May 8, 2003

by Steve Baldwin

On Top of The World

It was November, 1999: the absolute zenith of Europe's dotcom frenzy, and the 560-seat ampitheatre on London's Regent Street was packed with the masters of Europe's fashion universe. Ralph Lauren was there - and so was Alessandro Benneton and a couple of Versaces.

Bernard Arnault was there too - the urbane French billionaire whose $500 million investment fund had recently placed tremendous bets on hot Net startups such as, Datek, eBay, and @Home. Arnault had paid for this opulent party - the dancers, the free-flowing Moet champagne, and, most importantly, the Web site being launched tonight, which was called

A cluster of recessed ceiling lights dimmed as Hip Hop music swelled from a 15,000 watt sound system. Above the stage, a 22-foot projection screen displayed a bright orange rectangle: the splash screen of a Web site. Into the orange rectangle stepped a 16-foot tall animated female figure. When the figure spoke, it was in a soft, sexy, BBC-inflected voice:

"My name is Ms. Boo", she said. "Let's go shopping the Boo way". The avatar winked at the audience, flipped her virtual ponytail, and stepped into one of the Web site's "virtual boutiques".

She bought a set of Vans sneakers - the kind favored by skateboarders, and dropped them into her "Boo Bag". She entered another boutique, picked up a North Face jacket and a snazzy T-Shirt from Cosmo Girl. She made wry and sassy comments about the goods, and after spending about $600 danced over to the site's checkout counter.

The audience was rapt. Ms Boo moved like Laura Croft - the 3D wireframe heroine of Tomb Raider, a popular shoot-em-up adventure game. But instead of packing a 45-automatic, Ms. Boo toted a virtual shopping bag adorned with the logos of Europe's best brands, including Puma, Jil Sander, New Balance, FUBU, and Patagonia.

Ms. Boo was amazing, and Boo's site was astounding: a fully-immersive Web shopping experience that offered 40,000 different sportswear products to buyers in 6 different European countries. It was hip, multilingual sassy, and smart: a key selling point for the young, ultra-hip, sportswear buyers to whom it catered. Shopping at Boo was more like playing a 3D videogame than browsing through a dull old Web site, and by the end of the presentation, the audience was on its feet.

Boo was the biggest, most magnificent, most opulent e-commerce venture in history, and its launch party was the toniest event to hit London since Princess Di's funeral. Backed by some of the richest people in Europe, its destiny as "the of Europe" seemed assured.

Within 6 months, however, after consuming almost $130 million in cash, Boo would go down in flames.


It is unlikely that Malmstem, Leander, or Hedelin - the three 20-somethings who ran Boo into the ground, ever expected their names to be forever associated with the world's most spectacular Internet Failure. Right up until the moment of Boo's fiery demise, all three seem to have lived charmed lives.

Malmstem and Leander grew up together in the small Swedish town of Salur. In their early 20's they combined resources and started a publishing company called LeanderMalmstem that published fringe fiction from "up-and-coming" Swedish authors.

In 1994, LeanderMalmstem got lucky. It published a Swedish translation of Tin Tin in the New World that became a runaway bestseller.

In 1996, Malmstem and Leander took some of their profits from Tin Tin and started a Web-based book-selling business called The site became an immediate hit in Sweden, and stocked 1.5 million books by the end of 1997. Elle Magazine soon dubbed the pair "the Literary Rock Stars of Europe", and Wired Magazine characterized as " - an ecommerce site brimmed with uber-hipness".

Bertelsmann - the German giant - was so impressed by Bokus that it bought the site in May of 1998 for a whopping $30 million. The deal, brokered by Patrick Hedelin, an old friend of Leander and Malmstem, made them all multi-millionaires.

It was quite a coup - three young hipsters who had somehow managed to turn their literary hobby into a high-tech Interrnet business with global aspirations. And the European media ate it up: one Swedish newspaper went so far as to write: "Malmsten and Leander are now on the podium of the national idols, between ABBA and Bjorn Borg".

In late 1998, Leander and Malmstem were introduced to Bernard Arnault, the billionaire head of LVMH Moet Hennessey - a 33,000 employee luxury goods giant with annual revenues of $6 billion. Arnault - who surfed the Web recreationally, had become convinced that e-tailing was the wave of the future, and had put together a $500 million venture fund called Europe@Web. He took one look at the trio's track record and, convinced that he had found his dream team, expressed a strong interest in backing whatever fabulous Net venture the trio wanted to launch next.


