RFID About Systems, Not Tags
By George Anderson
Radio frequency identification (RFID) technology remains the hot topic in retail technology circles due in large part to Wal-Mart’s mandate that suppliers be RFID compliant by 2006.
Some, however, have questioned if the current costs of getting RFID programs up and running outweigh the benefits of using the technology. Others have identified technical hurdles RFID needs to overcome to be of value across the tracking board.
For most involved in the implementation of RFID, it’s not a question of whether retailers and their suppliers should pursue the use of the technology but what is the most prudent way to go about it.
Robert Malone, writing on the Forbes’ Web site, concludes, “The question is not do I or do I not tag but where, when and how much.”
According to Mr. Malone, “It is best to start with a pilot program of enthusiasts rather than committing to the technology across the board. Plan the application and understand the system, get the right suppliers, and test for quality, cost control, etc.”
He also points out that current and upcoming technologies make it clear that it is probably not wise for retailers and suppliers to put all their eggs in the RFID basket.
“Bar code is far from dead,” writes Mr. Malone, “and its broadly based use and the investment it represents are not about to fade from the scene quickly. RFID may be more robust in the sense of being able to deliver more information and without the need for proximity and clear sight lines as with bar code, but the devil can often be in the details. The devil can also be in an unwieldy system. Currently, the RFID system may err in the direction of complexity as processing the huge amount of information a set of tags can send out requires information integration of a high order and some kind of data warehousing to make sense. These require investments of consequence.”
As for up and coming technologies that some may see as competitive to RFID, Mr. Malone points to the potential use of “smart dust or microelectromechanical sensors (MEMS) made by companies like Dust.”
The objective of the technology is to create communicating sensors “the size of a grain of sand,” writes the author, adding, “Don’t laugh – much progress is being made even if you can’t see it.”
Ultimately, however, Mr. Malone says the benefits of RFID established in quantifiable studies means companies will need to dedicate human and financial resources to the technology.
He adds a caution and advice: “The standards for RFID are well along, but they are not cut in brass and therefore the smart money would bet on a limited investment and most of
that in research into the technology, the code standards, the vendors offering tools or services, and planning, planning and planning. Did we suggest planning?”
Moderator’s Comment: What is the current state of RFID in retailing? Where is it headed?
– George Anderson – Moderator
RFID is headed for adoption. Retailing is a very limited portion of the equation and the wants and need of retailers really don’t matter in the entire scheme of things (an area to which retailers don’t often pay enough attention. For instance:
“NATO Approves RFID Agreement (Today)
“The North Atlantic Treaty Organization (NATO) has approved Standardization Agreement (STANAG) 2233 for RFID.
“The draft agreement for CONSIGNMENT AND ASSET TRACKING BY RADIO-FREQUENCY IDENTIFICATION (NATO STANAG 2233) has received the votes needed for ratification as a STANAG. Formal promulgation of the STANAG will follow later this year. The STANAG is an agreement between nations to use a standardized RFID for tracking of consignments and assets between nations. The text of the agreement is not available for publication.
“STANAG 2233 references existing ISO standards for both active and passive RFID utilizing the infrastructures of the nations involved. Changes to the first edition already in progress reflect the rapidly evolving RFID market, and the applicable ISO standards. This STANAG applies only to NATO operations while encouraging use of the STANAG internally by the ratifying and implementing nations.”
The above will be implemented to harden supply chains against subversion of arms and materials that could be used to inflict pain on others. Additionally it will promote standardization, decrease inventory investment, etc. Don’t you sometimes wish you had an RFID device when you misplace your glasses or the TV remote control or the telephone?
LucB nails it. Retail RFID holds potential to eliminate the supply-chain “blindspot” that lies between the loading dock and the POS terminal. (Word on the street says this is the primary motivation behind Wal-Mart’s interest in this technology.)
Retailers need better operational analytics so their store and regional managers can regain a fighting chance as merchants. RFID could provide a real-time information flow in support of this goal, by feeding in to the local information loop that would allow store managers to efficiently stay in-stock, on-price, and on-program.
Not only would headquarters gain a clearer picture of store conditions, but managers would gain an edge on maintaining them.
For all the hype in the press about RFID tags, the reality is that they should be tested by willing participants, not forced by a mandate. There are still too many technical problems that need to be resolved – tag read reliability for all package types including glass and metal, having the systems and processes to handle the huge amount of data that will be generated, and then being able to analyze the data that is captured to make decisions.
Until the problems are solved, it should only be in test by those willing to absorb the cost.
I think we are looking at the sizzle, what may come from RFID at this point and proceeding like those benefits already exist.
One of the great potential applications for RFID hasn’t been mentioned here so far.
The ability to read the presence of individually identifiable items–on the floor, in the stockroom, or in the warehouse–will very nearly eliminate error in merchandising data.
In the past, a merchant was able to verify that 100 units of a SKU arrived at the shipping dock. From that point forward, stock on hand was determined in one of two ways: 1) subtracting sales from the number of units received, or 2) a physical count of merchandise. RFID allows the data providers within a retailing organization to verify–in real time–that the number of expected units on hand can be accounted for and located.
As a data collection technology, RFID has many inherent advantages that I believe does mean that some day it’s just the way virtually all things will be tracked (including some of the offshoots of RFID mentioned above).
However, even the questions being asked here have it wrong. A technology should not be pursued for the technology’s sake. Far too often, it is the RFID cart chasing the business problem horse, in part because individuals are making careers, for now, as RFID gurus/champions within corporations.
The reality is that, today, many of the problems being chased with RFID could be solved much less expensively with existing technologies. Notice also that the issues and benefits targeted for RFID keep changing, including at Wal-Mart.
