On the Web, every click and jiggle of the mouse helps e-tailers customize sites and maximize the likelihood of a purchase. Brick-and-mortar stores have long wanted to track consumers in a similar fashion, but following atoms is a lot harder than following bits. For the most part, they’ve been forced to rely on consumer surveys, says Herb Sorensen, an adviser at market research firm TNS Retail & Shopper in London. “The problem with survey research is the consumer can say one thing and do another.”
To get a better understanding of their customers in real time, mall operators are monitoring shoppers’ behavior with devices that track mobile-phone signals, while retailers including Montblanc, T-Mobile, and Family Dollar Stores are finding new uses for old tools such as in-store security cameras. The goal is to divine which variables affect a purchase, then act with Web-like nimbleness to deploy more salespeople, alter displays, or put out red blouses instead of blue. Until recently, “stores have been a black hole,” says Alexei Agratchev, chief executive officer of consultancy RetailNext. “People were convinced something was true and spending tens of millions based on that” without evidence to back it up.
Agratchev says RetailNext was founded in 2007 to change that. It helps retailers build systems to better understand customer behavior. In most cases, the company relies on the video from a store’s existing security camera system. That feed is run through RetailNext’s software, which analyzes the video and correlates it with sales data. The software can also integrate data from hardware such as radio-frequency identification (RFID) chips and motion sensors to track how often a brand of cereal is picked up or how many customers turn left when they enter a store. The company now has 40 retailer clients, including American Apparel and Family Dollar, and another 20 are testing the systems. RetailNext’s data sometimes refutes conventional wisdom. For instance, many food manufacturers pay a premium for their products to be displayed at the end of an aisle. But customers pay greater attention to products placed in the center of an aisle, according to RetailNext’s analysis.
Luxury retailer Montblanc began testing RetailNext’s video analytics at a store in Miami six months ago. Employees have used it to generate maps showing which parts of the store are best-trafficked and to decide where to place in-store decorations, salespeople, and merchandise. Rodrigo Fajardo, Montblanc’s brand manager in Miami, says RetailNext’s analysis helps his team make decisions faster. “We aren’t taking six months to make a change,” he says. “We analyze one week, and the next week we are making the changes.” He says the software has helped boost sales 20 percent and that Montblanc plans to expand its use to a dozen locations.
T-Mobile employs similar technology from San Francisco’s 3VR, a maker of security systems. Two years ago, 3VR executives realized that its cameras could be used to gather consumer data, according to the company’s CEO, Al Shipp. He says T-Mobile, in Bellevue, Wash., uses 3VR’s technology in some of its retail stores to track how people move around, how long they stand in front of displays, and which phones they pick up and for how long. T-Mobile declined to comment. Now 3VR is testing facial-recognition software that can identify shoppers’ gender and approximate age. The software would give retailers a better handle on customer demographics and help them tailor promotions, Shipp says. “You’ll have the ability someday to measure every metric imaginable. We’re scratching the surface.”
Some retailers are installing gear to track shoppers via cell phones. Path Intelligence, a company in Portsmouth, England, started selling a technology in 2009 that records a phone’s cellular signal and follows its owner through a building. Today it’s used primarily by malls in Europe, and the company says its technology records the paths of more than 1 million customers every day. Some retailers use the data to figure out where in a mall to place their stores, says Path Intelligence CEO Sharon Biggar. Others use it to find out the nationality of their visitors using the country code at the start of their phone numbers.
Shopping centers using FootPath post signs near entrances and mall maps informing shoppers that their mobile phones are being tracked and to turn off their phones if they don’t want to be monitored. But such tracking still concerns privacy advocates. David Jacobs, a fellow at the Electronic Privacy Information Center, says it’s “impractical” to suggest shoppers turn off their phones because so many people use them to meet up with friends. Path Intelligence says it doesn’t record anyone’s identity and alters some of the digits in each phone number before storing it. “We have designed this service so that it’s impossible to detect any personal information or link the number to a person,” says Biggar.
Not everyone is reassured. In November the Short Pump Town Center in Richmond, Va., and the Promenade Temecula mall in California began testing Footpath, the first such trials in the U.S. The test was suspended after one day following complaints from Senator Charles E. Schumer (D-N.Y.) that the technology could compromise shoppers’ privacy. “We would like to address the privacy concerns before moving forward,” says Julia Yuryev, a spokeswoman for Forest City Commercial Management, which owns both shopping centers. J.C. Penney tested the technology in one store but has no plans to implement it, says Rebecca Winter, a spokeswoman for the chain. “The more focused you get on the shopper, the greater their risks,” says Jacobs. Mark Rasch, director of cybersecurity and privacy at CSC, a consulting firm in Falls Church, Va., says that tracking phones or using cameras to glean shopping habits is “no more intrusive than what online retailers do.”
These tools are likely to become more common if other retailers can replicate Montblanc’s success at boosting sales. “It’s really a game-changing experience, and this is only the beginning,” says brand manager Fajardo. “Before we were just working based on certain know-how and intuition. This is designing a retail business based on real statistics.”