TECHNOLOGY TRENDS

Webvan's plan involves big money - its $400 million in private financing is the most ever for an Internet company- big risks and difficult technology. To win, itmust become the first retailer to regularly deliver goods as various as ice crea, paint and TV sets to consumers' homes on a massive, national level.

If successful, it would be as revolutionary as when Sears Roebuck began using the mail to deliver goods to consumers at the turn of last centruy, overturning a retailing model that relied on the local general store.

"This could be the biggest company ever to come out of Silicon Valley." says David Beirne, the venture capitalist who originally financed the Webvan concept, which was dreamed up by Louis Borders of Borders Books fame.

The plan took a little detour Monday. Webvan announced it will buy HomeGrocer, an Internet grocery funded by former Netscape CEO James Barksdale, for $1.2 billion in stock. The deal gives Webvan more customers and revenue sooner and more chaply than originally planned. That's important now that diving stock prices are restricitng ways to raise money and investors are demanding that Net companies be financially healthy.

"It looks like a new pragmatism on Webvan's part," says Randall Stross, author of eBoys, which in part details the creation of Webvan, Webvan will be able to conserve the cash it now has and hold out longer, waiting for the business to catch on and hit critical mass. That's a good decision for Webvan, Stross says. "They're going to need a lot of runway."

Which brings up the hitch. On one hand, Webvan has gotten off to a solid start. The first local market - the San Francisco Bay Area - is up and running and proves that the concept and technology can work. The Atlanta market just opened. Seattle is being built.

But in each market, Webvan spends $35 million just to build the distribution center - the highly automated 300,000-square-foot warehouses that can move more goods than 18 grocery stores. Then comes marketing, staffing and other costs. Webvan has to spend a ton of money up front, and then hope of money up front, and then hope that consumers are ready - ASAPI - for internet ordering and home delivery.

There, the jury is out, in the Bay Area, the number of customers and size of their orders is ratcheting up, but the service hasn't caught fire. The average order, about $72 in September, is up to $90 now and has to get to $103 to break even. Customers who try it love it. Yet, in general, "Thre is no buzz about it." says Trip Hawkins, CEO of 3DO, a computer game company in Silicon Valley. "None of my friends or employees has mentioned using it. I've not used it. That suggests that consumer habits, as usual, are changing very slowly."

Groceries to start The first thing Webvan has to get past is it image as a Web grocer, lumped in with struggling companies such as Peapod and NetGrocer. The point of Webvan isn't food. It isn't even, necessarily, the Web. The point is automation and home delivery on a scale never before attempted. The point is creating a new kind of service - like an old-fashioned milkman, but one who will hand-deliver anything that can fit inside a Saweway or a Wal-Mart.

"You won't hear any of us talking about the Intenet grocery business," says CEO George Shaheen, lured to Webvan in 1999 from his job as CEO of Andersen Consulting. "It's not our endgam."

Instead groceries are a good place to start for two reasons.

First, people order a lot more groceries than books or Cds, so selling groceries allows Webvan to build its system on a massive scale right away, then add more categories later. Earlier this month, Webvan announced it will carry consumer electronics, delivering DVD players along with the milk and bananas.

Second, people buy groceries often. That gives Webvan what Shaheen calls "frequency of contact"- a way to build a relationship with consumes. "Our people come into your house." Shaheen says, "I believe that two or three companies are going to win the right to go into your house. We'll be one of them."

To communicate the full impact of Webvan, Shaheen likes to show off the Oakland distribution center. It is a wonder of automation. Plastic totes filled with orders zoom overhead on conveyors, guided by computers loaded with artificial intelligence software written and maintained by 16 Ph.Ds. In a refrigerated room, five carousels, each the height of two men and the length of a football field, carry items to a single packer, who places them in bins as they pass by.

Once a customer uses the Web site to place an order, Webvan's computers pray the order to all the appropriate parts of the distribution center, whether to dry goods or forzen foods or the deli or the fresh fish area. The totes travel around collecting the orde, then wind up on the loading dock. Dan Vucovich, who runs the Oakland center, says orders are 99.9% accurate.