Just before Christmas, 1998, Malmsten, Leander and Hedelin locked themselves into a Stockholm hotel suite to conceptualize on the perfect e-commerce project to offer Arnault. All were aware that the e-clothing market was huge: analysts projected $60 billion in worldwide spending by 2003 with "first-mover advantage" accuring to any outfit that, like Amazon in the US, had the resources to forcefully project its global brand into the public mind (even if doing so put a serious dent in its bottom line).

"Boo" - as the project would later be named -- would be Bokus done right - a multilingual, multicultural site leveraging the global reach of the Internet to sell Europe's best, most exclusive brands to an international clientele. Unlike Bokus, which didn't ship goods outside Sweden, Boo would sell sportswear not just in Europe, but in the U.S.A., making it a truly global enterprise.

Leander cobbled together a mission statement for the project, which read as follows:

  • "THE VISION: To become the worlds's largest online fashion and sportswear retailer"
  • THE VEHICLE: A cutting edge Web site staffed by a team of dynamic individuals, hand-picked from the worlds of fashion, new media, marketing, business, technology and finance.
  • THE OUTCOME: An awesome virtual shopping experience that surpasses anything that exists on the Web"

The trio realized Arnault hadn't just handed them a global e-commerce mission, but had also provided them a golden opportunity to break through all aesthetic barriers that had retarded the Web's evolution beyond what Malmsten called "a boring mass of static HTML and pixelated JPEGs"

The trio's pronounced distaste for the way goods were sold on the Web wasn't just an aesthetic consideration. Leander, who had traveled in fashion circles for years, had repeatedly heard luxury clothiers complain that the Web was "a trashy medium". Most believed that Web exposure would only "cheapen" their brands. Ralph Lauren had gone so far as to announce to the press that he would rather "go to his grave" than suffer his goods being sold by some crappy Web site.

At the same time, European luxury retailers were becoming increasingly paranoid about the global encroachments of US-based Web superstores, especially Amazon, which were already adding designer name luxury goods to their catalogs. If Euro-retailers simply stood on the sidelines, the global e-commerce game would be lost to the rumpled Americans.

What was needed - and what the trio could provide - was a Web site completely unlike any that had ever been built before - a site glitzy enough to please the whims of Europe's high-value retailers. In order to convince the commanders of Europe's tony brands that the Web "wasn't crappy", the site had to provide a shopping experience as classy as the opulent physical stores that retailers on Carnaby Street and Fifth Avenue had traditionally used to display their goods.

After two days of intensive concentration, the trio developed the skeleton of a functional specification. On a yellow pad, Malmsten wrote down the features to be incorporated into the site, which Leander -- in a flight of inspiration - had named "Boo" (like Bokus, Boo was a meaningless term that she thought was "hip"). These features included:

  • 1) A 3D, wisecracking avatar programmed to behave exactly like a Liberal-arts educated shop clerk.
  • 2) Rotatable, zoomable 3D "virtual mannequins" that let Web users examine clothing from every possible angle.
  • 3) A frequently-updated Web-based magazine that would add hipness and "stickiness" to the Web site.

They got a demo together and gave it to Arnault - a friend of Kajsa's was a good Flash designer and she mocked up a bunch of cool-looking pages on a CD-ROM. In January, the trio flew back to France to show the demo to Arnault. He flipped, and immediately pledged $30 million to build Boo.

Within a couple of weeks, other investors pledged funds to the nascent project, convinced by Arnault's bold commitment that it was a safe bet. 21 Investment, an arm of the Benetton family holding group, was the first to get on board, and Omnia, a fund backed by Lebanon's wealthy Hariri family, was second. Sedco, a fund backed by Saudi Arabia's Mahfouth banking family came next - bringing up the rear was J.P. Morgan, a New York investment bank that had traditionally avoided investing in tech companies. Finally a commitment from Goldman Sachs raised Boo's war chest to $108 million.

It was the largest amount of money ever raised by a Net startup in its first-round of financing.


By March of 1999, the investors' checks had cleared Boo's corporate bank account, and Boo's founders began to put some of the investors' money to work. The company moved into a 12,600 square foot office on London's Carnaby Street - the birthplace of the "Mod" style movement of the 1960's. Here, a workforce of some 200 people was assembled - HTML coders, Web designers, Java programmers, and other functionaries working to transform the demo that Arnault had seen on the CD-ROM into a fully-functional e-commerce site.

Boo opened branch offices in New York, Paris, Stockholm and Munich, to give BOOM - its inhouse style magazine -- bureaus in each important "capital of street culture". They hired fashion writers, reporters, and photographers to staff each office. The New York office was by far the biggest: 90 people, including a team of developers whose job it was to build Boo's back-end commerce system. To house them, Boo leased 8,000 feet of prime real estate in the West Village with a panoramic view of the Hudson River.

