initial public offering (IPO). Many customers who had used Webvan’s delivery service loved it; and for three straight quarters in 2000, it had been voted the best online grocer of 12 in a survey.7 In some markets, the company had exceeded the average consumer order size that it had projected for itself at this point in time and in the last quarter of 2000, it had posted a gross margin of 27% that compared favorably with the 27% to 30% gross margins typically seen by large conventional grocers. It had also continued to innovate, initiating a successful program to take advantage of under utilized daytime delivery capacity. However, Webvan had also projected in its 1999 prospectus that the expensive distribution centers (DCs) it used were likely to be operating at breakeven capacity within five quarters of being launched.8 At the end of fourth quarter of 2000, six quarters had passed since Webvan’s flagship DC in Oakland, CA, began operation and the company had not yet hit this widely publicized target. In addition, the company had faced several unanticipated difficulties with its operations and with convincing enough consumers to purchase groceries often enough. In September 2000, George Shaheen, the company’s CEO, announced that Webvan would indefinitely delay expansion into the Washington, D.C., Baltimore, and Bergen Country, NJ markets as had originally been planned although the company had already spent $45 million on the DCs to serve these areas.9 At the start of 2001, the company had $212 million in cash and a perilously high cash “burn rate.” It chose to cut back marketing expenses stating that it would instead focus on getting current customers to order more frequently.10 Customers were using the company’s services only 1.7 times a month, and were often having difficulty getting a close-in “delivery window
602-037 Webvan Observers of Webvan had long voiced doubts about the companys viability. As news of the losedown spread, many people recalled an April 1999 Wall Street Journal article that wondered whether the company would become the Internet eras equivalent of the movie"Waterworld, "a disaster so epic it becomes American legend. This had, in fact, come to pass, but why? Why did Webvan fail so dramatically? And why had no other company decided to purchase its assets ts including elaborate nationwide physical and IT infrastructures? Were these really without values? Company Founding Genesis of idea Louis Borders, the founder of bookseller The Borders Group Inc, had an epiphany when he opened a Fed Ex package containing some Japanese spices and other specialty items that he had ordered from a catalog in 1997. He recalled, "I kept thinking that I would need a loading dock outside my house to use the Internet to buy things. And on top of that, I'd have to pay $10 per package He came to believe that online retail would never make it unless there could be a faster of the growing number of people making purchases online by creating an enterprise that would offer greater variety than conventional stores and yet provide the convenience and instant gratification chat online shoppers missed Borders incorporated Intelligent Systems for Retail in December 1996 and in April 199 ts name to the Webvan Group. The mission of the new Foster City-based company was"to deliver the last mile of e-commerce, "meaning the delivery of merchandise from a DC to a customers doorstep. The last mile was also referred to as"the gilded pathway into the homes and hearts of Americas consumers that everyone from Amazon. com to Wal-Mart to Fred Smiths FedEx lusts The Webvan Group planned to begin by offering groceries which people shop for frequently build critical mass, order frequency, and economies of scale. With an established customer base, it then planned to leverage its distribution system to expand to other categories, adding items such as consumer electronics and books whose profit margins were considerably greater than for groceries but were ordered less frequently. That is, they planned to attract an audience first and then monetise those eyeballs"to bring in additional revenue and do this on a global scale. From 1997 to May 1999, the company focused on developing the web store and constructing its first distribution and fulfillment center in the San Francisco Bay Area. Following a trial of its grocery delivery service in May 1999 to 1,100 people, the web store was launched in June 1999 as a venture, whose mission was to deliver everything from groceries to palm pilots to consumers in a cheap and Borders stated, " Intuitively, I knew I'd have a great financial model if I could eliminate osts. However, where he saw"a cornucopia of opportunity, "there were those who wondered whether people were in fact going to migrate their purchases of groceries and many other items to the Internet, and whether the new company could crack the"last mile"problem
602-037 Webvan 2 Observers of Webvan had long voiced doubts about the company’s viability. As news of the closedown spread, many people recalled an April 1999 Wall Street Journal article that wondered whether the company would become the Internet era’s equivalent of the movie “Waterworld,” a disaster so epic it becomes American legend.”11 This had, in fact, come to pass, but why? Why did Webvan fail so dramatically? And why had no other company decided to purchase its assets, including elaborate nationwide physical and IT infrastructures? Were these really without values? Company Founding Genesis of Idea Louis Borders, the founder of bookseller The Borders Group Inc., had an epiphany when he opened a FedEx package containing some Japanese spices and other specialty items that he had ordered from a catalog in 1997. He recalled, “I kept thinking that I would need a loading dock outside my house to use the Internet to buy things. And on top of that, I’d have to pay $10 per package.”12 He came to believe that online retail would never make it unless there could be a faster, cheaper and more efficient way of delivering items to consumers.13 Borders wanted to take advantage of the growing number of people making purchases online by creating an enterprise that would offer greater variety than conventional stores and yet provide the convenience and instant gratification that online shoppers missed. Borders incorporated Intelligent Systems for Retail in December 1996 and in April 1999 changed its name to the Webvan Group. The mission of the new Foster City-based company was “to deliver the last mile of e-commerce,” meaning the delivery of merchandise from a DC to a customer’s doorstep. The last mile was also referred to as “the gilded pathway into the homes and hearts of America’s consumers that everyone from Amazon.com to Wal-Mart to Fred Smith’s FedEx lusts after.”14 The Webvan Group planned to begin by offering groceries which people shop for frequently to build critical mass, order frequency, and economies of scale. With an established customer base, it then planned to leverage its distribution system to expand to other categories, adding items such as consumer electronics and books whose profit margins were considerably greater than for groceries but were ordered less frequently.15 That is, they planned to attract an audience first and then “monetise those eyeballs” to bring in additional revenue and do this on a global scale.16 From 1997 to May 1999, the company focused on developing the web store and constructing its first distribution and fulfillment center in the San Francisco Bay Area. Following a trial of its grocery delivery service in May 1999 to 1,100 people, the web store was launched in June 1999 as a venture, whose mission was to deliver everything from groceries to palm pilots to consumers in a cheap and efficient manner. 17 Borders stated, “Intuitively, I knew I’d have a great financial model if I could eliminate store costs.”18 However, where he saw “a cornucopia of opportunity,” there were those who wondered whether people were in fact going to migrate their purchases of groceries and many other items to the Internet, and whether the new company could crack the “last mile” problem.19
