RFID Asset Tracking for Business

RFID asset tracking replaces the annual clipboard hunt with radio. Each asset carries a UHF RFID tag with a unique identifier; fixed or handheld readers capture hundreds of tags per minute, without line of sight; and reconciliation software turns those reads into register movements — found, missing, moved — automatically. The result is not just a faster count but a different operating model: one in which the register stays accurate continuously rather than being rebuilt once a year.

CPCON is unusual in owning the whole stack. We have developed and refined our own RFID hardware configurations and reconciliation platform across thousands of inventory and asset projects worldwide over more than three decades, delivered by our own teams. That means your system is designed around your estate and your ERP — not around a single vendor’s product catalogue — and that we can be honest about where RFID earns its keep and where a simpler approach wins.

How passive UHF RFID works

The technology is simpler than it sounds. A passive UHF tag has no battery. When it enters the field of a reader’s antenna, it harvests a little of the transmitted radio energy, powers up its tiny chip, and reflects a signal back carrying its unique identifier — a technique called backscatter. The reader decodes many of these replies every second and applies anti-collision so that hundreds of tags in the same field are read without talking over one another. Because the link is radio, not light, it needs no line of sight: tags are read inside cabinets, through packaging and several metres away.

Three things determine whether a system performs: the tag (and how it is matched to the asset surface), the antenna and reader power, and the environment (metal and liquids absorb and reflect UHF). Get those right and read rates are excellent; get the tag wrong on a metal asset and the same system fails. That is why tag engineering per asset class is the heart of our design work, not an afterthought.

The four parts of an RFID system

We deploy passive UHF to the global EPC Gen2 / ISO 18000-63 standard, on the UHF band allocated for RFID in the UK and Europe (the 865–868 MHz range), using hardware certified for that band so the system is both legal to operate and interoperable across tags and readers. Each tag carries an EPC (Electronic Product Code) in its memory — the unique identifier that the reconciliation platform matches to a line in your register. Most tags also expose a small block of user memory in which we can encode an asset class or site code, useful as a cross-check when a tag is read in isolation.

Passive, active and other locating technologies

“RFID” is often used loosely, so it is worth being precise about what we deploy for asset tracking and why. The workhorse for counting and verifying assets is passive UHF, because it is cheap per tag, battery-free and fast in bulk. Other radio technologies exist for different problems, and we use the comparison below to set expectations honestly rather than let a buzzword drive the design.

For the overwhelming majority of fixed-asset and stock estates, the right answer is passive UHF, sometimes combined with a handful of active tags or beacons for the few assets that genuinely need live location. Designing for the asset population — not the most impressive technology — is how the system stays affordable and actually gets used.

Frequencies and read range

RFID is not one frequency but a family of them, and the band dictates range, read speed and what the technology is good for. The reason UHF dominates asset tracking is simple: it is the only band that combines metres of range with bulk, many-at-once reading. The table below places the bands in context.

Two practical points follow. First, the UHF band differs by region — the UK and Europe use 865–868 MHz, while North America uses a higher slice of spectrum — so hardware must be certified for the band it operates in; the equipment we deploy on UK estates is rated for 865–868 MHz, which is what keeps the system legal and interoperable. Second, the “typical range” figures are properties of the whole system, not the tag alone: the same tag reads farther with a higher-gain antenna and more reader power, and shorter against metal or liquid. That is precisely why we prove range and read rate in a site pilot rather than promising a single headline distance.

Fixed infrastructure vs portable readers

One of the first design decisions on any RFID programme is how much reading should be done by people carrying handhelds and how much by permanent infrastructure mounted in the building. The two are not rivals; they answer different questions, and the skill is in the blend.

In practice almost every estate uses both: handhelds for the broad, periodic counting of the whole population, and a small number of fixed portals at the choke points that matter — the goods-out door through which nothing should leave unrecorded, the boundary between a secure store and the floor, the despatch bay where returnable assets are shipped. Fixed readers can also drive simple physical controls — a light or alarm if a high-value tagged asset approaches an exit it should not — turning the portal from a passive logger into an active loss-prevention measure. We size the number and placement of fixed readers against your actual movement, loss and audit pressures, rather than wiring every doorway by default.