With Boo's staff and infrastructure in place, its founders took to the air, and throughout April and May, spent hundreds of hours jetting between London, Milan, Paris, Cologne, and Amsterdam. Their job was to sign up as many of the most popular sportswear brands in Europe as possible. Armed with an even flashier demo than the one thay had showed Arnault, by the beginning of June they had signed up DKNY Active, The North Face, Puma, Everlast, Jil Sander, New Balance, FUBU, Converse, Helly Hansen, and Patagonia.

The international business press went through the roof. Forbes Magazine added Boo to its list of "Top Cool Companies of 1999". Fortune Magazine went so far as to put the stylish trio on the cover. The Industry Standard, the New York Post, Business Week, Newsweek and Time all fell over themselves to cover Europe's brashest, biggest Web startup - a grand convergence of fashion, hipness, and Net-savvy entrepreneurship. One analyst even declared that Boo was already worth between $200 and $300 million dollars, even though Boo's Web site was just a placeholder that wasn't capable of selling a single shoelace.

The press was breathing hot and heavy, but Leander and Malmstem hungered for more exposure - big time exposure -- and to achieve this end they began spending money at a big time rate. After huddling with PR and advertising consultants, an impressive PR blitz was set into motion that included billboards, TV spots, movie shorts, and "street" sticker campaings. Magazine ads were booked in Details, GQ, ESPN Magazine, Vogue, and Elle. Boo's TV spots were directed by Michael Coppola - son of Francis Ford. By August, Boo was spending more than $5 million a month on "communications" expenses. These expenses would total more than $25 million by the end of the year.

As befit their influential status in an industry where image is more important than substance, Boo's founders insisted on doing everything "first class". When Malmstem and Leander crossed the Atlantic to woo a US apparel manufacturer, they travelled on the Concorde and stayed in first-class hotels. On the ground, they rented Ferraris and Dodge Vipers. They boasted of taking part in all-night drinking sessions with clients that featured $150 shots of rare Scottish malts. When in London, they breakfasted daily at a top Soho hotel, running up huge tabs.

Befitting Boo's grand and global mission, its Carnaby Street were signficantly more opulent than those of a typical Internet startup: Malmstem and Leander went so far as to decorate every toilet in the building with Boo's signature orange logo. They gave each of Boo's senior managers a complimentary mobile phone and an $800 Palm V handheld computer. Each day, cartloads of fresh organic fruit, imported chocolate and soft drinks were trucked into the Carnaby Street offices. Workers seldom worked past 6:00 PM - an unusual thing for a dotcom crunching on a hard and fast deadline (Boo was slated to launch in July). Vast, throbbing dance parties for Boo's staff were also held on a regular basis at trendy Notting Hill music clubs. These eleborate shindigs featured hired dancers, live music, open bars, and thousands of Euros worth of free apparel give-aways.


3,000 miles across the Atlantic, in Boo's New York office, a much less cheery situation presented itself. By the late Spring of 1999, Boo's backend developers were already struggling with serious delays trying to build Boo's commerce software.

Boo's software had always been an issue with the New York office. None of Boo's American geeks could fathom why Leander and Malmstem had chosen Interworld's Commerce Xchange system - a notoriously buggy, hard-to-customize commerce system, to serve as the core of Boo's back-end.

Dealing with London was difficult - Leander and Malmstem didn't communicate well with techies, and often walked away out of the room when they got bored. But by far the most pressing problem haunting the New York developers was the extraordinarily complicated nature of Boo's back-end software.

It was multi-lingual, handled multiple currencies, and did on-the-fly tax calculations in 6 different tax zones. It also had to integrate with the back-end systems of Boo's shipping partners: UPS and DeutschesPost, and communicate with the warehouse inventory systems of each of Boo's clothing suppliers.

Nobody, not even Microsoft, had ever tried to build such an elaborate commerce system with so many disparate pieces assembled from a raft of different suppliers. When things went right, the victory was short-lived, and bugs soon reappeared. When things went wrong, the whole Rube Goldberg-like scheme crashed. As the summer wore on, Boo's original launch deadlines began to slide, and the New York developers began to seriously wonder whether Boo could possibly launch in time for the all-important Christmas selling season.

Some of the New York developers also were concerned about Boo's Web site, which had existed as a placeholder since May. By mid-summer, the site had already developed a profoundly negative reputation on mailing lists and Usenet groups for being slow. Even without any merchandise on display, the site's elaborate Flash and Quicktime VR animations took many minutes to load. Sometimes they locked up the user's browser completely, forcing a system reboot. London, however, didn't want to listen to any complaints made about the site's elegant front-end by the "back-end" boys in New York. As far as Malmsten and Leander were concerned, the front-end was theirs to control: a "a design issue", not a technology issue.