Webvan 602-03 Management Team To realize his vision, Borders recruited a management"dream team. "Borders himself served as first Chairman and CEO of the company. Borders had developed the advanced information systems used by the company he had founded earlier, Borders Books, to manage inventory across the count and had also co-founded another company, Synergy Software, in 1989. He was also instrumental i designing the warehouses with intricate processes that the Webvan Group would rely on In September 1999, Borders recruited George Shaheen from his role as CEO of Andersen Consulting. Shaheen, whose first job was bagging groceries and trimming beef part-time in Elmwood, IL, had spent 32 years at Andersen and had helped create the world's largest consulting firm. He stood to collect a $10 million bonus upon retirement from Andersen Consulting in 11 months plus additional payouts from his group's venture investments. At the Webvan Group, with a salary of $500,000, a ninth of what it was at Andersen Consulting, Shaheen held 1.25 million shares of stock, in addition to 15 million options. If the company did well, Shaheens stake would have been worth $100 million. The company also hired senior executives from Goldman Sachs Co, Oracle Corporation, Federal Express, American Stores Company, Marriott International, and General Electric. See Exhibit 1 for more background on the companys management team. David Beirne, a managing member of Benchmark Capital, served as a member of the board in addition to Christos Cotsakos who was the ceo and Chairman of the board for e+Trade. Tim Koogle, the CEO of Yahoo! Inc, and Michael Mortiz, general partner at Sequioa Capital, also sat on Webvan's board funding Sequoia Capital, two leading venture capital firms. There had been no online grocery successes at this point in time: the best known online supermarket, Peapod Inc, had a $21 million loss on revenue of only $69 million in 1998 and another competitor, Netgrocer Inc. had laid off employees in the fall of 1998 after poor results. The Webvan Group, however, pointed to the inability of these online grocers to offer same-day delivery, a narrow delivery window or a highly automated and scalable business model. 24 $400 million in four rounds of venture capital financing, the most for an Internet company in 1999 Commenting on the companys valuation, Jim Barksdale, the former CEO of Netscape, observed, "I've never seen anything like it. Webvan announced in August 1999 that it planned an initial public offering where it hoped to raise an additional $300 million. David Beirne of Benchmark Capital, one of the first to finance the company noted, " This could be the biggest company to come out of Silicon Valley. By the time of its postponed initial public offering on November 5, 1999 which was underwritten by Goldman Sachs Co, the company had a only a meager operational record and had collected only $4 million in revenue. Nonetheless, the company managed to raise an additional $400 million in its IPO. On its first day of trading, the companys initial share price of $15 rose at one point to $34 giving it $15 billion capitalization before falling to $25 by the end of the day( Exhibit 2 provides a chart showing the company's stock price
Webvan 602-037 3 Management Team To realize his vision, Borders recruited a management “dream team.” Borders himself served as first Chairman and CEO of the company. Borders had developed the advanced information systems used by the company he had founded earlier, Borders Books, to manage inventory across the country and had also co-founded another company, Synergy Software, in 1989. He was also instrumental in designing the warehouses with intricate processes that the Webvan Group would rely on.20 In September 1999, Borders recruited George Shaheen from his role as CEO of Andersen Consulting. Shaheen, whose first job was bagging groceries and trimming beef part-time in Elmwood, IL, had spent 32 years at Andersen and had helped create the world’s largest consulting firm. He stood to collect a $10 million bonus upon retirement from Andersen Consulting in 11 months plus additional payouts from his group’s venture investments. At the Webvan Group, with a salary of $500,000, a ninth of what it was at Andersen Consulting, Shaheen held 1.25 million shares of stock, in addition to 15 million options. If the company did well, Shaheen’s stake would have been worth $100 million.21 The company also hired senior executives from Goldman Sachs & Co., Oracle Corporation, Federal Express, American Stores Company, Marriott International, and General Electric. See Exhibit 1 for more background on the company’s management team. David Beirne, a managing member of Benchmark Capital, served as a member of the board in addition to Christos Cotsakos, who was the CEO and Chairman of the Board for E*Trade. Tim Koogle, the CEO of Yahoo! Inc., and Michael Mortiz, general partner at Sequioa Capital, also sat on Webvan’s board. Funding By April 1999, the Webvan Group had already attracted $120 million in funding from high-profile backers such as CBS Inc., Knight-Ridder Co., Softbank Co. of Japan, as well as Benchmark Capital and Sequoia Capital, two leading venture capital firms.22 There had been no online grocery successes at this point in time: the best known online supermarket, Peapod Inc., had a $21 million loss on revenue of only $69 million in 1998 and another competitor, Netgrocer Inc. had laid off employees in the fall of 1998 after poor results.23 The Webvan Group, however, pointed to the inability of these online grocers to offer same-day delivery, a narrow delivery window or a highly automated and scalable business model.24 By July 1999, Webvan had garnered an additional $275 million by selling a 6.48% stake to Goldman Sachs & Co., Softbank Co. and Sequoia Capital.25 In the end, it had raised a total of about $400 million in four rounds of venture capital financing, the most for an Internet company in 1999.26 Commenting on the company’s valuation, Jim Barksdale, the former CEO of Netscape, observed, “I’ve never seen anything like it.”27 Webvan announced in August 1999 that it planned an initial public offering where it hoped to raise an additional $300 million. David Beirne of Benchmark Capital, one of the first to finance the company noted, “This could be the biggest company to come out of Silicon Valley.”28 By the time of its postponed initial public offering29 on November 5, 1999 which was underwritten by Goldman Sachs & Co., the company had a only a meager operational record and had collected only $4 million in revenue. Nonetheless, the company managed to raise an additional $400 million in its IPO. On its first day of trading, the company’s initial share price of $15 rose at one point to $34 giving it $15 billion capitalization before falling to $25 by the end of the day30 (Exhibit 2 provides a chart showing the company’s stock price
602-037 Webvan point,bristled,"They have the sales of two of our stores and one fourth of our market cap. "29 over time). Its day-end $8 billion market capitalization was about 40% of the market value of the nation's largest supermarket chain, Kroger, which had 2, 200 stores and a share price of $23. Noting the companys valuation, one Safeway executive whose company had a share price of $40 at that Size of market In its prospectus filed with the Securities and Exchange Commission( SEC) in November 1999, the Webvan Group noted that the total US market for groceries, drugstore merchandise and prepared meals alone was over $650 billion in 1998, and that International Data Corporation(IDC)estimated that the 63 million web users in the United States at the end of 1998 were projected to grow to 177 million users by the end of 2003. Consumer purchases of goods and services online were projected to increase from $12.4 billion in 1998 to $75 billion by 2003. Specifically, online grocery purchases that amounted to $235 million in 1998 and $519 million in 1999 were expected to grow to $3. 