The case for item-level RFID

Adapted from the forthcoming book, My Incredible Supply Chain Journey ... And What You Can Learn From It. Reprinted by permission.

I am convinced that one of the main pillars of a comprehensive business management process is manageable and user-friendly technology. The right technology, properly applied, can make the difference between adequate and excellent.

For example, companies struggle with dramatic forecasting errors and other supply chain management (SCM) challenges that are largely the result of inaccurate information. But by using technologies like item-level radio frequency identification (ILRFID), retailers in particular can make big gains in accuracy in terms of inventory tracking, demand forecasting, and other activities that impact their ability to have products on store shelves where and when consumers want them.

Those who know me know that I am passionate about ILRFID. I am completely convinced that this technology will, over time, replace the bar code, and that it will have a major impact on supply chain management across all industries. In addition, ILRFID will facilitate omnichannel merchandising, which seeks to provide consumers with the ability to order product any time, from anywhere, to be picked up from or delivered to the location of their choosing. Inaccurate inventory information will truly have a major impact on a retailer's ability to meet the expectations of the consumer who chooses one or more of the order and delivery options. Because item-level RFID dramatically improves inventory accuracy, it will be a key enabler of omnichannel commerce.

A brief introduction to ILRFID
RFID—radio frequency identification—is a technology that uses radio waves to transmit an item's unique identifying information, such as a serial number, or product and shipping information, from a tag to a reader. The reader takes that information and sends it to a computer system. (I won't be able to go into great detail on this complex subject here; for more information, I highly recommend RFID Journal, including its magazine, website, and educational events.)

ILRFID involves tagging and tracking individual items such as apparel or other products in the retail supply chain. This technology can help address and overcome some of the common problems associated with inaccurate inventory information that retailers continue to struggle with today, as they have since consumers began shopping at retail outlets. To understand how it does so, it helps to first understand what those problems are.

The current system of inventory control has a history of errors at every touch point, and at each of those touch points, there is a cost associated with the error. On top of that, the current methods of capturing product information provide marginal accuracy. For example, reading the bar code with a scanner requires line of sight and being within 9-12 inches. Given this circumstance, physical inventories take a long time and are very costly. Consequently, most companies will take a physical inventory once or twice a year, at their distribution center and in their stores, most often using an outside contractor. Some companies use cycle counting, which means they inventory product groups at different times of the year, hoping this will solve their problem. And this could be true, but there are a lot of conditions to be considered; for example, the number of products, importance of date codes, and the number of manufacturing and distribution points.

Typically, when a physical inventory is about to be taken, the accuracy of the product inventory—which takes into consideration the size, color, and style of a stock-keeping unit (SKU)—is in the range of 65 percent to 70 percent. The inventory count is somewhat accurate after a physical; however, accuracy declines at a rate of 2-3 percent per month. So it is understandable that at the end of a season or sales cycle, markdowns are prevalent; they are an attempt to sell product that didn't move during the season.

Among the typical reasons why that happens is that the product is in the wrong store because the forecast was seriously in error, or the product was hidden in the storage area of the store and not brought out to the sales floor, or if it was on the sales floor, it was on the wrong rack or rounder and was not noticed by the sales clerks. As a result, it was missed by the consumer. From a practical standpoint, just walk the garment floor of a retailer and ask yourself, "How difficult is it to find a particular garment if it is on the wrong rack?" I just went through this experience myself at a retail store. I enlisted the sales associate to help me find the garment I was looking for that frustrating afternoon. Well, she checked her inventory report, and the product was shown as being available, but it was nowhere to be found. She promised to follow up, find the garment, and call me. It's now three weeks later, and the call never came! The purchase was never made.

Forecasting software will make adjustments, given the expectation of errors, and apply algorithms and triple exponential smoothing (a statistical-analysis method that captures important data patterns over a period of time). To account for the expected forecasting errors, inventory management systems will provide safety stock, typically in the 2-3 week range.

The major message to be gained from the scenarios outlined above is that incorrect inventory information results in a substantial amount of extra effort, cost, and lost sales! When the product isn't available when the consumer is ready to buy, it causes numerous problems in terms of consumer confidence. The damage done to consumer confidence is exacerbated when the product is on promotion and not on the sales floor. And just imagine the damage to the reputation of the store if the inventory information is incorrect and the product is ultimately found and moved to retail, but too late—the consumer has discovered the same item on sale at a competitor's store and buys it there.

These problems can be avoided with item-level RFID. ILRFID is a "magic wand" that can read tagged product even when it is inside a case!