Communications between London and New York were also maddeningly erratic. One project manager practically had a coronary shortly after he hired on in August and discovered that Boo didn't even have a project plan in place for Boo's backend software deployment. Only by hastily improvising a plan using a borrowed copy of Microsoft Project was absolute chaos averted. In another instance of communications breakdown, a lead developer waited for 36 hours to receive an answer on a mission-critical issue, because Leander and Malmstem had taken the day off to have their portraits done for Britain's National Portrait Museum. Other developers fumed when Boo's senior management frittered away precious time in meetings "whose only purpose was to decide the subject of the next meeting".


It wasn't until mid-September that Malmstem and Leander woke up to the fact that Boo's November 1 launch was in serious jeopardy. No matter how many 24-hour shifts were worked by New York's bleary-eyed crew of Solaris and Oracle specialists, the software just wouldn't behave.

In a tense early morning conference call in late August, Boo's founders put the New York office on notice that they would be held responsible for meeting, or failing to meet, the November deadline. When London asked for a recommendation on how to speed up the work, the developers suggested that the original 6-country launch be scaled back to just one country - preferably England, to meet the deadline. London wouldn't have it: Boo was going to launch in 6 countries, or it wouldn't launch at all.

The front end of the site - the part with Ms. Boo and all the other glitzy fashion frills - was already complete, but without an operational back-end was there, the whole thing was a joke - a gorgeously-bedecked body without a head.

Throughout October, developers in Boo's New York office fought a desperate race against time, and finally won. In the early hours of October 31st, the pieces all fit together. Boo's elaborate commerce back-end was joined to its feature-rich front-end, and the site was launched. Everybody at Boo breathed a deep sigh of relief, believing the worst to be over.


But within less than a week of Boo's November 1st launch, a sinister plume of negative publicity began to swirl across the Internet. Something - and nobody seemed to know exactly what it was - was crashing the Web browsers of thousands of people who came to the Boo and tried to shop.

Boo's developers frantically searched for the cause, and pored through the site's Flash, Java, Quicktime VR, CGI, and back-end code, but it was too late - hundreds of thousands of people who had come to see the site were unable to view it. Boo's cutting-edge 3D Virtual Boutique technology was bombing for all but a handful of high-bandwidth users, and thousands of angry customer-emails started pouring in to Boo's crew of customer support people. All told, more than 98% of Boo's prospective customers gave Ms. Boo a Bronx cheer, and went elsewhere.

The media - once solicitous - turned malicious. They ran a skein of unflattering articles whose headlines read "Ms. Boo Gets Booed" and "Long-Awaited Fashion Site Craps Out". One journalist tested Boo out on 6 different machines, and gleefully noted that the site crashed each one. Others noted that some of Boo's customers who were actually able to order items frequently received the wrong items, and found them nearly impossible to return.

As the firestorm of media criticism burned through Boo's public image, Malmstem and Leander were forced to defend Boo's performance in public. Malmstem's response was feeble and unresponsive: "We're a broadband proposition . We're not regressing the site - we built a site that's optimum and perfect for new technologies. There are no current plans to make it more accessible." Leander defended frustrating all but 2% of Boo's potential audience by saying, "our mission is to sell to the young and the hip, and they all have T1 connections. We don't care if the other people go away - we encourage them to shop elsewhere."


Shortly after Christmas 1999, Patrick Hedelin decided it was time to jump ship. Stating that he needed to "spend more time with my family", Hedelin left London and returned to Sweden. The loss of Hedelin - the level-headed money man -- removed the last vestige of financial credibility once enjoyed by the Boo team. Alessandro Benneton soon angrily withdrew his support from the project after Arnault refused to insist that Boo's founders "take a back seat role".

At the end of January, Boo's quarterly sales figures came in, providing extremely troubling news. In the three months of the sites operation during the all-important Christmas buying seasion, Boo had only been able to move a paltry $680,000 worth of merchandise. Malmstem and Leander knew that the remaining investors would soon demand a serious restructuring, or would move to have them thrown overboard.

On February 11, Malmsten bit the bullet and laid off 100 Boo staffers - plus 20 telephone support people deemed surplus (it turned out that boo's customers preferred to use e-mail instead of the telephone). He also fired the designer who had been responsible for Boo's first unsuccessful Web incarnation (although all she did was implement the founders' design features to the letter). The firings were carried out brutally - in a way that one staffer called "a thoroughly unprofessional manner" that left a few fuloughed staffers crying at their desks.