5 billion by 2002, $10.8 billion by 2003, and rise to $16.8 billion by 2004; they would represent 2% of the total grocery market. Forrester Research projected that 5% of U.S. households would be buying groceries online in a few years while Jupiter Communications forecast in 1999 that the online grocery market would be worth about $3.5 billion in 2000 and $6.5 billion by 2003. Webvan's aim was to capture a substantial chunk of this growing market. The company noted that traditional grocery stores had to limit the number of items they carried in their stores. In addition, traditional stores had to face the costs of building and operating stores near residential areas,and had significant costs associated with personnel, store set-up and inventory. Webvan sought to provide a more cost-effective solution. Ultimately, Shaheen never saw groceries as the companys"end-game"-he saw the market as $1.5 trillion, an IDC projection for 2003, which encompassed all web-based purchases. Original Business model Based in Foster City, CA, the Webvan Group aimed to deliver dry and perishable goods to consumers' homes, at competitive prices, within a specified 30-minute window. Customers who logged onto the Webvan Groups web site upon launch were able to choose from 15,000 grocery tems at prices that were initially competitive with off-line retailers. Offerings at the company drugs.Customers were able to place orders anytime, pay by credit card wel s sakery items, non- included fresh produce, premium meats, fresh seafood, prepared-meals perishable items found in conventional supermarkets, wines, cigars,as s non within a 30-minute window the same day or up to four days later. Webvan did not require any membership fees, and delivery was to be free for orders over $50 with smaller orders carrying a $4.95 surcharge. With its staff of courteous and friendly drivers who did not expect tips, the company hoped to quickly gain a large number of customers who would be happier about convenience and they planned to expand across the country to other markets as well as expand their product offerings The challenge faced by the Webvan Group was the delivery of perishable and dry goods at competitive prices to consumers within a 30-minute window chosen by the customer on a scale never
602-037 Webvan 4 over time). Its day-end $8 billion market capitalization was about 40% of the market value of the nation’s largest supermarket chain, Kroger, which had 2,200 stores and a share price of $23.31 Noting the company’s valuation, one Safeway executive whose company had a share price of $40 at that point, bristled, “They have the sales of two of our stores and one fourth of our market cap.”32 Size of Market In its prospectus filed with the Securities and Exchange Commission (SEC) in November 1999, the Webvan Group noted that the total US market for groceries, drugstore merchandise and prepared meals alone was over $650 billion in 1998,33 and that International Data Corporation (IDC) estimated that the 63 million web users in the United States at the end of 1998 were projected to grow to 177 million users by the end of 2003.34 Consumer purchases of goods and services online were projected to increase from $12.4 billion in 1998 to $75 billion by 2003.35 Specifically, online grocery purchases that amounted to $235 million in 1998 and $519 million in 1999 were expected to grow to $3.5 billion by 2002, $10.8 billion by 2003, and rise to $16.8 billion by 2004; they would represent 2% of the total grocery market.36 Forrester Research projected that 5% of U.S. households would be buying groceries online in a few years37 while Jupiter Communications forecast in 1999 that the online grocery market would be worth about $3.5 billion in 2000 and $6.5 billion by 2003.38 Webvan’s aim was to capture a substantial chunk of this growing market. The company noted that traditional grocery stores had to limit the number of items they carried in their stores. In addition, traditional stores had to face the costs of building and operating stores near residential areas, and had significant costs associated with personnel, store set-up and inventory. Webvan sought to provide a more cost-effective solution.39 Ultimately, Shaheen never saw groceries as the company’s “end-game”—he saw the market as $1.5 trillion, an IDC projection for 2003, which encompassed all web-based purchases.40 Original Business Model Overview Based in Foster City, CA, the Webvan Group aimed to deliver dry and perishable goods to consumers’ homes, at competitive prices, within a specified 30-minute window. Customers who logged onto the Webvan Group’s web site upon launch were able to choose from 15,000 grocery items at prices that were initially competitive with off-line retailers.41 Offerings at the company included fresh produce, premium meats, fresh seafood, prepared-meals, bakery items, nonperishable items found in conventional supermarkets, wines, cigars, as well as non-prescription drugs.42 Customers were able to place orders anytime, pay by credit card, and schedule deliveries within a 30-minute window the same day or up to four days later.43 Webvan did not require any membership fees, and delivery was to be free for orders over $50 with smaller orders carrying a $4.95 surcharge. With its staff of courteous and friendly drivers who did not expect tips, the company hoped to quickly gain a large number of customers who would be happier about convenience and service than getting discounts. Thereafter, with an established customer base in one geographic area, they planned to expand across the country to other markets as well as expand their product offerings. The challenge faced by the Webvan Group was the delivery of perishable and dry goods at competitive prices to consumers within a 30-minute window chosen by the customer on a scale never
Webvan 602-032 before attempted. The grocery category also provided a challenge because of its razor-thin margins of 1% to 1.5% and the wide product range and temperature requirements. Webvan knew that the last mile to the consumer" posed an enormous logistical problem, especially with groceries, but as Shaheen expressed it, " two or three companies will legitimately earn the right to cross into a persons home. We intend to be one of those Customers Motivating consumers to shop online was viewed as an uphill battle since many people did not want to pay delivery charges and everyone demanded high service standards. Moreover, going to a grocery store on a weekly basis was highly ingrained behavior. However, the company believed that the"market demand for high quality reliable grocery services [was] enormous and [was] very much like the pent-up demand for high quality wide-bandwidth communications. 6. Shaheen saw the company's target customers as women and, more specifically, soccer moms- that is, mothers in urban two-income families with a combined income of more than $75, 000 and little spare time. He noted that 1% to 3% of this target market would make the economics of the new company work. People made an average of 2. 4 trips to the grocery store per week and spent an average of $4, 600 annually in the process, about 10% of their income. In a survey the company conducted in San Francisco to learn how many people would be likely to use an online grocer, 6% of respondents noted "absolutely, "23% said probably. Shaheen believed that 35%of the market would be buying groceries on the Internet by 2003-2004. The Webvan group also noted that most people viewed grocery shopping as an inconvenience and that nearly 55% of all Americans considered time put little economic value on their time in the supermarket. e s had apparently shown that consumers Operations Warehousing and order management To tap into the online consumer market, the Webvan was a highly e pproach that was capital and technology intensive. Central to Webvan's approach tomated central DC that could carry as many as 50,000 items and process 8,000 orders that moved through a series of put-away and pick pods, where goods could be loaded by workers G-4 a day. The company constructed its first DC in Oakland, CA, a 330,000 square feet warehouse that ould serve as many people as 18 conventional supermarkets and contained cooking facilities as well as multiple temperature zones to store items such as wine, cigars and ice cream. The DC had 4 carousels with a storage capacity of 107, 000 locations and contained about 5 miles of conveyor belt Goods received at the distribution center were broken down immediately from cases to"eaches (singular items) and transferred to trays, which were the containers used to store inventory. Each tray carried a unique bar code license plate, which was used to track all movement in the system. Product barcodes were scanned by workers and the number of items on each tray were entered in conjunction with each trays barcode license. To increase throughput, Webvan spread high-demand products across multiple carousels e Q When a customer placed an order online and specined a delivery time, the information was Iveyed to the central DC. There, the appropriate quantities of colored totes were assigned to the der; yellow totes were for used dry groceries, green totes for chilled products and blue totes carried frozen goods. Each tote was given a bar code license plate that linked to a specific customers order and could then be tracked throughout the fulfillment process. A custom-written piece of software
