Grocers Count On IT To Keep Cash Registers Ringing

Webvan may be dead, but its technology and its dream live on. Coming soon to New York: Fresh Direct, an online grocery-delivery service, running in part on some of the much-heralded and now-defunct Web grocer's equipment, bought at auction this summer.

Fresh Direct's debut is still under wraps, and company executives declined comment, but it appears they've learned a thing or two from history. First, they're keeping their delivery area small, restricting it to the population-dense Big Apple. They're keeping things simple, too: The merchandise selection will be light on nonperishable goods, just enough to cover the bases, and heavy on fresh produce and meat, which have the highest profit margins. Delivery will be available only at night, when more people are home. If a customer doesn't answer the door, drivers will leave an order with a supervisor-behind the wheel of his own van in a central location -who will return later to make the delivery.

"They learned from the mistakes of those before them," says Timothy Laseter, VP at consulting firm Booz Allen Hamilton. Laseter has talked with two of the company's top executives, Steve Yeboli and Jason Ackerman, who have experience setting up a gourmet grocery. But because it has opted not to operate brick-and-mortar stores, Fresh Direct faces the same distribution problem that hamstrung Webvan: It had to spend millions on a distribution center, now being upgraded to handle food processing. But without physical stores, the company has no other choice. In mid-1999, Webvan signed a $1 billion contract with Bechtel Corp. to build grocery distribution centers in 26 U.S. cities in a two-year period. The company said a large part of its expense-$25 million to $35 million for each distribution center-was an investment in fulfillment infrastructure.

When Webvan Group Inc. filed for bankruptcy in July and officially bit the dust after burning through $830 million in two years, many in the industry sounded the death knell for online grocery-delivery services. It's a niche business, industry experts and analysts said. "Online grocery is a murderously difficult market," Miles Cook, a VP at consulting firm Bain & Co., said at the time. The media pointed at neighborhood grocers, who've been carrying milk and bread to doorsteps for decades, as the only viable delivery service.

One of those is D'Agostino Supermarkets Inc., a family-owned chain of 22 stores, mostly in New York. D'Agostino always offered delivery-that's how many of its Manhattan customers prefer to shop-and charges $3 for the privilege. In June, a few weeks before Webvan went belly up, D'Agostino began to take orders over the Internet; by Labor Day, the service was available from all of its stores.

MyWebGrocer.com runs the company's Web site, as it does for 28 other regional grocers across the country. It charges D'Agostino $6 per order; the grocery chain, in turn, charges its customers $10 on top of the $3 delivery charge. About 250 people order via the Web every week, says Nick D'Agostino III, VP of corporate administration, whose grandfather founded the first store in 1932. "We run it as a service," he says. "That was one of Webvan's biggest problems. They didn't realize it was a service."

D'Agostino doesn't have a multimillion-dollar automated distribution facilities or fancy conveyor belts zipping through warehouses; it picks items from its store shelves to fill orders. When shoppers log on to the Web site, they enter the neighborhood in which they live. That's the same approach taken by some of the world's largest grocers: Tesco in the United Kingdom, and Albertson's, Safeway, and Royal Ahold in the United States. Each has anted up tens of millions of dollars to set up sites that fill orders from their own brick-and-mortar stores.

They've had mixed results. Tesco has been the most successful, largely because net margins for grocers in England are close to 8%-four times higher than on this side of the Atlantic, Cook says. Tesco has done such a good job-online sales were about $335 million last year-that its competitors are beginning to cede the market. In late November, after less than eight months of an eight-store trial, Safeway plc in the United Kingdom shut down its online service. The chain, which is separate from U.S. Safeway, wasn't even delivering-customers ordered on the Web and picked up their groceries at a store.

If nothing else, the breakneck turnover of inventory requires the industry to rely on technology. The average household shops at a grocery store 2.2 times a week and buys 40 to 50 items; the average consumer spends about $4,000 a year there, AMR Research director Peter Abell says. "It's a low-margin, high-volume business," he says. "You better be using IT, otherwise you're going to be hurting to make a profit."

Grocers such as Big V Supermarkets Inc. and drugstore chains such as Longs Drug Stores Corp. have used PriceCenter software, developed by DemandTec Inc., to measure the net profitability of individual items for sale and then adjust the prices to sell more of those items that carry higher net margins.