On March 5th, the agony of Boo's users was eased a bit when a "low-bandwidth" version was belatedly added to the site. Ms. Boo was reduced to a background character no longer capable of making wry comments about anything. Malmstem also abruptly changed Boo's "no-discount" policy by instituting a 40% discount across Boo's line of merchandise. Some of Boo's users came back, but Boo's revenue - $1.1 million in March -- remained a trickle - not the flood that Boo's increasingly worried investors were demanding. JP Morgan was the next to leave the Boo board - citing a "conflict of interest" due to its 4% share of Boo.

Boo was completely aflame by mid-April, and few expected that the project had much time left. A second restructuring plan that proposed closing Boo's international offices as a cost-cutting measure was drafted. But when Malmsten and Leander went back to Arnault in early May to get an additional $30 million to keep the site going, he turned them down flat. Arnault had plainly had enough - the recent NASDAQ meltdown had reduced the value of high-tech holdings by more than 30%, and he was plainly not in much of a mood to spend another $30 million on a venture whose only talent seemed to be burning cash.

For two weeks in May, Boo's founders assured the press that "nothing is wrong with Boo" as they tried to line up a buyer for the site and its vaunted international commerce technology. They approached Nike and Puma, but the talks fell through. Soon, nothing could prevent the behemoth from hitting the ground.

On May 16th, Boo's Carnaby Street staff was called into a conference room and given the bad news: all were going to be losing their jobs, except for Leander, Malmstem, and three others, who would be kept on staff to continue demo-ing Boo just in case additional investors appeared. When these investors didn't show up, KPMG was brought in to liquidate Boo's assets, which a KMPG official acknowedged were mainly "a brand name and a Web site".

KMPG found less than $800,000 remaining in Boo's coffers, and immediately set about trying to raise enough cash to pay down Boo's $25 million debt to DeutschePost and two advertising agencies. The liquidator put Boo's vaunted "world-class" commerce software on the block. More than 30 companies wanted to bid, but after KMPG raised the minimum bid to $1.5 million, most of them went away. On May 28th, Boo's back-end software was auctioned off to BrightWorks - a UK Internet company - for what the acquiring company admitted was "a song": $374,900.

A song it was - what BrightWorks paid was less than 1% of what Boo had spent to develop it.


You're on the web a lot. You've seen many a dead site. You've forgotten our email address... and you don't feel like coming back here to get it.

What do you do?

(javascript required)

The Ghost-o-Meter opens a small, movable window... if you've found a Ghost Site, fill in the blanks, fire it off, and go back to foolin' around. Its that easy.

You can also use this form:

What the ??!

Well, this is all very interesting, but what the heck is Ghost Sites anyway? Why devote a live site to Dead Sites?

If you're interested in this Ghost Sites thing, it is a project that I began in the summer of 1996 while I was working for Time-Warner's Pathfinder. Late in the evening of July 4th, while piloting a small craft across Long Island Sound, I had what only can be described as an epiphany.

From out of the depths came a cruel vision of the World Wide Web. It wasn't a friendly place - an innocent place of community, commerce and chat. It was a great and utterly pitiless electronic ocean that swallowed up sites, careers, and venture capital like a ravenous killer whale. Great sites - sites like Mecklerweb and iGuide - were going down with all hands. Great fortunes were collapsing and proud content sites lay wrecked on the bottom. No one seemed to care. The future was a vast abyss - who would record these days of New Media folly, disaster and despair?

Back on shore, but still haunted by this vision, I launched Ghost Sites as a modest attempt to document the great disappearing fleet of web sites sinking beneath the waves. This project briefly made me spectacularly famous, and then I was quickly, and completely forgotten.

By March of 1997, Ghost Sites had succumbed to the same deadly entropy that had settled over the Internet, and became a crewless wreck itself. For six cruel months, it drifted like a despised garbage barge, broke its keel in a summer squall, and finally washed up on Geocities.

On an icy November morning, Morbus boarded the wreck, inspected the damage, and offered the captain a safe harbor. The bilge pump was started, and the squealing, rusty hull lifted off the sands again. It soon arrived here - in the dark, unquiet waters of Disobey.Com.

If you have a favorite rotting site that you'd like to mention, email me at

Ghost Sites has appeared in a number of places including Time Magazine, ZDNet, The Netly News and more. For a list of all those we know of, as well as links to online counterparts, click here. You can also take a look at the limited edition t-shirt we once offered.

Previous Issues

Text issues? Click here.
Web issues? Click here.