Webvan 602-037 5 before attempted. The grocery category also provided a challenge because of its razor-thin margins of 1% to 1.5% and the wide product range and temperature requirements.44 Webvan knew that the “last mile to the consumer” posed an enormous logistical problem, especially with groceries, but as Shaheen expressed it, “two or three companies will legitimately earn the right to cross into a person’s home. We intend to be one of those.”45 Customers Motivating consumers to shop online was viewed as an uphill battle since many people did not want to pay delivery charges and everyone demanded high service standards.46 Moreover, going to a grocery store on a weekly basis was highly ingrained behavior. However, the company believed that the “market demand for high quality reliable grocery services [was] enormous and [was] very much like the pent-up demand for high quality wide-bandwidth communications.”47 Shaheen saw the company’s target customers as women and, more specifically, soccer moms— that is, mothers in urban, two-income families with a combined income of more than $75,000 and little spare time.48 He noted that 1% to 3% of this target market would make the economics of the new company work.49 People made an average of 2.4 trips to the grocery store per week and spent an average of $4,600 annually in the process, about 10% of their income.50 In a survey the company conducted in San Francisco to learn how many people would be likely to use an online grocer, 6% of respondents noted “absolutely,” 23% said probably. Shaheen believed that 35% of the market would be buying groceries on the Internet by 2003-2004.51 The Webvan Group also noted that most people viewed grocery shopping as an inconvenience and that nearly 55% of all Americans considered time to be their most precious commodity. However, other studies had apparently shown that consumers put little economic value on their time in the supermarket.52 Operations Warehousing and order management To tap into the online consumer market, the Webvan Group took an approach that was capital and technology intensive. Central to Webvan’s approach was a highly automated central DC that could carry as many as 50,000 items and process 8,000 orders a day.53 The company constructed its first DC in Oakland, CA, a 330,000 square feet warehouse that could serve as many people as 18 conventional supermarkets and contained cooking facilities as well as multiple temperature zones to store items such as wine, cigars and ice cream.54 The DC had 41 carousels with a storage capacity of 107,000 locations and contained about 5 miles of conveyor belts that moved through a series of put-away and pick pods, where goods could be loaded by workers.55 Goods received at the distribution center were broken down immediately from cases to “eaches” (singular items) and transferred to trays, which were the containers used to store inventory. Each tray carried a unique bar code license plate, which was used to track all movement in the system. Product barcodes were scanned by workers and the number of items on each tray were entered in conjunction with each tray’s barcode license.56 To increase throughput, Webvan spread high-demand products across multiple carousels. When a customer placed an order online and specified a delivery time, the information was conveyed to the central DC. There, the appropriate quantities of colored totes were assigned to the order; yellow totes were for used dry groceries, green totes for chilled products and blue totes carried frozen goods. Each tote was given a bar code license plate that linked to a specific customer’s order and could then be tracked throughout the fulfillment process.57 A custom-written piece of software
602-037 Webvan would then devise the optimal "picking plan"for each tote, or the route it would follow through the DC. The totes routings were determined by the time required to pick each part f an order. th scheduled delivery time and the existing workload. Routes were also calculated so that the biggest, heaviest or most solidly packed items came first so that smaller, more delicate items would not be crushed. A series of scanners, fixed or hand-held, read each tote's barcode as it passed through the tem Workers known as"pickers"were stationed at the pods, and were responsible for assembling orders. Each picker could work on as many as 16 orders simultaneously and the company claimed that in a single hour, pickers could pack 450 grocery items-nearly 10 times the productivity of a shopper wheeling a cart through a supermarket. Pickers assembled orders from rotating carousels that carried items to them so that they would not have to walk more than 19 feet. An MIT trained mathematician, Borders reasoned that within 10-feet, not enough items could be near the"picker while 30 feet was too long for a"picker"to walk. "You want to keep people stationary so you can keep them busy", noted Gary Dahl, Webvan's vice-president for wholesale As a tote came to a pod, carousels would move into position to place the ordered items within reach of the picker. A"light tree"display next to the carousel bins would then tell the picker which items to pick and a"sort bar"display on the conveyor belts would indicate which items to put into which tote. Once an item was placed in a tote, the picker hit a button to confirm that the pick was complete; the next item then appeared or a signal let the picker know that the tote was complete After completion, the picker would close the tote and place it on an express carousel that took it to the shipping area. The process was timed to ensure that all totes in an order arrived at the shipping area at approximately the same time The Webvan Group had exclusive relationships with vendors who supplied the company with the onveyors, carousels and light displays. In fact, the company was so protective of its carousel technology that it did not allow the carousels to be photographed. Such automation, the company felt, would allow it to increase its offerings without a proportional increase in personnel. The company also claimed that as a result of the automation, grocery products were only handled an average of 8 times as compared to an average of 14 times for a conventional supermarket; Webvan maintained that its produce would therefore be of higher quality. However, not everyone was impressed with the technology. Some believed that the automation would only save the company 1% of the cost of groceries being sold. Others noted that 40% of the companys product volume was fresh and frozen food that the company was not able to automate because of the labor-intensive nature of picking/preparing fresh food. Another 25% of the companys offerings were fulfilled from the less automated mechanized pods that were used to store heavy or fast-moving items such as soda. The remaining 35% of items were picked using the highly automated carousel pods and analysts doubted that this would yield significant cost savings to Webvan. Delivery Once at the DCs dock, orders were separated by delivery route and loaded onto dollies that would be wheeled onto trucks that had ambient, chilled, and frozen compartments. The totes were then transported to one of 10 to 12 manned depots within a 50-mile radius of the dC (Federal Express served as the model for the companys hub and spoke delivery system). From the manned depots, individual orders were put on one of about 60 delivery vans that would then take ds to the customers. By design, van drivers, called"couriers"would not have to travel more than miles in any given direction to reach a customer The Webvan courier delivered the order to a customers home and unpacked the totes. Using wireless hand held device the courier could print out a receipt and an itemized list of the order and