Webvan has effectively used the automation to overcome some of the difficulties of grocery delivery. Produce looks as good as anything on a grocery's shelf. Ice cream arrives hard enough to bend a spoon.

From the loading dock, large trucks carry orders to nine local holding facilities. There, the orders are put on smaller trucks that go into neighborhoods. The drivers- couriers, as Webvan calls them- will put up and carry the order intoyour kitchen. Delivery comes the day after you order and its promised within a 30-minute window that you determine. Vucovich says 8% of orders are on time, and customers don't dispute that.

The hub-and-spoke system allows the big trucks to move to the local centers at night, avoiding traffic. It also extends the reach of distribution centers. Webvan now serves Sacramento from the Oakland center.

Put it together, and the Webvan system could rewrite the rules of retailing, at least in groceries. The company figurs it can make profit margins of 12% per order, vs. the industry average of 8%. In each market, Webvan can build the equivalent of 18 grocery stores for $35 million. At $5 million to $6 million each, those 18 stores would cost a mainstream company more than $90 million.

And the system seems to work, which was not a certainty when Borders first talked with Beirne three years ago. The goal now is to build on that system to creae a sustainable business - and eventually, perhaps, one of the world's biggest businesses.

Lots of ambition

Webvan plans to build distribution centers near ever major city. It's an expensive and slow way to win customers. The company has a good chunk of cash. In addition to the $400 million from four rounds of private financing, it raised $405 million with its initial public offering in November. But $805 million will only buy 23 distribution centers, which doesn't even figure in other costs and covering continuing losses.

The stock has been hammered, along with all the other Web retailers. From a 52-week high of $24.88 a share, it's now down to $7.31. The first quarter of 2000. Webvan lost $57.8 million on revenue of $16.2 million. The good news there is that revenue was four times higher than the quarter before.

Clearly the markets are sending Webvan and other Net companies different signals today than a year ago. The, the feeling was go-go-go-spend money, build fast and get scale. "The rulenow is watch your cash, conserve cash, and where possible consolidate your industry and make sure you're part of the remaining market leader." Shaheen says. That's where the Home Grocer deal comes in.

HomeGrocer isknd of Webvan light. its distribution centers are about 100,000 square feet, one-third the size of Webvan's. Its automation isn't quite as sophisticated. It doesn't carry as many products. Unlike most other Net grocers. HomeGrocer runs its own flett of trucks, though its delivery window is 90 minutes instead of Webvan's 30.

Meshing the two, says Stross, "seems to be a dilution of what Webvan sees as its key advantages" - the supersize distribution centers and deep automation.

But the deal could be the jump-start Webvan needs. Webvan hadplanned to be in 15 markets by the end of 2001. Now, by convertng HomeGrocer sites to Webvan, it will be in 13 markets this year. Over time, Shaheen says, Home-Grocer sites can be expanded and fitted with Webvan technology.

"This is a very nice move," says Mary Alice Taylor, CEO of Home-Grocer. The combined company "doubles its revenue and ctus it scapital requirements in half."

The deal also gives Webvan greater mindshare. It has marketed aggressively in the Bay Area and Atlanta. The new San Francisco Giants baseball stadium is loaded with Webvan messages. But outside those vities, Webvan is barely recognized by consumers. The bigger footprint will give Webvan a greater presence.

The challengers.

Competitors? Webvan seems to keep its eyes on two: Amazon.com and Wal-Mart. Aqmazon.com might someday decide to build local distribution centers for faster delivery. Wal-Mart could use its stores as distribtuiion centers and use the Web and a fleet of trucks to start home delivery.

Shaheen discounts both, for now. Matching a Web-based business toward so-called bricks-and-mortar assets, or marching a bricks-and-mortar business like Wal-Mart toward the Web,, is a major obstacle. "I'd be an idoiot to write Wal-Mart off," he says. But, he adds, "The landscape is littered with people who couldn't do it."

Of course, Webvan hasn't done it yet. For now, it has achieved liftoff and is tilting toward the sky. Whether it stays on course of craters remains to be seen.