Here's an example. We have successfully used a process called the electronic proof of delivery. Here's the way it works. The supplier packs out a case of product—let's say several pairs of socks, a few T-shirts, and for good measure, a few boxer shorts. The supplier prepares a packing slip identifying the customer and contents of the carton. The supplier electronically "reads" the contents of the case to ensure there is agreement with the packing slip. The information from each case that is packed and read is used to complete an Advance Ship Notice (ASN), which is transmitted to the retailer. When the order is received by the retailer, the case is read again, and it's highly probable that there is agreement in the 99.8 percent range.

Typically today, using the bar-code system, the retailer will pull up to 10 percent of the cases off the receiving line to check them. If there is an inaccuracy—the ASN and the case contents are not in agreement, for example—then the retailer will short-pay the invoice and complete reconciliation when the next inventory is completed.

Now, suppose that item-level RFID had been employed in the example above. Imagine the productivity that could be realized on the receiving floor, the rapidity of turning the door and truck around, the ability to read each individual item inside the case, and more. Imagine how, with ILRFID, a retailer could run product off an inbound truck and directly to an outbound carrier destined for a store. It's about customer satisfaction, solid retailer and seller relationships, and streamlined supply chain operations. One can easily come to the conclusion that every touch point can be positively affected by the use of ILRFID.

I'd like to take a moment to bring in the concept of Electronic Product Code (EPC)-enabled ILRFID—the nirvana that industry has been seeking since the advent of the bar code. My friends at GS1 US, the standards organization that now encompasses VICS (Voluntary Interindustry Commerce Solutions Association), have done a lot of work in this area. Here's how they describe it on the group's website:

The use of the Electronic Product Code (EPC) with RFID provides an effective bridge from today's barcode-based systems to RFID without losing your current investment in identification systems, transaction systems, or other information sharing techniques. EPC-enabled RFID, coupled with serialization techniques, represent a way for your company to extend your current investment in the GS1 Standards to take advantage of the benefits of RFID as a data capture technology, while still using the GS1 standards for identification and data exchange.

We are almost there and will definitely see this great technology increasing by leaps and bounds in the very near future. You can read more about this subject at www.gs1us.org.

Item-level RFID moves forward
My first hands-on experience with ILRFID was when I was with VICS and visited El Puerto de Liverpool in Mexico City in 2007. Liverpool is a Mexico-based department store chain that at that time had 78 stores with annual sales of US $3 billion. The previous year, the retailer had launched a comprehensive RFID solution for its 200 suppliers to help improve its inventory management processes. Liverpool's goal was to increase merchandise availability to near-perfect levels.

I was thoroughly impressed with how Liverpool had Levi Strauss & Co. source-tag jeans, and with how jewelry and eyeglasses were item-tagged in the retailer's distribution center. I was especially impressed with the jewelry that was in open-front displays, which allowed the shopper to inspect the product without a sales clerk present! This was also true for expensive eyeglasses. Security was ensured because a store clerk had to deactivate the tag before merchandise could leave the store; otherwise, an alarm would sound. Consequently, shrinkage was almost nonexistent.

Levi's source-tagging led to a smoothly operating supply chain. It allowed for the replenishment of product based on consumer purchases, with a relatively short turnaround time. This was a major factor in ensuring that product was available to meet consumer demand. (Today Liverpool's sales exceed US $4 billion across 85 stores, and the company continues to pursue its goal while expanding the use of item-level RFID to other product categories.)

I found myself intrigued by item-level RFID, and followed up with a conversation with Bill Hardgrave, director of the University of Arkansas (UARK) Walton College RFID Research Center. We discussed how VICS and the university could work together to support the use of ILRFID, starting with item-level tagging for apparel/footwear. I also contacted Mark Roberti, owner and editor of RFID Journal. He shared the history of RFID and how the concept was first developed during World War II to identify aircraft, and he explained how commercial applications expanded following work done by the Massachusetts Institute of Technology's Auto-ID Lab. This helped me understand the development of this technology and the credible organizations that had been involved for a number of years. (You can read his fascinating article on the history of RFID.)

Early on, RFID was mostly used on cases and pallets for tracking purposes. Wal-Mart Stores was among the first to do this, mandating that some of its suppliers use RFID. Although this was a thought-leading idea, the value was not there for the suppliers, and that first initiative did not really take hold. Nevertheless, the technology did work! Walmart's experience reinforces the idea that RFID has to be a collaborative initiative, with both the retailer and the supplier realizing benefits.