602-037 Webvan 6 would then devise the optimal “picking plan” for each tote, or the route it would follow through the DC. The totes’ routings were determined by the time required to pick each part of an order, the scheduled delivery time and the existing workload. Routes were also calculated so that the biggest, heaviest or most solidly packed items came first so that smaller, more delicate items would not be crushed.58 A series of scanners, fixed or hand-held, read each tote’s barcode as it passed through the system.59 Workers known as “pickers” were stationed at the pods, and were responsible for assembling orders.60 Each picker could work on as many as 16 orders simultaneously and the company claimed that in a single hour, pickers could pack 450 grocery items—nearly 10 times the productivity of a shopper wheeling a cart through a supermarket.61 Pickers assembled orders from rotating carousels that carried items to them so that they would not have to walk more than 19 feet. An MIT trained mathematician, Borders reasoned that within 10-feet, not enough items could be near the “picker” while 30 feet was too long for a “picker” to walk. “You want to keep people stationary so you can keep them busy”, noted Gary Dahl, Webvan’s vice-president for wholesale.62 As a tote came to a pod, carousels would move into position to place the ordered items within reach of the picker. A “light tree” display next to the carousel bins would then tell the picker which items to pick and a “sort bar” display on the conveyor belts would indicate which items to put into which tote. Once an item was placed in a tote, the picker hit a button to confirm that the pick was complete; the next item then appeared, or a signal let the picker know that the tote was complete. After completion, the picker would close the tote and place it on an express carousel that took it to the shipping area.63 The process was timed to ensure that all totes in an order arrived at the shipping area at approximately the same time. The Webvan Group had exclusive relationships with vendors who supplied the company with the conveyors, carousels and light displays. In fact, the company was so protective of its carousel technology that it did not allow the carousels to be photographed.64 Such automation, the company felt, would allow it to increase its offerings without a proportional increase in personnel.65 The company also claimed that as a result of the automation, grocery products were only handled an average of 8 times as compared to an average of 14 times for a conventional supermarket; Webvan maintained that its produce would therefore be of higher quality.66 However, not everyone was impressed with the technology. Some believed that the automation would only save the company 1% of the cost of groceries being sold.67 Others noted that 40% of the company’s product volume was fresh and frozen food that the company was not able to automate because of the labor-intensive nature of picking/preparing fresh food. Another 25% of the company’s offerings were fulfilled from the less automated mechanized pods that were used to store heavy or fast-moving items such as soda. The remaining 35% of items were picked using the highly automated carousel pods and analysts doubted that this would yield significant cost savings to Webvan.68 Delivery Once at the DCs dock, orders were separated by delivery route and loaded onto dollies that would be wheeled onto trucks that had ambient, chilled, and frozen compartments.69 The totes were then transported to one of 10 to 12 manned depots within a 50-mile radius of the DC (Federal Express served as the model for the company’s hub and spoke delivery system). From the manned depots, individual orders were put on one of about 60 delivery vans that would then take goods to the customers. By design, van drivers, called “couriers” would not have to travel more than 10 miles in any given direction to reach a customer. The Webvan courier delivered the order to a customer’s home and unpacked the totes. Using a wireless hand held device the courier could print out a receipt and an itemized list of the order and
Webvan 602-032 check items against the order list. The mobile device also gave a courier access to a customers back orders so a customer could be immediately credited for any returns they might have. If a delay occurred, customers would be informed, refunded $3, and given the option to reschedule delivery If a customer was not home, the company added a $4.95 charge for a re-delivery attempt. Speaking they can be our saving grace if there are problems, m2 Ou of the companys couriers, Shaheen commented, Couriers are our ambassadors of reliability and Information Technology The Webvan Group spent three years and hired 80 software programmers to create the proprietary systems linking and automating many aspects of its business process. It spent about $15 million on this process in 1999 up from $3 million the previous year and 0.2 million in 1997. In its 1999 annual report, the company claimed that it had spent 50 person-years of effort on software development The system would be aware of the location of every stock-keeping unit or"SKU, "and how those SKUs were on hand at a DC at any point in time. The company also invested heavily in inventory forecasting. Forecasts incorporated expiration dates, and could be changed on a daily basis to deal with changes in demand. Webvan also relied on internal benchmarking of its invento and distribution systems(for example, pick rates of products were monitored so that if demand lagged, the placement of products in the DC could be modified)"and claimed that it followed suggestions from its employees, many of whom had been in"pick environments"before &n the entire operation was supported by a warehouse management system from Optum Inc,which allowed integration of the companys custom-designed inventory, materials handling, order tracking, accounting, and transportation management systems. The design and complexity of Webvan's system was, however, something Optum had not dealt with before and the project pushed the company to expand its ability to interface with the material handling equipment. Webvan used Descartes Systems software to optimize delivery routes for a set of orders given a set of resource constraints such as number of delivery vans available. In addition, the software was used to optimize delivery schedules. When a customer placed an order, delivery slots were either taken; the system could rule it out based on the time and location of other committed deliveries. 2 as marked available or unavailable. Unavailability of a slot did not necessarily mean that the slot Ultimately, the set of scalable and replicable systems used, claimed Webvan, would no ot require as many warehouse employees, stock personnel or cashiers seen in conventional supermarkets. That is the system would allow an increase in volume without a concomitant increase in human resources Bud Grebey, the companys spokesman proclaimed, "If you want to fly to Mars, you don't build a rocket ship to carry you halfway there. The key is software that handles the logistics At the launch of its Oakland DC, the company's distribution system operated five days of the reek although the company anticipated full-week operation in a few months post launch. Supplier relations an efi pliers. In its IPO filing with y had to establish and develop relationships with wholesalers To stock its warehouses, the com and su SEC in 1999, the company noted that it offered its suppliers cient supply model, which would be reflected in the discounts and pricing of goods it received. For instance, Webvan noted the lower costs to suppliers who no longer had to come as
