Retailers now swim in more data than they know what to do with. And they’re working overtime to digest that data — collected from e-commerce transactions and via merchandising, CRM and POS systems — to glean useful insights. Many are turning to predictive analytics in an effort to use cutting-edge data science to forecast trends and personalize messaging. Data even plays a role in brick-and-mortar stores, where new metrics allow retailers to study in-store behavior at a level of detail never before possible, says Andy Wong, a partner at digital retail consultancy Kurt Salmon Digital. “As we build up more behavioral data on both customers and associates in-store, we’ll continue to find new ways to dynamically optimize the in-store experience and new levers for engagement and conversion,” he says.
Here are six important ways retail and data work hand in hand:
1. Throughout the supply chain
Data is helping retailers obtain end-to-end visibility of their supply chains, which helps improve quality control efforts, among other things, says Jim Hayden, vice president of solutions at Savi, which offers sensor analytics systems. “Companies now use data to track and trace in-transit goods throughout the supply chain, in real time,” he says. For example, motion and light sensors may reveal areas of theft, sensors that detect shock may reveal areas of damage, and sensors may help reveal whether or not food and drugs remained within safe temperatures.
Retailers are also using data to provide more accurate forecasts of the freshness of perishable grocery products, says Venkat Viswanathan, chairman and co-founder of LatentView Analytics. Take yogurt, for instance: “The variety in SKUs in flavors, fat content and other product and packaging features make it increasingly essential for retailers to have more accurate forecasts for a specific SKU type, as consumers are very discerning,” he says. “Sophisticated neural-networks-based algorithms are now able to provide more accurate forecasts that consider regional variations and changing consumer preferences.”
2. In the fitting room
Several retailers, including Neiman Marcus and Nordstrom, have tested “smart” fitting room mirrors that can collect information on customer shopping habits and make recommendations. That data can also help retailers understand how customers react to products, says Oliver Guy, global retail industry director at Software AG. For example, knowing that a given shirt is often taken into the fitting room with a given pair of jeans may be an important insight for merchandisers. Or a correlation between sizes tried and sizes purchased can indicate whether the sizes of various items are accurate. “We are seeing technologies that allow this kind of data to be harvested,” Guy says, adding that data can even help in-store staff estimate wait times outside fitting rooms so they can direct shoppers to alternative facilities if a line is too long.
3. For personnel decisions
Thanks to POS data and data collected via mobile devices, retailers have more information about the performance of their sales associates than ever before, says Wong. The metrics include number of customers assisted, average response time and cross-sell frequency, and that data is available by region, store or department, and even for individual employees. “This kind of granularity of performance has allowed retailers to start optimizing the labor model in stores and measuring results,” he says. For example, he explains, one of Kurt Salmon Digital’s retail clients uses data to optimize staffing levels, and not just around peak periods but based on real-time needs in specific parts of the store — based on what they learn from the data, managers can move employees nimbly from department to department as the need arises.