But people had been thinking about item-level RFID for quite a few years. In 2007, Bill Hardgrave, David Cromhout, and Justin Patton conducted research on why retail was slow to adopt item-level RFID. Not long after that, VICS and UARK established a partnership with the aim of launching some test pilots for the application of ILRFID in retail.

Meanwhile, the UARK RFID Research Center in 2007 had partnered with Dillard's, J.C. Penney Co., and Macy's-Bloomingdale's to study improvements in inventory accuracy through item-level RFID tagging. The retailers agreed to conduct two-year pilots in collaboration with the Research Center team and several technology providers.

The Dillard's pilot showed consistent inventory-accuracy improvements thanks to item-level tagging, with the results published in April 2009; the Macy's/Bloomingdale's results, published in August 2009, showed improved inventory accuracy and reduced shrinkage; and the J.C. Penney study showed improved inventory accuracy, with the results published in April 2010. A big surprise was that each retailer also achieved an increase in sales because in-stock rates were higher. (The results of the pilots, as well as reports on UARK's research on applications for apparel, pharmaceuticals, and other industry segments, can be found here.)

There is little doubt that the success of these pilots spring-boarded item-level RFID into global prominence, creating substantial interest due to the fact that each of the retailers realized an increase in sales. The most significant interest and activity has been in the apparel category, with 19 of the top 30 apparel retailers investing in or deploying RFID, according to Bill Hardgrave, who is now dean of the College of Business at Auburn University.

To help move the adoption of ILRFID forward, in 2010 representatives of retailers, suppliers, technology companies, educators, and industry associations decided to launch the VICS Item-Level RFID Initiative (VILRI). The group describes its purpose this way:

VILRI made amazing progress in a relatively short period of time. By March of 2012, membership had grown to more than 250 individuals and 90 companies. This important, collaborative effort continues today, under the auspices of GS1 US. You can learn more about the initiative, the benefits of item-level RFID, and new developments in the technology at www.vilri.org.

What can ILRFID do for you?
What are the benefits of tagging items with RFID? Some answers to that question were developed in a joint effort among UARK, the VICS Item-Level RFID Initiative committee, and technology providers. It's reprinted here with the kind permission of Retail TouchPoints, an electronic publication for retail executives. It originally appeared in "VICS Item-Level RFID Initiative Leads Charge to Retailing's Future," an article I contributed to the March 2012 issue.

Reducing out-of-stocks—}This is more critical than ever, not only for apparel, but in all product groups. Keeping store inventory to an absolute minimum is critical to success. Sacrificing efficiency in order to meet consumer expectations—and vice versa—is no longer an option. Today's highly competitive marketplace dictates that companies must do both to be successful.

Increasing sales and increasing store and supply chain productivity—This will certainly make a contribution to the retailer's and the supplier's bottom line. Pilots have indicated that having the right mix of products that consumers want to buy will maximize sales, maintain adequate shelf availability, and help keep prices competitive.

Lowering the cost of inventory—Avoiding out-of-stocks and markdowns is very important, but the limited selling seasons of apparel and the frequent reconfiguration of products on the sales floor makes this challenge very difficult without item-level RFID.

Improvement of speed to market—With many products, trends and consumer preferences emerge rapidly, especially in the fashion business. Products have to get from design to the store faster than ever, sell, and be replenished as quickly as possible, while consumer interest is at its highest and before the next selling season begins.

Reducing labor—Retailers routinely slash prices to move merchandise, which results in lower gross margins. Improving store-level inventory accuracy and the number of SKUs that can be read with item-level RFID vs. the bar code reduces labor for taking inventory and searching for products. As source tagging is broadly implemented, retail and manufacturer labor productivity will improve. The supplier will also have fewer order discrepancies to resolve along with order status.

Generating data to maximize programs—The marketplace is rapidly embracing marketing programs with a goal of driving store traffic. Retailers need accurate, real-time information to determine what promotions are working and which ones aren't. Suppliers need this information to determine which marketing campaigns they will continue to fund. Retailer promotional programs represent about 25 percent of sales, so the impact can be substantial.

Preserving brand integrity—Counterfeiting is a significant problem, estimated to be at US $1 trillion worldwide. It erodes brands and margins, and it causes confusion in the marketplace. Consumers don't know if they are buying legitimate, branded products. Item-level RFID, with GS1 US EPC-enabled serialization, is the one approach that can comprehensively address this challenge.

Consumer satisfaction—Harvard Business School research has consistently found that when a product is out of stock, the interested customer is highly unlikely to return to that outlet, and will shop for the product in another store. Research by the University of Arkansas found that the retailers participating in item-level RFID pilots each saw an increase in sales as a result of having product in stock (about 6 percent).