Webvan 602-037 7 check items against the order list. The mobile device also gave a courier access to a customer’s back orders so a customer could be immediately credited for any returns they might have. If a delay occurred, customers would be informed, refunded $3, and given the option to reschedule delivery.70 If a customer was not home, the company added a $4.95 charge for a re-delivery attempt.71 Speaking of the company’s couriers, Shaheen commented, “Couriers are our ambassadors of reliability and they can be our saving grace if there are problems.”72 Information Technology The Webvan Group spent three years and hired 80 software programmers to create the proprietary systems linking and automating many aspects of its business process.73 It spent about $15 million on this process in 1999 up from $3 million the previous year and 0.2 million in 1997.74 In its 1999 annual report, the company claimed that it had spent 50 person-years of effort on software development.75 The system would be aware of the location of every stock-keeping unit or “SKU,” and how many of those SKUs were on hand at a DC at any point in time.76 The company also invested heavily in inventory forecasting.77 Forecasts incorporated expiration dates, and could be changed on a daily basis to deal with changes in demand. Webvan also relied on internal benchmarking of its inventory and distribution systems (for example, pick rates of products were monitored so that if demand lagged, the placement of products in the DC could be modified)78 and claimed that it followed suggestions from its employees, many of whom had been in “pick environments” before.79 The entire operation was supported by a warehouse management system from Optum Inc., which allowed integration of the company’s custom-designed inventory, materials handling, order tracking, accounting, and transportation management systems.80 The design and complexity of Webvan’s system was, however, something Optum had not dealt with before and the project pushed the company to expand its ability to interface with the material handling equipment.81 Webvan used Descartes System’s software to optimize delivery routes for a set of orders given a set of resource constraints such as number of delivery vans available. In addition, the software was used to optimize delivery schedules. When a customer placed an order, delivery slots were either marked available or unavailable. Unavailability of a slot did not necessarily mean that the slot was taken; the system could rule it out based on the time and location of other committed deliveries.82 Ultimately, the set of scalable and replicable systems used, claimed Webvan, would not require as many warehouse employees, stock personnel or cashiers seen in conventional supermarkets. That is, the system would allow an increase in volume without a concomitant increase in human resources.83 Bud Grebey, the company’s spokesman proclaimed, “If you want to fly to Mars, you don’t build a rocket ship to carry you halfway there. The key is software that handles the logistics.”84 At the launch of its Oakland DC, the company’s distribution system operated five days of the week although the company anticipated full-week operation in a few months post launch.85 Supplier Relations To stock its warehouses, the company had to establish and develop relationships with wholesalers and suppliers. In its IPO filing with the SEC in 1999, the company noted that it offered its suppliers an efficient supply model, which would be reflected in the discounts and pricing of goods it received.86 For instance, Webvan noted the lower costs to suppliers who no longer had to come as
602-037 Webvan often to replenish products as they had to at conventional supermarkets. When the company opened for operations in 1999, it was purchasing goods from 10 distributors and directly from 160 vendors Purchases were based on a combination of anticipated and actual demand. Because of the 24-hour minimum lead-time for an online order the company had more demand-based information that it could transmit to suppliers. To communicate this information to suppliers and to ensure just-in-time order fulfillment and real-time drop delivery of its products, Webvan partnered with Harbinger.net, in January 2000. The companys top tier suppliers began using its system soon thereafter and the company planned to have all its suppliers linked in this fashion in the future. The company provided its suppliers with information on consumer choice and wanted merchandise sent to it by suppliers in packages and forms that would make Webvan s merchandise processing more efficient Specifically, unlike bricks-and-mortar retailers that dealt in pallets, Webvan dealt in "eaches (singular items) and wanted its inbound shipments to arrive in a form that could be broken down quickly and moved to the pods and carousels. Shaheen noted, "We are not a pallet operation Were working with suppliers so they can re-engineer their supply chains that come to us Shaheen also wanted to avoid the costs associated with fully stocking all of Webvan's DCs Instead, he planned to create connections with suppliers and manufacturers along the companys supply chain so that only fast-moving items would be stocked at Webvan,'s DCs. Webvan's partners would carry the rest and delivery them on a just-in-time basis; Webvan's DCs would include"cross docking"facilities where these supplier deliveries would be integrated with items picked from within the dC to complete each customers order. In exchange for these considerations, Webvan chose not to charge manufacturers the"slotting fees"they normally had to pay to get a product on a store's For brand name suppliers such as Pillsbury, General Mills and Proctor Gamble, the online grocery market was uncharted territory, and one where they did not have to compete with generics, to them, Webvan offered to the opportunity to deliver product samples directly to consumers. In addition, Webvan could also provide manufacturers information such as what individual consumers bought and whether a particular product made it on a customer's"re-purchase"list. Marketing To entice prospective customers in the San Francisco Bay Area to try its service upon launch in June 1999, the company waived delivery fees regardless of order size through July 31, 1999. In addition, the Webvan Group advertised that its prices were 5% lower than conventional grocery stores. Following the launch of its service in June 1999, Webvan planned to increase spending substantially on its radio and newspaper advertising campaigns to build brand identity and loyalty It also planned to leverage its relationship with media investors such as CBS and Knight-Ridder And with the data it gathered on consumer purchasing, preferences and behavior, it planned to offer targeted and customized services. In 1999, the company announced that it planned to spend $6 to $9 million annually in advertising for each city and a total of $200 million on advertising in the year Growth and Efficiency Projections Webvan planned a national rollout to achieve economies of scale since all of the 26 DCs the ompany planned would conform to a single design, and be managed with a uniform technology
602-037 Webvan 8 often to replenish products as they had to at conventional supermarkets.87 When the company opened for operations in 1999, it was purchasing goods from 10 distributors and directly from 160 vendors.88 Purchases were based on a combination of anticipated and actual demand. Because of the 24-hour minimum lead-time for an online order the company had more demand-based information that it could transmit to suppliers. To communicate this information to suppliers and to ensure just-in-time order fulfillment and real-time drop delivery of its products, Webvan partnered with Harbinger.net, in January 2000.89 The company’s top tier suppliers began using its system soon thereafter and the company planned to have all its suppliers linked in this fashion in the future.90 The company provided its suppliers with information on consumer choice and wanted merchandise sent to it by suppliers in packages and forms that would make Webvan’s merchandise processing more efficient.91 Specifically, unlike bricks-and-mortar retailers that dealt in pallets, Webvan dealt in “eaches” (singular items) and wanted its inbound shipments to arrive in a form that could be broken down quickly and moved to the pods and carousels. Shaheen noted, “We are not a pallet operation. We’re working with suppliers so they can re-engineer their supply chains that come to us.”92 Shaheen also wanted to avoid the costs associated with fully stocking all of Webvan’s DCs. Instead, he