RFID, IoT and sensors

RFID is increasingly a component of a wider Internet-of-Things picture rather than a standalone counting tool. The connection works in two directions, and both rest on the same foundation — the tag is the asset’s identity, and the register is the system of record.

• RFID events as an IoT stream. Every fixed-reader read is itself a timestamped event — “asset X passed portal Y at time Z” — and a continuous stream of those events is exactly the kind of data an IoT platform consumes. Fed into dashboards and alerting, it turns asset movement into something you can monitor and react to, not just reconstruct after the fact.
• Sensors attached to identity. Sensor-enabled and battery-assisted tags can carry more than an identifier — temperature for cold-chain or climate-sensitive equipment, shock and tilt for assets that must not be dropped, light or tamper status for security. The RFID read binds that sensor data to a specific, register-matched asset, so a temperature excursion or a shock event is attributed to a known item rather than an anonymous reading.
• One identity, many systems. Because the EPC is the anchor, the same tag can serve the asset register, a maintenance system, a security platform and an environmental monitor at once — each subscribing to the events it cares about — without any of them becoming a rival source of truth.

We deliberately keep this layered. The RFID and sensor infrastructure captures and attributes events; the reconciliation platform turns them into clean movements and exceptions; and your asset system remains the single record everyone reports from. That discipline is what stops an IoT deployment fragmenting into a dozen disconnected dashboards.

Tag security and privacy

Because RFID is read by radio, at a distance and without line of sight, organisations reasonably ask what that means for security and privacy. For fixed-asset tracking the exposure is modest, but it is worth designing for deliberately rather than assuming it away.

• Least data on the tag. A passive UHF tag normally carries only an opaque identifier — not asset descriptions, values or any personal data. The identifier means nothing without your register, which stays behind your own access controls, so a stray read reveals almost nothing of use to an outsider.
• Access and kill passwords. EPC Gen2 supports access and kill passwords and lockable memory. Memory can be locked so the tag cannot be trivially rewritten or cloned, and a tag on an asset leaving your control — sold, disposed of, returned to a lessor — can be permanently disabled (“killed”) so it cannot be read again.
• Read-zone control. The same antenna design and read filtering that keep counts accurate also limit where tags can be read from, reducing the scope for casual reading of tagged assets from outside a controlled area.
• People and consumer-goods concerns rarely apply. The privacy debates around RFID generally concern tagging individuals or retail items a person carries away; tagging plant, IT and equipment that stays within your estate does not raise the same questions. We nonetheless design to the principle of least data on the tag as a matter of course.

From days to hours: a worked example

The efficiency case is easiest to grasp as a before-and-after. Picture a distribution operation that had always counted its fixed assets and high-value stock wall-to-wall once a year. The exercise tied up several staff for the best part of a working week: aisles and mezzanines walked methodically, asset numbers read off labels and typed into a spreadsheet, racking and cabinets opened to find hidden items, and then days more spent reconciling the count against the register and chasing the discrepancies the manual process inevitably introduced. The whole thing was disruptive enough that it could only be faced annually — which meant the register was demonstrably accurate on exactly one day a year and drifted for the other 364.

With the estate RFID-tagged and the reconciliation platform in place, the same count changed shape entirely. One auditor walked each zone with a handheld held at a steady sweep, the screen filling with a live tally as hundreds of tags answered through boxes, racking and cabinet doors that previously had to be opened by hand. Areas that had taken a day were read in under an hour; the full site, once a multi-day operation, came down to a single shift. The reconciliation that used to take further days was near-instant, because the platform matched the captured EPCs to register lines automatically and presented only the genuine exceptions — found, missing and moved — for a person to judge.

The deeper change was not the one-off speed-up but what it made possible. Because a count now cost hours rather than days, the annual ordeal could be replaced with a rolling cycle count: high-value and mobile classes swept weekly or monthly, the rest on a sensible rotation, so the register was never more than a short cycle away from the truth. The year-end position stopped being a scramble and became a report — and the labour that used to be consumed by counting was freed for work that actually needed a person.

RFID vs barcode: the honest comparison

Barcodes are not obsolete — for small, stable estates they are often the right answer. The decision turns on volume, movement and how often you need to count. The table below is the comparison we walk clients through in the design phase.

