SHOPPER TRACKING TEST YIELDS CLUES AT MARSH

DENISE ZIMMERMAN

INDIANAPOLIS -- An innovative shopper-tracking system at Marsh Supermarkets has begun yielding clues about how to boost aisle traffic and increase transaction size and time spent shopping the store.

"We found that by moving key destination categories around and eliminating dead spots in the store, we were able to increase traffic up to 30% in some of our slower-moving aisles," said Mark Heckman, director of market research for the chain here.

The test system, in place at one Marsh store since last October, relies on two-way infrared technology linking transmitters on shopping carts with sensors in the ceiling to track shopper movement. Fifty shopping carts and four handbaskets have been equipped with 2-inch transmitters. Marsh is about halfway through a one-year test of the system.

The key to creating a more profitable store layout and merchandising scheme is finding the optimum mix of destination categories -- those commodity items shoppers typically come to the store determined to buy -- and high-margin impulse products, Heckman said in an interview with SN.

By closely monitoring shopper activity through the tracking system, the chain has identified several dead areas in the store, areas that shoppers were avoiding. In response, the chain began experimenting with alternative category mixes to encourage shoppers to explore aisles they typically had been passing by, Heckman said.

In one scenario, for example, "we eliminated some slower-moving [stockkeeping units] in cosmetics, expanded greeting cards space, added more destination categories and created a promotional section in one aisle," he said. The result was a 30% increase in traffic in those areas. "We were drawing people to the center of the store who were shopping the ads and specials, as opposed to having them shop [only] on the perimeter."

In another scenario, Marsh examined the traffic patterns around an endcap display of Coca-Cola and learned that 55% of shoppers who bought that sale item never went down that aisle.

"You have to ask yourself, 'What are we creating in our stores?' " Heckman said. "Are we creating an environment where people are trained to buy sale items off the ends? If they are trained in that manner, what are we losing by not sending them down the aisle to get the product?"

Marsh also used the Coca-Cola endcap to examine the impact on sales when shoppers were exposed to a product twice, in two locations.

About 30% of shoppers bought the product off the end cap display without traveling down the aisle. But when shoppers traveled through the aisle and a saw the product a second time on the end cap, 34% of them bought Coke.

Heckman stressed, however, that the test findings were only preliminary and the results anecdotal.

"We did the same thing with a toilet tissue product on sale, for instance, and didn't see [the same results]. I don't know if I have a conclusion. But I want to show the kinds of things that can be measured with an in-store electronic tracking system," he said.

The tracking system was developed by Codem Retail, Hollis, N.H. Codem installed the device at its own cost to give the system its first full-scale field test at Marsh. The device is "transparent to the customer, so you're not asking the customer to get involved -- to wear a badge or fill out a survey -- so we think it has a lot of inherent advantages," Heckman said.

Two years ago Marsh did ask its customers to get involved with research, surveying them extensively at the exit after observing their shopping behavior from atop catwalks in the store -- and the result was the Marsh Super Study.

The electronic tracking system being tested today is a logical extension of that five-store Super Study, which had costly and laborious conventional research methods, Heckman said.

Results derived from the electronic shopper tracking test are expected to augment the Super Study's findings. Whereas the Super Study produced strategies to maximize dynamics within a particular category, Heckman said the tracker system may reveal clues about relationships between different categories.

"That's the element the tracker brings to the category management paradigm, in that we are using category location, adjacencies, relationships as a viable factor," he said. "That's what the tracker system brings to the table."

Though its data-gathering methods are not as arduous as the Super Study, the electronic tracking system is expensive, he said. The chain probably will not expand the tracking system to other stores without enlisting partners willing to share the expense.

Heckman suggested that a direct-store-delivery vendor, for example, could use the tracking system to monitor its own operations and evaluate the efficiency of unloading and taking product from the back room to the display.

"Doing in-store research is wonderful, but when you're talking about a system that costs anywhere from $75,000 to $100,000 per store, the return on investment is not going to be there just by doing in-store research for the chain supermarket that's involved," Heckman said. "There's going to have to be some outside involvement."