Shelf science in supermarkets

Professors' software notes how products relate to each other

March 17, 2002|By Lorraine Mirabella | Lorraine Mirabella,SUN STAFF

Most grocery shoppers don't give a second thought to where Frosted Flakes cereal or Green Giant peas sit on their supermarket's shelves.

But for grocery managers, deciding which products to sell and how to arrange them is a highly specialized science that can boost profits or hurt sales.

In a recently published study, researchers specializing in company operations say they've found a better way to manage categories of food products. Professors at three universities say grocery stores that use the professors' new models - based on information from consumers and grocery managers - can boost profit by as much as 25 percent.

"A lot of stores struggle with the issue of limited shelf space and try to figure out what kind of products to carry on the shelves," said Christopher S. Tang, a professor at the Anderson Graduate School of Management at UCLA and one of the study's authors. The research looked at "how to utilize the shelf space optimally to generate the highest profits," Tang said.

Doing so effectively has become more crucial because of the explosion of new products, Tang said. Shelf space has grown as supermarkets have gotten bigger, but the number of products introduced each year has grown much faster. The average grocery store carries 30,000 different items, but "there are hundreds of thousands to choose from," Tang said. "If you have the wrong products, customers don't come in, and products sit on the shelf."

Food retailers have relied on what is known as "category management" for items such as soft drinks, bottled water and salad dressing to determine which products to stock, which to remove, how much to carry and where to put them, according to the Washington-based Food Marketing Institute. The 8-year-old practice has become more sophisticated over the past few years, thanks to improved technology.

"It's a way for retailers to differentiate themselves from their competitors, but also to position themselves with the consumer," said Patrick Walsh, director of industry relations for the FMI. "If consumer needs are not met, it won't translate into sales."

Retailers track what is selling by scanning products at check-out. Information goes to databases, then into category management software programs, which analyze what shoppers are buying and what they're leaving on the shelves.

But that's often the wrong approach, argued Tang and the other researchers, Teck-Hua Ho, of The Wharton School at the University of Pennsylvania, and Juin-Kuan Chong, of the Business School at the National University of Singapore.

About 70 percent of the time, supermarket chains manage categories by pulling slow-selling items from the shelves.

"They're not thinking about the relationship of one product to another product," or realizing that removing one item could affect the sales of remaining, related items, Tang said.

For example, if a brand of lobster bisque happens to be a slow seller, most category management models would direct store managers to remove it in favor of a better-selling soup. But if the bisque has a unique quality, for instance, that it can be mixed with other soups to make a sauce, removing it could hurt sales of remaining soups. It might make more sense to remove one of two brands of, say, cream of chicken soup, the research shows.

"Our model would tell you, under certain conditions, to remove one of the cream of chickens, given limited shelf space," Tang said.

The findings, which focused on categories including ice cream, potato chips, barbecue sauce, coffee, frozen pizza and sausages and hot dogs, were published in January in the trade journal Manufacturing and Service Operations Management. The data was based on interviews with grocery store managers at some of the nation's biggest chains.

Researchers said their approach also differs from commonly used "category management" in that the models are based on detailed consumer behavior as opposed to sales.

The consumer portion of the research, based on two years' worth of scanner data from five large Chicago supermarkets, showed that shoppers are less loyal to brands than they used to be and are more interested in variety.

"By and large, they are variety-seeking," Tang said. "They may not stick with the same brand or flavor. They try out different things."

For supermarkets, that means consumers prefer a wider variety among fewer brands instead of more brands of similar products.

Supermarkets should be asking: "Are you offering different varieties to consumers, rather than duplicating brands with the same formula?" Tang said.

In the ice cream category, for instance, supermarkets "may want to reduce the redundancy of ice creams that have the same flavor. They might want to carry more flavors, so customers feel the selection is bigger," Tang said. And, he said, "You don't need to carry so many toothpastes with the same formula. Consumers would rather have different formulas than different brands. ... That's how they perceive variety."

Baltimore Sun Articles
Please note the green-lined linked article text has been applied commercially without any involvement from our newsroom editors, reporters or any other editorial staff.