The headline advantages of RFID — no line of sight and bulk, many-at-once reading — are exactly what collapse counting time and make continuous counting affordable. Both technologies share the same foundation, though: durable tagging and a disciplined asset number. RFID is barcode tagging done with radio.

What an RFID programme delivers

Where RFID asset tracking is used

Real-time tracking vs periodic counts

The same tags support two operating models, and most estates blend them:

• Periodic and cycle counting (handheld). Audit teams sweep a zone with a handheld reader and capture every tag in range in seconds. Because it is so fast, the once-a-year count can become a rolling cycle count of high-value and mobile classes, keeping the register accurate all year rather than at a single point.
• Event-level movement tracking (fixed). Fixed readers and antennas at doorways, despatch bays and key thresholds record the moment a tagged asset passes — near-real-time visibility for the high-value or high-risk items where knowing when something moved justifies the infrastructure.

The art is proportion: handhelds everywhere, fixed portals only where the value of real-time movement data outweighs the cost. We size that balance in the design phase against your actual loss, movement and audit pressures.

How we implement it

1. Site survey and pilot. We test tag types and reader placement against your real materials and layouts, and prove read rates before any rollout decision — no estate is committed on a brochure figure.
2. Baseline verification and tagging. Assets are physically verified and RFID-tagged, and the register is cleaned so the system starts from truth rather than inheriting old errors.
3. Deployment. Handheld readers for audit teams, fixed portals for doorways and despatch where movement tracking matters, and the reconciliation platform configured to your locations and asset classes.
4. Handover or managed service. Your teams run counts in-house, or CPCON operates recurring counts as a managed service with SLA-backed accuracy reporting.

What an RFID count actually looks like

It helps to picture the day-to-day reality rather than the technology. Take a floor of an office or a bay of a warehouse that previously took a two-person team most of a day to count by reading and typing asset numbers from a clipboard. With a tagged estate and a handheld reader, one person walks the zone holding the reader at a comfortable sweep, the screen filling with a live tally as tags answer. A floor that took a day is read in well under an hour.

When the sweep is done, the handheld holds a list of every EPC it saw. The reconciliation platform compares that list to the register lines expected in that location and resolves three outcomes automatically:

• Found. The tag was expected here and was read — confirmed in place, verification date updated.
• Missing. A register line was expected but no tag answered — flagged for investigation; persistent misses become impairment or disposal candidates.
• Moved or unexpected. A tag was read that the register placed elsewhere — the asset is relocated and its location corrected, or it is an untagged-then-found item that needs adding.

Because the cost of a count has collapsed, the once-a-year event becomes a routine. High-value and mobile classes can be swept weekly or monthly as a rolling cycle count, so the register is never more than a short cycle away from the truth, and the year-end position is a report rather than a scramble.

Data flow: from radio to system of record

The thing that turns RFID from a gadget into an asset-control system is the software layer that sits between the readers and your ERP. Raw reads are noisy — the same tag is seen dozens of times in a sweep, tags from the next zone occasionally intrude, and a fixed portal sees an asset every time anyone walks it through a doorway. Our reconciliation platform turns that noise into clean, attributed events.

Crucially, your ERP stays the system of record. The RFID layer does not become a second, competing database of asset truth; it feeds the one you already run, which is what keeps finance, audit and operations working from a single set of numbers.

Governance: why 99% is a people problem, not a radio problem

Well-designed UHF estates routinely read 99%+ of the tags that are present. The gap that stops an estate reaching 100% is almost never the radio — it is governance. New assets arrive and are put into use before anyone tags them; assets are disposed of without the paperwork; items move between sites with no transfer record. RFID reads what is tagged and recorded; it cannot read an addition nobody tagged or know about a disposal nobody logged.

That is why we never sell RFID hardware in isolation. A rollout is paired with a baseline verification and register clean-up so the system starts from truth, and with simple operating procedures — tag-on-receipt, log-on-disposal, record-on-transfer — so it stays there. The technology removes the cost of counting; the governance removes the cause of error. You need both.

