Home delivery is an entitlement in the United States. You shop and the store delivers. The practice dates back 100 years. The old economy service works so well in our local Food Emporium in New York that customers who pay $3.99 to have their shopping delivered have to hurry home before the delivery man gets there, so as to pay the tip, also regarded as an entitlement.
But physically going to the grocery store, loading up a squeaky wire trolly and queueing at the checkout was supposed to become a thing of the past after on-line shopping arrived a few years ago.
That's when companies like California-based Webvan.com tried to change our habits with virtual supermarkets.
Webvan was the hottest of the food-supply dotcoms. It delivered groceries free in big green and yellow plastic boxes. These were packed by automated carousels and conveyors in high-tech warehouses costing $25 million each. In 26 US cities Webvan guaranteed delivery of toilet rolls and baked beans within a 60-minute window.
The billions of dollars needed to pay for this nation-wide network was provided by prestigious backers such as Goldman Sachs and Sequoia Capital.
Surely they knew what they were doing?
But last week Webvan went out of business, one of the biggest casualties yet of the Internet bust.
Its operating loss went from $13 million in 1998 to $438 million in 2000. Together with HomeRuns.com, a smaller east coast virtual supermarket which also collapsed last month, it blew $860 million.
So what went wrong? Webvan clearly over-estimated the number of customers willing to embrace the idea of ordering their avocados on PC screens, and under-estimated the sensuous pleasure of real shopping, where the customer can squeeze the avocado to test its ripeness, or pick out the best-looking side of beef.
Basically Webvan lacked a working business model. It over-invested in a business with a net margin of just one per cent. Some 1.3 million wired households signed on, but it was nowhere near enough to ensure profitability.
"We made the assumption that capital was endless and demand was endless," said chief executive Robert Swan, who took over the sinking enterprise in April.
"We had started to think of the customer as a statistic or number instead of a Tom or Sally."
By then Webvan had switched to low quality produce, which just sent customers who got damaged avocados back to the supermarkets.
This does not mean that online shopping at virtual stores has no future. Some pure retail dotcoms are doing well because they did not try to go too far too fast. SimonDelivers.com in Minneapolis has 43,000 customers buying an average $4,200 in groceries a year. Peapod.com, the largest remaining virtual American grocery, now owned by Dutch chain Royal Ahold, has survived by confining itself to New England and Chicago.
The New York research firm, Jupiter Media Metrix, estimates that e-grocers have the potential to become an $11 billion business in the US within five years, though that's still only 2 per cent of the market.
The lesson for on-line retailers may however be that it is better to have an established, old fashioned, walk-in store before going on line.
The Albertsons chain of 2,500 stores, the second-largest food and drug retailer in the US, is experimenting with five stores in Seattle where assistants equipped with wireless devices fill internet orders from the shelves, and they are finding it works.
As Webvan went Albertsons tripled their online sales.
American stores are actually a bit behind Europe in adding Web-based initiatives. In Ireland Superquinn already has an on-line delivery service used by more than 1,500 shoppers who live in the vicinity of its 17 large stores and eight shopping centres and who are willing to pay a £5 delivery charge.
Tesco.com, the on-line service of Britain's biggest food retailer Tesco, is breaking even after two years and is making a foray into the United States.
Last month it signed a deal to export its on-line expertise to Safeway, which has 1,650 stores in the US and Canada. Safeway's GroceryWorks.com shopping channel had been losing money for the same reason as Webvan; it operated from high-tech warehouses with limited choice.
On-line retailers have been forced by the dotcom meltdown to take more account of the human factor. Just because something new and innovative works doesn't mean people will buy it. Cisco Systems for example has developed software for wired-up refrigerators which can alert on-line suppliers when the orange juice runs out. But who needs it?
Perhaps the most user-friendly concept is the "store-pick" model, being developed by Safeway and Albertsons. It works this way. After you punch in your website order, the groceries are collected not from a cavernous warehouse but from your nearest supermarket, by a company employee sent to pick the items from the shelves. It cuts transport costs and speeds delivery times. Of course, just as 100 years ago, you still have to tip the delivery guy when he rings the doorbell.
