What it costs to do RFID Asset Tracking the right way the first time.
This is the second in a two part blog about RFID for Asset Tracking. Part one of the RFID blog explained a simple five step process for building an RFID Asset Tracking proof of concept (POC). Our goal is to get those "bar code losers" stuck in the 70s into the next century. We want to make you cool again, and keep you from losing assets and failing audits. RFID can deliver and ODIN is the coolest company in RFID so if anyone can give you the celebrity makeover it’s us. We also have a no-charge eBook on RFID Asset Tracking you can download that goes into more detail.
Last time we met, you were building an RFID Asset Tracking POC and gathering data. Now you have to start putting together the RFID business case by getting details of RFID tags, RFID readers, and RFID software. To do that you need some cost data and then you need to know what traps to look out for as you move forward with your POC. Like a lot of new things there are plenty of shysters out there (you may have heard Bernie Madoff started an RFID company from lock-down but that’s never been confirmed).
Detailed Cost of RFID for Asset Tracking
RFID has finally come of age for Asset tracking: RFID tags follow one universal standard (ISO 18000-6), RFID readers are smaller and much cheaper than just a couple of years ago and RFID software or middleware is scalable and easily integrates to systems of record like SAP or Maximo. Now that you kow it works what does RFID cost?
As a general rule of thumb you can use the following costs (assuming you want 99.9% accuracy and you want to know directionality, if tags are moving or stationary, etc):
- RFID hand held readers $3,500
- RFID fixed portals for office & data center environment $9,000 (Includes readers, antennas, housing, mounting and installation at each read point)
- RFID tags for metal mounting on servers, etc. $ 3.00
- RFID operating system $1,500 per device
- RFID asset management/intelligence/location layer $75,000 per site
- Engineering services $75,000 per site
Those are rough numbers for an enterprise class RFID system that will ensure 99%+ read accuracy and easily integrate into an existing system of record - the type of system you can count on for accurate audit results and security. Like any other technology you can get away with something for $10,000 that has a single terminal and a green screen with some ancient RFID readers. Conversely you can spend a couple million bucks to get the ”I want the IBM Commercial where my assets show up in 3D images on the inside of my glasses” system. But a couple hundred grand can solve just about anyone’s asset tracking problems and we have found in more than 500 RFID projects that the ROI tends to be about a year on something like that and audit accuracy increases three to four fold.
This is not your father’s Wal-Mart RFID
The RFID revolution really started in 2003 when Walmart tried to force all their clients to use RFID. It was a disaster. The 800 pound gorilla tried to flex it’s muscle to make their suppliers use a technology they had never even heard of. It also spawned a whole slew of venture capital investment in RFID companies. Well over a billion dollars was invested in RFID companies to please Walmart. When no CPG companies bought those system designed for Walmart warehouses the VC said to their companies “sell that crap to someone or we’ll find someone who will”. So a lot of the warehouse-focused companies starting looking to do asset tracking. That explains the horrif deployments you may have heard about.
Theses first generation companies didn’t listen to clients who wanted RFID asset tracking. They listened to their VC investors. Many current RFID asset tracking solutions look like some high school science project – even one of the largest IT Companies in the world selling routers and servers last year deployed big ugly Walmart antennas bolted to a wall and then sent their engineers around wearing Home Depot Aprons– like they were $15 an hour dock workers. Hardly the cutting edge system more often associated with the company. These are supposedly the smartest IT guys in the world yet they deployed an RFID system meant for Walmart not one meant for a data center or office. They didn't know the real possibilities.
The RFID infrastructure needs the right hardware and the optimal RFID software. The right hardware should be very thin, aesthetically pleasing wings that either mount flush on a wall or inside a wall and make use of 2011 RFID technology. They should be an inch or so thick, fully enclosed and able to tell what direction things are moving without cumbersome GPIO devices like photo eyes or pressure pads. They should mount inside the wall or on the wall and look like they belong in an office.
The right hardware works with any of the top commercially available RFID hardware. It should be software that is designed first and foremost for read performance. You have likely spent millions of dollars on SAP, Maximo, HP Asset manger, Sunflower or some of the purpose-built asset management system – get the most from that system don’t try to replace it. The RFID software you need to increase read accuracy, limit maintenance and management of an RFID network, reduce network traffic, and provide basics like filtering data, smoothing out trends, location awareness, provide business process dependencies between read points (what does it mean if something goes through Read Point A first then Read Point B), tell if an RFID tag is moving or stationary and help facilitate tuning of the system.
Most importantly RFID software or middleware should be stable code that scales with your organizations need. The best place to find good RFID software is by talking to analysts who cover the RFID industry – like Gartner Group or ABI Research.
What does RFID success look like?
Our client Stephanie, we talked about last week, at one of the world's big bank, can now produce an automated report in less than 1/5 the time it used to take to do an inventory – and the RFID inventory has virtually no errors. They are taking the program one step further and working with Dell and Cisco to get their assets RFID tagged at the factory so a box coming in the door gets added to their system before it even gets unpacked. That’s 21st Century Cool. But not all RFID programs went so smoothly.
At least two of the top three Banks in the United States suffered false starts. One, the largest bank in America, spent millions of dollars only to rip it out and start over. Why? The project manager thought he knew enough about RFID to do a system with a start-up software company and have success. They installed a system, not based on physics, but instead based on Microsoft BizTalk first and foremost. On paper the integration looked easy (and it was to their BizTalk server) but the RFID component yielded barely 80% read accuracy. The interesting thing was they thought enough about deploying 21st Century technology and replacing that 40 year old bar code to try again. But you don't need to make the same mistake. Think physics and stable software built for RFID accuracy.
For more detail on how to move forward with RFID asset tracking check out the eBook on RFID Asset Tracking (PDF).
You can prevent false starts and build a system that gets 99% plus accuracy right from the start if you focus on the physics. At ODIN that’s what we live for. We started by hiring the guy who ran MIT’s Auto-ID lab and just got geekier from there. We have PhD’s from MIT, Stanford, GW, you name it - we have the big foreheads who really want to make RFID solutions work. Why do they want to make the solutions work? Because our goal is to Create Supremely Satisfied Clients. That's been one of our core ideologies from the start. It’s what we were built for – getting our customers promoted, and making them cool again.
If you want help with your IT asset tracking program and you think it’s finally time to get rid of the mullet, stop playing Dungeons and Dragons and earn the respect of your 16 year old. Give us a call at ODIN, we’ll take care of you or download the eBook RFID PDF.