Why RFID pilots fail — and how we avoid it

A surprising number of RFID projects stall at the pilot, and the reasons are consistent. We design around them deliberately:

• Wrong tag on metal. Generic labels detuned on steel produce poor reads and a verdict that “RFID does not work here”. We specify and test on-metal tags per asset class before any volume is committed.
• No baseline. Switching on RFID over a dirty register just measures the mess faster. We verify and clean first, so the pilot proves read rates against known truth.
• Over-reach. Trying to track every low-value item in real time inflates cost and complexity. We scope RFID to the classes where it pays and leave the rest on barcode.
• No governance. Without tag-on-receipt and log-on-disposal, accuracy decays within months. We build the procedures in from day one.

The limitations — stated plainly

We would rather you knew the constraints than discovered them after rollout:

• Metal and liquids. Both absorb and reflect UHF. On-metal tags and considered placement solve it, but the wrong tag on a steel asset simply will not read.
• Dense stacks. Tightly packed items can shadow inner tags; antenna design, read angles and movement during the sweep mitigate it.
• Stray reads. A reader can pick up tags from an adjacent zone; antenna power, orientation and read filtering keep counts attributed to the right location.
• Cost at small scale. Tags and readers cost more than barcodes. For small, low-movement estates a barcode regime can be the better economic choice — and we will say so.

The return on investment

The dominant return is labour. A wall-to-wall count that consumed team-days or team-weeks becomes a walk-through of hours, repeatable as often as you need it — which is what makes continuous accuracy affordable. On top of that sit reduced loss and shrinkage, fewer duplicate purchases of assets that were simply lost on paper, lower over-insurance once values are based on what genuinely exists, and faster, cheaper external audits because every figure is backed by a recent, scan-based count. The investment is the tags, readers and rollout; the payback is the gap between an annual disruption and a register that stays right.

Measuring success: what we report

“Better visibility” is not measurable; read accuracy and count time are. Whether you run the system in-house or take it as a managed service, we hold the programme to figures you can see:

• Read accuracy. The proportion of present tags read in a controlled sample — the headline measure of whether the tag-and-reader design is working, tracked by location and asset class.
• Register accuracy. The proportion of register lines confirmed in their expected location after a count — the measure finance and audit actually care about.
• Count time. Elapsed time and labour per zone, before and after, evidencing the efficiency case rather than asserting it.
• Exception ageing. How quickly missing, moved and unrecorded items are resolved — the governance health of the estate over time.

As a managed service, those measures sit behind an SLA, so accuracy is a commitment with evidence, not a hope.

Assets and stock on one infrastructure

The same readers, tags and software that track fixed assets also count stock, work-in-progress, tools and returnable containers — the radio does not care whether an item is capitalised or held for sale. That means a single RFID investment can serve both finance and operations: a fixed asset verification one week, a wall-to-wall stock count the next, tool control in the workshop in between. Designing the system with both jobs in mind from the start is usually cheaper than buying twice, and it concentrates the counting skill and the data in one place.

RFID is the means, not the end

RFID is a means, not the end: the end is a fixed asset register and stock records that stay accurate between audits — supporting FRS 102 reporting, statutory stocktaking records, the IT asset inventory your security assessments sample, and insurance values you can defend in a fixed asset verification. If a simpler barcode regime fits your estate better, we will tell you so in the design phase rather than sell you radio you do not need.

Getting ready for RFID: what to expect

Organisations weighing an RFID programme usually ask the same practical questions. The honest answers shape a realistic plan:

• “Do we have to tag everything at once?” No. Rollouts are normally phased by site or asset class, starting where the case is strongest — the high-value, high-movement classes — and extending as the model proves itself.
• “Will it disrupt operations?” Tagging is done around your working day, and because counts are so fast once live, the recurring disruption of the old annual count actually falls.
• “What about the assets already barcoded?” They can stay on barcode, be over-tagged with RFID where the case justifies it, or migrate class by class. The asset number and the register discipline carry across either way.
• “Who runs it afterwards?” Your teams, with the platform and training handed over; or CPCON, as a managed service with SLA-backed accuracy. Many estates start managed and move in-house as confidence grows.
• “What if RFID is overkill for part of our estate?” Then we say so. The design splits the estate into the classes that justify radio and the classes that do not, rather than forcing one technology across everything.

The common thread is proportion. RFID is powerful where volume, movement and count frequency make it pay, and unnecessary where they do not — and a good design is the one that draws that line correctly for your estate.

Why CPCON