Reducing shrink—The overall average for retail shrink is 1.45 percent of sales. The average in apparel is 1.87 percent, computer and electronics is 0.97 percent, cosmetics and perfume are 1.79 percent, jewelry is 1.06 percent, shoes is 0.85 percent, and vehicle parts is 1.77 percent. There is a major opportunity to reduce shrink, including customer and employee theft, with item-level RFID.

Numerous studies by academic researchers and consulting firms have confirmed and continue to confirm the value of ILRFID. For example, according to Item-Level RFID Tagging and the Intelligent Apparel Supply Chain, a white paper from Motorola Solutions, companies that have implemented item-level tagging are achieving inventory accuracy rates of 98-99.99 percent and have seen sales jump by 4 percent to as much as 21 percent. Another example: Research on EPC-enabled item-level RFID in supplier operations, which was conducted by the University of Arkansas and sponsored by GS1 US and the American Apparel and Footwear Association, found that RFID-based audits experienced an error rate of only 0.01 percent. Manual audits had error rates of between 1 and 5 percent.

Despite all the successes and the good work being done to promote item-level RFID, the retail industry does not seem to be giving as much attention to it as maybe it should.

Where do we go now?
Up until now, the majority of the ILRFID implementations have involved apparel, footwear, and the like. But I think the number of applications will grow.

Grocery, the industry where I spent most of my career, may ultimately benefit from item-level RFID as the technology develops and the pricing of individual items is no longer an impediment. However, there is no reason why item-level RFID can't be used in the logistics side of grocery. Dreamers can make it happen, but keep in mind, it will require collaboration.

Item-level RFID could get its biggest boost from the rapidly growing interest in omnichannel merchandising. There are aspects of omnichannel merchandising that will make it one of the most complex business processes that retail has ever experienced. The only system that can deliver accurate product information is item-level RFID. Unless the retailer makes ILRFID a priority omnichannel will not deliver on its full potential.

Consider the sheer complexity of serving and meeting the expectations of omnichannel consumers. They are encouraged to purchase products online, at the store, via mobile device, or any other method supported by the retailer. They expect to have that product delivered to any destination they request, on time and exactly as ordered. They can buy online and pick up at a store, or buy online and have it delivered to their home or office, and if a product isn't available at the store it will be shipped to the consumer from another store location, a company replenishment point, or a supplier.

Now let's consider the complexity and uncertainty that will be created. Retailers that are involved in omnichannel may use multiple stores as replenishment points in addition to distribution centers. Macy's, for example, was filling orders from nearly 300 of its stores by the middle of 2013, and has said it plans to have as many as 800 stores act as replenishment points, with ILRFID as the critical enabler. Now the question is, how does the retailer plan for replenishment of those stores? That will be a major challenge, as it will be difficult to establish shipment profiles on one hand and in-store demand on the other. Then there is the issue of forecasting for replenishment by the supplier, and whether the supplier will be responsible for shipping to a consolidation center or directly to the stores.

If omnichannel is going to meet consumers' expectations, the entire retail system has to take steps to improve inventory accuracy. The methods used today to take physical inventories are simply not accurate enough to support fulfillment from so many locations. The typical product inventory is 50 percent accurate after six months; accuracy deterioration begins right after a physical inventory is taken, and it continues until the next physical inventory. Inaccurate inventory combined with varying demand in retailers' omnichannel system combines to make accurate forecasting a major challenge for the retailer, which in turn affects the replenishment program on the supplier side of the business. There is no doubt in my mind that item-level RFID is the only technology that is capable of ensuring the level of inventory accuracy that omnichannel commerce demands.

I think ILRFID can be viewed as a "disruptive technology"—one that will fundamentally change the way business is conducted. I hope you will agree that we have made a strong case for item-level RFID as a technology that provides many benefits and is key to gaining visibility of timely and accurate information on the location and quantity of a product, and, in some cases, even the quality of temperature-controlled products.

There continue to be barriers to implementation: tight budgets for spending on technology, the lack of trust between trading partners and solutions vendors, and the inaccurate assumption that only big retail chains can afford and benefit from ILRFID. But I feel confident that this technology will be such a game-changer in terms of cost, accuracy, and efficiency that at some point in the future, adopting item-level RFID will be considered a no-brainer.

Editor's Note:My Incredible Supply Chain Journey ... and What You Can Learn From It will be published as an electronic book (e-book) in early 2014 by the Council of Supply Chain Management Professionals. More information will be available at cscmp.org prior to publication.