Big V, in Florida, N.Y., operates 31 Shop Rite stores and does about $1 billion in sales, mainly in New York and New Jersey. Its shoppers are motivated by price more than any other factor, CIO Kevin Sterneckert says. Big V, which is now in Chapter 11 bankruptcy proceedings, conducted its own pilot program with PriceCenter in 15 stores recently and increased its net profit 1.67% without increasing shoppers' bills. "I can't think of a retailer in the country that won't be interested," he says.

He hasn't met Ukrop's Super Markets, a 60-year-old family-owned chain of 28 stores in Richmond, Va. Jeff Thomas, Ukrop's director of category planning and analysis, isn't interested in the technology. The reason? Every time he changes a price, Thomas must print new labels, stick them on shelves or individual items, and then tell everyone from his advertisers to his point-of-sale devices to his warehouse billing system that he's made a change. So the savings don't add up, he says.

Eventually, he could use electronic shelf labels, which now cost $4 to $6 each, depending on volume, to change prices on the fly. But not yet. "We have an interest down the road," Thomas says. "The price will start to get more affordable."

Ukrop isn't alone. Many grocers are hamstrung by the labor costs of implementing the pricing changes suggested by DemandTec and vendors such as Khimetrics, whose software applies mathematical algorithms to sales and inventory data to either maximize profit or minimize markdowns.

"One of the problems of Khimetrics and DemandTec is they have great math, but you still need to execute" the price changes, AMR's Abell says. "Until there's a link with electronic shelf labels, you need to apply all those neat prices manually on a store-by-store basis. That's expensive."

Still, Sterneckert sees great potential for saving money if retailers are willing to take it a step further and link pricing decisions with replenishment systems. "It's on the verge of being a key part of delivering merchandise to the store," he says. "Today, replenishment systems don't use the effect of price in their calculations to determine replenishment quantity. It's very clear for most retailers that price significantly impacts how items are going to move. The replenishment system doesn't know you've changed the price a dime."

Consumer perception is important, too. Sterneckert concedes that he's careful not to make his shoppers feel they're being forced to buy certain items, such as if they notice the price of their favorite brand of cereal has been raised while a rival brand is cheaper.

The same applies to another technology grocers are eager to implement: self-checkout lanes. The lanes are great for supermarkets-they decrease labor costs and give what Abell calls the "illusion of quicker checkout." The biggest complaint about grocery shopping is waiting in line, but because most people are slower with a scanner than a cashier, the do-it-yourself aisle isn't necessarily always the solution.

"Self-serve checkout doesn't work for everybody, like a mother with 3-year-old twins and $300 worth of groceries, but it does work for a guy with a copy of Esquire and a six-pack," says Cathy Hotka, VP of IT at the National Retail Federation, a retail-industry trade association.

Supermarkets are installing self-checkout lanes from vendors NCR, Optimal Robotics, Productivity Solutions, and PSC at a 125% faster rate this year than in 2000, Abell says. The leader is Kroger, which has about 980 in its 2,400 stores. Because the self-serve lanes don't have long counters, Kroger can replace two checkout aisles with twice as many self-serve ones, where shoppers scan and bag items in one motion. A single clerk hovers near-by to help.

Kroger says self-checkout accounts for about 20% of sales and 40% to 50% of customer traffic-mostly from people with a few items who don't want to wait in a 15-items-or-less express line. "They feel more in control," Abell says.

Big V has operated four self-checkout lanes in one store for 16 months, and Sterneckert estimates the payback on his investment, about $100,000, to take 10 months. The price of the system varies by the configuration and number of stores installing it and the types of tender it will accept.

"The store manager swears by it," Sterneckert says. "The customers that don't use it still believe they benefit because the lines are shorter." About one-fifth of the stores' total sales pass through the self-checkout lanes.

What Abell is really excited about is still at least 18 months away: the development at MIT of the electronic product code, a radio-frequency chip designed to replace UPC bar codes. Attached to every product and every pallet throughout the supply chain, the tags would do more than just provide the physical location of an item-they would alert retailers to items that are close to being out of stock, reduce inventory, speed up delivery, and even check freshness.

There are benefits for consumers, too. "When that item is in your grocery cart, you may not ever have to check out," he says. "It would automatically pull all prices together, and on your way out, you might just show a credit card, and get a printout of your receipt."

The tiny devices are being tested now in field trials, and Abell is eager to see them adopted. "Grocers are 8% to 10% out of stock by the end of the day, on average. That's a lot of lost sales for that item for the manufacturer." Major consumer packaged-goods makers such as Johnson & Johnson, Proctor & Gamble, and Unilever all are chipping in to help with development.