planned to create connections with suppliers and manufacturers along the company’s supply chain so that only fast-moving items would be stocked at Webvan’s DCs. Webvan’s partners would carry the rest and delivery them on a just-in-time basis; Webvan’s DCs would include “cross docking” facilities where these supplier deliveries would be integrated with items picked from within the DC to complete each customer’s order.93 In exchange for these considerations, Webvan chose not to charge manufacturers the “slotting fees” they normally had to pay to get a product on a store’s shelves. For brand name suppliers such as Pillsbury, General Mills and Proctor & Gamble, the online grocery market was uncharted territory, and one where they did not have to compete with generics; to them, Webvan offered to the opportunity to deliver product samples directly to consumers.94 In addition, Webvan could also provide manufacturers information such as what individual consumers bought and whether a particular product made it on a customer’s “re-purchase” list.95 Marketing To entice prospective customers in the San Francisco Bay Area to try its service upon launch in June 1999, the company waived delivery fees regardless of order size through July 31, 1999.96 In addition, the Webvan Group advertised that its prices were 5% lower than conventional grocery stores.97 Following the launch of its service in June 1999, Webvan planned to increase spending substantially on its radio and newspaper advertising campaigns to build brand identity and loyalty.98 It also planned to leverage its relationship with media investors such as CBS and Knight-Ridder. And with the data it gathered on consumer purchasing, preferences and behavior, it planned to offer targeted and customized services. In 1999, the company announced that it planned to spend $6 to $9 million annually in advertising for each city and a total of $200 million on advertising in the year 2000.99 Growth and Efficiency Projections Webvan planned a national rollout to achieve economies of scale since all of the 26 DCs the company planned would conform to a single design, and be managed with a uniform technology
Webvan 602-032 platform. In July 1999, the company entered an agreement with the Bechtel Group, the engineering and construction firm, to construct these DCs across the country at a total cost of $1 billion. With each DC costing around $35 million the savings were projected to be tremendous since it would cost $90 million to build the 18 supermarkets that would be the equivalent of each DC. In its prospectus the 2000 and that an additional 7 centers were slated to open in 2001 t0 icago, and Seattle by the end of ompany stated that it Planned to begin distribution in Atlanta, The company had to contend with two big costs that a traditional grocer did not pay for-packing carts with customers' orders and delivering the ordered items. At the DC, the average 25-item order ould be assembled in less than one hour The central Dc proposed to service the equivalent of consumers from 18 conventiona supermarkets but with less than half the labor and double the selection of items. While a conventional supermarket was constrained by the number of items it could offer customers, Webvan's DCs were designed to stock 50,000 items. And each DC, the company claimed would require only 900 employees as opposed to the 2, 700 that would be required in 18 conventional supermarkets Ultimately, the highly automated warehouses that allowed a picker at Webvan's DCs to be nearly 10 times more efficient than a normal shopper wheeling a cart down a grocers aisle were expected to give the company a 10 percentage point edge on profit margins over traditional supermarkets With a DC operating at its designed capacity and as a consequence of the additional demand based information in the supply chain, the company also projected that inventory would turn 24 times annually as compared to the 9 to 11 times seen in traditional supermarkets Revenue At full capacity with 8,000 orders a day and an average order of $103, the company projected annual revenues of $300 million per DC. At half this volume, the company expected to break-even. An average supermarket, on the other hand, typically brought in $12 million a year And by situating its DCs in low-rent industrial zones, the company hoped to set real 1% of revenue(generated when operating at full capacity)as opposed to the 6% seen for conventional evidential The company also projected achieving operating profits of 12% of sales, triple the grocery industry average when its DCs would be operating at full capacity and after absorbing the costs of entering new markets and enlisting new customers. The resulting savings, it claimed would then allow it to underwrite the cost of deliveries to customers' homes and customers could expect free delivery on any order over $50. 10 Borders stated that if everything went according to plan, the Oakland DC would be profitable within six to 12 months and "other Webvan warehouses might break even in as little as 60 days The companys prospectus filed with the SEC in November 1999 was more conservative: it stated that a DC viewed as stand-alone business unit was likely to generate significant cash flow following five quarters of operation. Analysts expected the company to have $11.9 million in sales in 1999-an amount that was less than Safeway or Kroger made in a single busy afternoon but that would soon climb to $120 million in revenue in 2000 and over S500 million in 2001 1200 families, in the san Francisco bay area used its service consistently, t could achieve positive earnings(before interest, taxes, depreciation and amortization) for the Oakland DC, if the DC were viewed as a stand-alone business unit. In general, the company felt that it needed to achieve a 1%
Webvan 602-037 9 platform. In July 1999, the company entered an agreement with the Bechtel Group, the engineering and construction firm, to construct these DCs across the country at a total cost of $1 billion. With each DC costing around $35 million the savings were projected to be tremendous since it would cost $90 million to build the 18 supermarkets that would be the equivalent of each DC. In its prospectus the company stated that it planned to begin distribution in Atlanta, Chicago, and Seattle by the end of 2000 and that an additional 7 centers were slated to open in 2001.100 The company had to contend with two big costs that a traditional grocer did not pay for—packing carts with customers’ orders and delivering the ordered items.101 At the DC, the average 25-item order could be assembled in less than one hour. The central DC proposed to service the equivalent of consumers from 18 conventional supermarkets but with less than half the labor and double the selection of items. While a conventional supermarket was constrained by the number of items it could offer customers, Webvan’s DCs were designed to stock 50,000 items.102 And each DC, the company claimed would require only 900 employees as opposed to the 2,700 that would be required in 18 conventional supermarkets.103 Ultimately, the highly automated warehouses that allowed a picker at Webvan’s DCs to be nearly 10 times more efficient than a normal shopper wheeling a cart down a grocer’s aisle were expected to give the company a 10 percentage point edge on profit margins over traditional supermarkets.104 With a DC operating at its designed capacity and as a consequence of the additional demandbased information in the supply chain, the company also projected that inventory would turn 24 times annually as compared to the 9 to 11 times seen in traditional supermarkets.105 Revenue At full capacity with 8,000 orders a day and an average order of $103, the company projected annual revenues of $300 million per DC.106 At half this volume, the company expected to break-even.107 An average supermarket, on the other hand, typically brought in $12 million a year. And by situating its DCs in low-rent industrial zones, the company hoped to set real estate costs at 1% of revenue (generated when operating at full capacity) as opposed to the 6% seen for conventional supermarkets which had to operate near residential areas.108 The company also projected achieving operating profits of 12% of sales, triple the grocery industry average when its DCs would be operating at full capacity and after absorbing the costs of entering new markets and enlisting new customers.109 The resulting savings, it claimed would then allow it to underwrite the cost of deliveries to customers’ homes and customers could expect free delivery on any order over $50. 