Of course it makes sense to get to understand this powerful new technology, and in the course of doing that see whether it allows problems to be solved in a new or more cost effective way than other data collection approaches. But if people would just remember that RFID is just a data collection tool, among many others, rather than some huge force on its own, we would all be much better off.
The business applications to really use the data are years away.
LPs, 45s, 8 Track, cassette, LaserDisk (I actually bought one), videotape, digital. Hard drives, floppies, smaller floppies, ZipDisk, CD, DVD, flash memory. CRT, flat screen, HD, plasma, LCD. I’ve left out many other communication developments, but you get the point.
Media, technology, and formats for consumer electronics constantly evolve, and are acceptably priced because consumers embrace them for their benefits. In contrast, retailer technology expects to take a step and rest; take a step and rest; take a step and rest. Like a slinky. Why this difference?
In consumer electronics, a variety of choices simultaneously exist for recording, reading, storage, and playback. Movies are made available in both DVD and VHS. Music via CD and MP3. Software via download and CD. Portable digital storage via hard drive, flash, CD, and DVD. And consumers don’t mind. They simply purchase and use what they prefer and can afford. Whatever fits their needs. Then, newer stuff becomes cheaper and more generally accepted while older stuff drops off the rear of the evolutionary choo choo.
Perhaps product manufacturers can provide multiple retail reader options simultaneously, giving retailers the option of how to capture information based on their preferences and what they can afford. In other words, why can’t products ship with multiple communication technologies like UPC, RFID, and “Dust?” Newtech will grow cheaper and be embraced more widely, while oldtech falls off the back of the train.
If you really want to think about RFID as a technology, it is tags and sensors. The whole industry seems to equate RFID with tagging products – in the warehouse – a very worthy application. We have been tagging shoppers on the sales floor for four years. I don’t think it is too much of a stretch to point out that, once again, many retailers are more obsessed with their products than they are with their customers. This raises serious questions as to whether people even know what “customer-centric” is. You can chalk me up as a skeptic about the retail industry being customer centric.
The question (to paraphrase JFK) is not what retailers have done for RFID, but what can RFID do for them? The jury is still out – way out. The price is still high – way high. And…as one of my colleagues recently pointed out, Wal-Mart’s mandate has slipped a year.
RFID has to be looked at with much more specificity than it has in the past. Which retail and CP segments will benefit? At what level (pallet, case, item, component)? Active or passive tags? When is the price of the reader a gating factor, and when is the price of the chip a gating factor?
Finally, what are the business benefits of each application? These questions are overlooked under the umbrella of “RFID”.
There are really two very distinct aspects to RFID. First is the obvious ease of data collection created by automatic capture of identification (license plate) information for objects passing reader stations. By itself, this is going to generate a lot of data at the “edge of the network” that must be filtered and presented logically to the business applications. This whole process is where the new technology is involved and is where all the wrinkles have to be ironed out.
The second area is in the meaningful application of the new data within the business applications. A better terminology for this aspect is “serialization.” Whether it is done by bar codes, RFID, or some new technology, serialization refers to the tracking of individual units. It means FMCG retailers are all of a sudden going to be able to track individual units much as automobile or appliance retailers track serial numbers on cars and refrigerators. The fundamental question is “just because we can, does that mean we should?” Tracking the characteristics of a particular case means a data base somewhere which can be accessed by the applications. These characteristics can drive decisions on markdowns for dated products or unpopular flavors or colors. As George mentions, when combined with sensor data they can be used to detect mishandling. The challenge is that all the business applications have to become sensitive to this new data. Another challenge is that there will be a long transition phase during which both serialized and un-serialized containers will be handled. The applications will have to support business processes based on both circumstances while the supply chain gets rid of older containers.
Much discussion has occurred around the first aspect of RFID. This is where all the testing and new hardware are involved. It is the second aspect that will provide the payback and also involves the most preparation. Both applications and business processes have to be modified to take advantage of serialization. A company that still considers RFID “too risky” for their current involvement should nonetheless begin considering what serialization means to the way they do business.
The great leap in product code identification was the UPC (EAN in other countries) bar code. I have a feeling that RFID’s ROI is a fraction of the value of the UPC.
It appears that tech improvements do not all have the same quantum leap ROI. For example, the first widely used spread sheet pc program was Visicalc. Then came Lotus 123. Then came Excel. Lotus and Excel did not have the same degree of improvement that Visicalc did. The latter 2 programs were incremental improvements not quantum leap improvements.
Some say that the biggest potential ROI for RFID comes when the retailer knows the item is in the store’s stockroom or warehouse but not out on the shelf. But that application doesn’t seem to be the one getting the most focus.
RFID adoption will be speedy when the ROI is proven excellent. So far, the jury has not seen this evidence. They’ve heard a lot of theory, but they need some proof.
Individual item level information would be a boon. No more OOS driven by bad perpetual inventory data. No more under-ring theft, immediate total checkout, reduced total inventory levels because you don’t have to allow for bad data.
It all sounds great. The problem is the investment in the technology to equip the shelves/registers/backrooms to read the tags is ridiculous vs. the benefits, the reliability of all the equipment in a real retail environment does not exist, interference by other RF sources has not been overcome, and we can’t even get all the cases on a pallet to read 100% of the time going through a “portal” specially designed to do that.
When the technology works and the cost benefit ratio works, people will adopt it readily. Today RFID at retail is mostly flash with very little substance.
The biggest piece of this RFID puzzle has yet to be conquered, and that is the exorbitant cost of both tags and readers (portals). During a meeting I attended in the fall of 2004, one of the pioneers of this technology admitted that even the most optimistic analysts say that the cost per unit will never fall below a nickel each. How do you justify this investment on consumer goods that may only cost $0.15 to $1.00 each? It may make sense someday for more expensive consumer goods, but never for the everyday cheap items that are sold worldwide.