Grocers make another go at home delivery

After several failed attempts, grocery giant Safeway is easing back into the home-delivery business. But this time, the company thinks it has the right recipe: technology and know-how borrowed from its new partner, Tesco. The U.K.-based grocery chain operates the world's largest profitable online supermarket, with $422 million in sales annually for the unit.

"We have not lost faith in the home-delivery model," Safeway spokesman Brian Dowling said. "The difference between our previous efforts is that now we have got a model that works. We've got the advantage of Tesco's technology expertise and we're coupling that with our brand name."

Safeway launched home-delivery services in Washington state and Oregon two weeks ago and is quietly testing the system near company headquarters in Pleasanton, Calif., about 30 miles east of San Francisco. Dowling said the company would launch the service in other markets as well.

While there's no guarantee that Safeway has the right formula this time around, lessons learned from other failed efforts and the partnership with Tesco give this latest effort a shot at surviving.

Instead of relying on highly automated and expensive warehouses, Tesco employees pick and pack groceries straight off the grocer's shelves. And rather than wildly expanding into multiple cities at one time, Tesco waits until one operation is running smoothly before moving to another.

By comparison, Webvan operated large, expensive warehouses and expanded nationally before any single region was operating profitably.

When a shopper walks into one of Tesco's 900 stores in Europe and Asia, they might find employees plucking groceries--designated for online customers--off shelves and filling up shopping carts, or "trolleys," as the British call them. Tesco has bolstered the system with technology it has been developing since 1996, when it began home delivery.

Tesco outfits its shopping carts with computers that direct employees to the most efficient route through the store, allowing them to fill multiple orders simultaneously, Dowling said. The computer also scans bar codes on the groceries and electronically rings them up. The system has cut down the time of picking groceries by more than a third, Tesco representatives said.

Help from across the pond
When it comes to building an e-commerce operation, it's almost unheard of for the United States to look for help from outside its borders. With 90 percent of the world's online transactions rung up in the country, e-tailers overseas view companies in the United States as the e-commerce experts.

But no matter how they tweaked their business plans, U.S. companies botched most attempts at building successful Web supermarkets.

They overspent, struggled to make their complex systems work, and failed to entice enough consumers to try them out. Most of the top companies--Webvan, Streamline, ShopLink.com and PDQuick--have closed their doors. Investors sunk more than $2 billion into Webvan and Streamline alone.

"With only 25 percent of online grocery buyers placing orders more than once a month, online-only plays were destined to crash and burn," said Forrester Research analyst Robert Rubin.

Transporting goods to consumers' homes isn't exactly a new concept. A half-century ago, the neighborhood grocer carted jugs of milk or blocks of ice to people's doors. That ended with the emergence of supermarkets, which made it more efficient to stock goods in a large store.

The Internet was supposed to revive home delivery. Customers ordering over the Web could see photos of the products they wanted and electronically place an order. A few days later, a delivery person would lug the groceries to the home.

Although many consumers loved the idea, not enough participated to justify the massive investment needed to get the companies started.

The problems went beyond inefficient warehouses. Soaring gasoline prices and the logistics of trying to deliver within a specific time frame in a large area also hampered the efforts.

The most spectacular grocery flop was Webvan. The company, which had a market value of more than $8 billion soon after its stock was sold publicly, shut down last year after attracting more funding than any e-tailer other than Amazon.com. At its peak, Webvan had plans to expand into 26 cities and had signed a $1 billion contract with engineering company Bechtel to build a network of high-tech warehouses.

A home-delivery pioneer
Safeway, which operates more than 1,700 stores and sees annual sales of $34 billion, was a pioneer of modern-day home delivery. In 1990, Safeway launched a delivery service only to shut it down two years later. Safeway's next attempt came in 2000, when it paid $30 million to acquire a controlling interest in GroceryWorks.com, a regional Web grocer that operated in Texas.

Last year, GroceryWorks was near insolvency until Tesco invested $22 million for a 35 percent share of the company. Safeway moved GroceryWorks' operations near its California headquarters and handed Internet operations to Tesco.

"With Tesco's know-how and the Safeway brand, we have the perfect combination to bring grocery home-shopping to the world's largest market," Tesco executives said in a statement last month.

But success in the United States may be a bit harder won, some analysts say.

Cities in the United Kingdom are much more densely populated than sprawling U.S. metropolises, which could present higher fuel costs and sap efficiency. Americans are also more in love with their cars, Rubin said, and many don't mind driving to the local grocery store.

"Trolley jams" at Tesco
Cracks in Tesco's system have also emerged in the United Kingdom. Several British newspapers have reported that Tesco's online-order pickers have caused "trolley jams" in the aisles of its stores and at times have annoyed customers. The store went so far as to set up "highway code" for its workers in an attempt to eliminate clogged aisles.

A big problem Rubin sees with Tesco's model is the limitations it has tracking inventory. By fulfilling online orders via local stores, Tesco is unable to know what each store has or does not have in stock. That means consumers who order a six-pack of Pepsi from a store that has run out may be disappointed when they don't find it among the items they ordered.

"You impact operations when you're cramming aisles with pickers and their carts or running out of stock," Rubin said. "The in-store pick-and-pack is good for starting up, but as you get bigger you need to change your fulfillment."

Dowling said that "occasionally every store runs out of stock," but that Safeway "prides itself" on keeping shelves stocked.

Rubin said he believes Skokie, Ill.-based Peapod has adopted the kind of hybrid system--a melding of Webvan and Tesco--that eliminates some of the problems Tesco has faced and avoids the financial entanglements that plagued Webvan.

Peapod, the lone survivor among the U.S. group that launched national Internet operations during the 1990s, nearly went broke in 1999 before Royal Ahold, a Dutch grocery company, acquired it and infused it with $73 million. In the early 1990s, Peapod embraced the picking-and-packing model, and then, after the rise of Webvan, switched to a warehouse-based system.

Going hybrid
Under Royal Ahold's guidance, Peapod operates large, central warehouses and has already expanded its customer base in areas including Chicago and Washington, D.C. When it wants to move into a new area, Peapod stocks groceries in backrooms and loft areas within Ahold-owned Stop & Shop stores. This way, customers aren't bothered, and inventory can be tracked and controlled better.

"We start small, market like crazy and build up our business," said Peapod spokeswoman Paula Wheeler. "The mini-warehouses are usually about 1,000 feet and carry 8,000 (different items). Our belief is that once a market is built up, it's more efficient to build larger and centralized warehouses."

While it's an open question whether the Web grocery business can ever turn a profit in the United States, the service did attract a certain audience.

Amy Bohutinsky, a public relations executive for an Internet travel site, was a Webvan customer for more than a year when the company filed for bankruptcy last spring. The defunct grocer was always on time, employed "the politest delivery people" and was easy to use, she said.

"There aren't too many full-service grocery stores in San Francisco, so it was very easy to log on at work and have them bring them by when I got home," Bohutinsky said. "I certainly don't keep as many groceries in my refrigerator since they went out."

Safeway said that its online customers can choose from a range of delivery times, and that it will deliver goods the same day if customers place the order before 10 a.m.

Should Safeway.com start taking orders in San Francisco, Bohutinsky says she'll sign up if the cost for the service is reasonable. In the Pleasanton area, Safeway charges a flat fee of $9.95 for all orders.

"One of the things that Webvan did that bugged me was that they offered the service for free and then started charging unless your order was over $50," she said. "Then they kept raising the minimum to $75 and then to $100...I'll try (Safeway) as long as they treat me right."