110 Borders stated that if everything went according to plan, the Oakland DC would be profitable within six to 12 months and “other Webvan warehouses might break even in as little as 60 days.”111 The company’s prospectus filed with the SEC in November 1999 was more conservative: it stated that a DC viewed as stand-alone business unit was likely to generate significant cash flow following five quarters of operation. Analysts expected the company to have $11.9 million in sales in 1999—an amount that was less than Safeway or Kroger made in a single busy afternoon but that would soon climb to $120 million in revenue in 2000 and over $500 million in 2001.112 Market penetration The company had stated that if approximately 1% of households, about 120,000 families, in the San Francisco Bay Area used its service consistently, it could achieve positive earnings (before interest, taxes, depreciation and amortization) for the Oakland DC, if the DC were viewed as a stand-alone business unit.113 In general, the company felt that it needed to achieve a 1%
602-037 Webvan to 3% penetration to achieve positive earnings on a similar basis in other markets. However,even then in 1999, analysts sounded alarms that Webvan might expand faster than consumer demand Skeptics such as Dan Rabinowitz of Peapod, a competitor online grocer, noted that the company would have to do eight times the volume of a typical store to be successful and said, " Were in 10 markets, and we,ve seen how difficult it is to build up scale. Others thought that the company was in over its head"stating that the problem with the companys delivery model was largely one of density and that"It]ransportation should be in the hands of transportation professionals who know how to do it However, speaking of the Webvan system, Shaheen said, This is not your mothers grocery store t took a total reengineering of the supply chain and a total e-engineering of the business model to put this together. And Kevin Czinger, the companys chief financial officer insisted that the Webvan Group would be highly cash generative"and break even within five quarters after launchin 8 Changes and Performance over Time Since its launch in the San Francisco Bay Area in June 1999, sales at the company improved steadily as did its revenues(see Exhibit 3 for the companys quarterly income statements). However, losses continued to mount at a high rate given the companys capital-intensive business plan. In keeping with its expansion plans, by the third quarter of 1999, the company had signed leases in locations that would allow it to serve the Washington D. C, Dallas, Chicago and Seattle markets And by the end of 1999, it had signed an additional set of leases to serve Denver, northern New Jersey, Philadelphia, New York City, Boston, Orange Country and Baltimore Early Performance in the San francisco Area The Webvan Group chose the San Francisco area for its debut, citing residents' food and Web savvy. The highly automated system that the company built was, however, vulnerable to bottlenecks misaligned totes, misread displays and Shaheen acknowledged in January 2000 that incomplete orders were a challenge to overcome. Even at this early point in its operations, getting a preferred delivery time was a problem for customers. To address this problem, Webvan intended to guarantee repeat customers the same delivery slot week after week. One quarter after opening, at the end of September 1999, the Webvan Group was operating its 330,000 square foot flagship San Francisco Bay Area DC located in Oakland, CA, at less than 20% The DC was operating for fiv he company employed 630 people. Its offerings had increased since its launch in June 1999 from 15,000 items to 18,000 items at the end of September 1999. Repeat customers ordered every 10 days on average(from launch through the third quarter of 1999)and each customer ordered, on average, 20 items per order. And consistent with the companys claim that its produce would be of higher quality owing to lesser handling, the company reported in November 1999 that 17% of its revenue could be attributed to produce sales compared to the 10% seen by conventional supermarkets By the end of 1999, the company had extended its distribution operations at its Oakland dC to six days a week but it was still operating at less than 25% capacity and stated that it now dealt with 130 distributors and about 230 vendors. By January 2000, it was operating 7 days a week and
602-037 Webvan 10 to 3% penetration to achieve positive earnings on a similar basis in other markets.114 However, even then in 1999, analysts sounded alarms that Webvan might expand faster than consumer demand.115 Skeptics such as Dan Rabinowitz of Peapod, a competitor online grocer, noted that the company would have to do eight times the volume of a typical store to be successful and said, “We’re in 10 markets, and we’ve seen how difficult it is to build up scale.”116 Others thought that the company was “in over its head” stating that the problem with the company’s delivery model was largely one of density and that “[t]ransportation should be in the hands of transportation professionals who know how to do it.”117 However, speaking of the Webvan system, Shaheen said, “This is not your mother’s grocery store. It took a total reengineering of the supply chain and a total e-engineering of the business model to put this together.”118 And Kevin Czinger, the company’s chief financial officer insisted that the Webvan Group would be “highly cash generative” and break even within five quarters after launching.119 Changes and Performance over Time Since its launch in the San Francisco Bay Area in June 1999, sales at the company improved steadily as did its revenues (see Exhibit 3 for the company’s quarterly income statements). However, losses continued to mount at a high rate given the company’s capital-intensive business plan. In keeping with its expansion plans, by the third quarter of 1999, the company had signed leases in locations that would allow it to serve the Washington D.C., Dallas, Chicago and Seattle markets.120 And by the end of 1999, it had signed an additional set of leases to serve Denver, northern New Jersey, Philadelphia, New York City, Boston, Orange Country and Baltimore.121 Early Performance in the San Francisco Area The Webvan Group chose the San Francisco area for its debut, citing residents’ food and Web savvy. The highly automated system that the company built was, however, vulnerable to bottlenecks, misaligned totes, misread displays and Shaheen acknowledged in January 2000 that incomplete orders were a challenge to overcome.122 Even at this early point in its operations, getting a preferred delivery time was a problem for customers. To address this problem, Webvan intended to guarantee repeat customers the same delivery slot week after week.123 One quarter after opening, at the end of September 1999, the Webvan Group was operating its 330,000 square foot flagship San Francisco Bay Area DC located in Oakland, CA, at less than 20% capacity. The DC was operating for five days a week and the company employed 630 people. Its offerings had increased since its launch in June 1999 from 15,000 items to 18,000 items at the end of September 1999. Repeat customers ordered every 10 days on average (from launch through the third quarter of 1999) and each customer ordered, on average, 20 items per order.124 And consistent with the company’s claim that its produce would be of higher quality owing to lesser handling, the company reported in November 1999 that 17% of its revenue could be attributed to produce sales compared to the 10% seen by conventional supermarkets.125 By the end of 1999, the company had extended its distribution operations at its Oakland DC to six days a week but it was still operating at less than 25% capacity and stated that it now dealt with 130 distributors and about 230 vendors. By January 2000, it was operating 7 days a week and