Platt Retail Institute Finds RFID-Based Inventory Accuracy, Sales and Satisfaction Gains at Macy’s

By Claire Swedberg

The study, undertaken during the past 15 months, found that the retailer not only improved its on-shelf and display compliance, but made single-unit sales possible when a store's inventory may be down to its last unit for a particular SKU.


The use of radio frequency identification technology at Macy’s stores has boosted the global retailer’s rate of on-shelf display compliance and overall inventory accuracy, while also lifting customer satisfaction and enhancing omnichannel fulfillment based on in-store, single-unit accuracy. That was the finding of a report released by The Platt Retail Institute (PRI), developed in cooperation with the Northwestern Retail Analytics Council (RAC).

But those findings are just the beginning, says Steven Keith Platt, the Platt Retail Institute’s director and research fellow and RAC’s research director. According to Platt, the technology offers retailers such as Macy’s the opportunity to conduct a wide variety of analytics based on product tag reads, including better management of fitting rooms, determining best merchandise placement techniques, and managing pricing and trends predictions.

Steven Keith Platt

PRI’s working paper, the research for which took about 15 months to complete, quantifies the benefits of RFID data for Macy’s. The researchers tracked Macy’s technology implementation results in four use cases. These consisted of RFID’s use with display items in the women’s shoes departments across all of Macy’s U.S. stores, the overall inventory accuracy as measured by the gross unit variance (GUV) averages from 2013 to 2015, single unit fulfillment for omnichannel or in-store sales and, finally, back-room to store-front replenishment (see Macy’s Launches Pick to the Last Unit Program for Omnichannel Sales, Macy’s Expands RFID and Beacon Deployments and Macy’s to RFID-Tag 100 Percent of Items).

The Platt Retail Institute is an international research company that focuses on technology’s use and impact on the customer experience, while RAC is a consumer shopping behavior-based research organization consisting of researchers from PRI and Northwestern University. PRI launched the research not only to present detailed findings regarding the use of RFID in the retail environment, but also to demonstrate how RFID-generated data can be integrated with other information managed by a store to provide greater business insights.

“We found that the value of RFID data is already high, but when you start to use it in other applications, the value rises even further,” Platt says. “RFID is such a rich source of information—the benefits go way beyond the supply chain or inventory management.”

Case one of the four-case study found that when it comes to display compliance in women’s footwear, the use of RFID brought the rate of errors (incidents of available shoes not being displayed on shelves) down. Macy’s carries more than 250,000 stock-keeping units (SKUs) of women’s shoes, and a typical shoe department features about 800 different styles. The high volume of styles can lead to a lack of compliance in terms of displaying each style for customers.

Prior to the RFID deployment, the rate of errors (non-displayed shoes) was 30 percent. With the RFID system in place, the research indicates, that noncompliance was reduced to between 4 and 6 percent. Customer satisfaction was also higher than in other store departments, presumably due to the women’s shoes department’s greater display compliance—more products are on the shelf where customers can see them.

In case two, the study focused on GUV in the men’s departments at eight New Jersey Macy’s stores. Here, the retailer carries an average of 297,666 SKUs. GUV variances occurred due to employee or vendor error, merchandise loss or theft, and infrequency of cycle counts. The researchers compared two non-RFID-enabled brands against an RFID-enabled brand, and found that RFID-enabled brands showed between 17 and 24 percent less year-end variance than those that were non-RFID-enabled. The data also found that better inventory accuracy, in general, led to fewer markdowns.

In case three, the study examined single-unit fulfillment, which is necessary for omnichannel sales. Without RFID technology, Macy’s fulfillment system only enabled an online purchase if it could determine that multiple units were available at a particular store, since inventory accuracy could not be fully trusted. Approximately 20 percent of Macy’s sales are single-unit purchases, meaning that the ability to sell a single item from any store is especially important.

Additionally, the study examined the results of Macy’s six-month fulfillment test of online orders of women’s social-occasion dresses at eight stores (four RFID test and four control) throughout 2014. It found that making single merchandise units visible and available for sale was possible with RFID. Test stores using the technology outperformed the control stores based on fulfillment requests, units picked and units sold. During the five-month period of the case-three study, the researchers found that the rate of locating and selling merchandise with RFID was about 6.1 percent higher than at the control stores. At the four RFID test stores, the visibility of single items produced an increase of 293 percent in unit requests.

Finally, case four examined the accuracy of inventory in back-room to store-front replenishment. Macy’s has reported to researchers that 13 to 20 percent of its merchandise is kept in store stock-rooms. Of that product, prior to the RFID system’s implementation, 5.5 percent of merchandise that should be on the sales floor was not. The study further found a strong correlation between sales and units brought to the store front for customers to view. Not surprisingly, the research determined that products not sold during the year, once picked from the back-room and put on display, sell at a higher rate than those not displayed.

In addition to these key areas of research, the study found a potential for better demand forecasting, merchandise trends understanding, dynamic pricing, fitting room use, in-store marketing and effective merchandise placement with RFID. For instance, Platt says, demand and trend forecasts are key to effective merchandise management, and RFID can enable managers to understand inventory levels, view trends, and thereby make better buying decisions and reduce overstock. Variable pricing can be adjusted according to sales and browsing trends, he adds. Tracking merchandise brought into fitting rooms also enables retailers to identify units that are popular, as well as those that are tried on but not purchased—which could indicate a potential problem with a fit, for example.

The study was sponsored by Avery Dennison, which has partnered with Macy’s to develop custom inlays and labels for its product categories, and Tyco Retail Solutions, whose inventory software the retailer is using. It was co-sponsored as well by Zebra Technologies, which provides Macy’s with its MC3190-Z mobile RFID reader.