It identifies the item rather than detecting a presence, so the system knows what left and whether it was paid for.

A conventional pedestal produces a binary alarm with no information attached. Staff have to work out what caused it.

An RFID pedestal reads the unique identity of every tag passing and checks it against the sales record.

That removes the largest category of false alarm - the item that was actually purchased - and it tells staff which product to ask about.

It also produces data: what leaves unpaid, when, and in what combinations, which is a loss-prevention analysis conventional EAS cannot support.

The cost is a harder installation, a more expensive tag and a supply chain that has to apply and encode it.

Because the same tags allow a full stock count in a fraction of the time, and stock accuracy drives availability and sales far more than shrinkage does.

A handheld reader counts an area in minutes, so counts can happen weekly rather than annually.

Accurate stock records make click-and-collect and ship-from-store viable, which is why adoption accelerated with omnichannel retail.

Availability improves because gaps on the shelf are found while the stock is still in the back room.

Against that, the exit application alone rarely justifies the tag cost and the programme effort.

Retailers who buy RFID as a loss-prevention system usually under-invest in the inventory processes and get a poor return on both.

UHF readers detect tags well beyond the doorway - on nearby shelves, in a stockroom, or in bags that are not leaving - and filtering those out is the main commissioning task.

The read range that makes RFID useful is also what makes the exit noisy.

Antenna selection and aiming, transmit power control and physical shielding shape the read zone.

Direction detection - working out whether a tag is moving in or out - removes a large share of the remainder.

Software filtering by dwell time, signal strength and read count handles the rest, and needs tuning against real traffic.

Budget commissioning time properly. An RFID exit installed and left at default settings produces alarms nobody trusts within days.

It cannot be deactivated in the way an EAS label can, so the options are removal, a kill command, or marking the item as sold in the system.

Most retail deployments simply record the sale, so the exit reader recognises the tag as legitimate and does not alarm.

The tag remains readable afterwards, which is what raises the privacy question.

Some tag types support a permanent kill command that disables them at the till, at the cost of losing the tag for returns and after-sales.

Removable tags - on a swing ticket or a hard tag - avoid the issue entirely by leaving the tag in the store.

Decide the policy before rollout, because it affects tag choice, till process and the customer-facing disclosure.

It varies by jurisdiction, and the recurring expectations are disclosure, a means of deactivation or removal, and not linking the tag to an identifiable customer.

A tag that remains readable after the item leaves the store can in principle be read elsewhere, which is the substance of the concern.

Several territories expect retailers to inform customers that items carry tags and to offer removal or deactivation.

Linking a tag identity to a named customer - through a loyalty scheme or a card payment - engages data protection law directly in most regimes.

Guidance and law here have developed differently in different regions, so this needs a local answer rather than a global policy copied across.

Document the approach as part of the deployment rather than after a complaint.

It can, and many retailers run both during transition - but only if tagging coverage is genuinely complete.

RFID protects only tagged items, and an item without a tag passes unnoticed.

That makes source tagging effectively a prerequisite: partial coverage produces a system with known blind spots.

Where suppliers cannot all participate, retailers commonly keep conventional pedestals for untagged lines, which means two systems and two till processes.

Some pedestals house both technologies in one housing, which solves the entrance but not the counter.

Plan the transition as a supply-chain programme with a coverage target, not as a hardware replacement.

Reader antennas at the doorway, a reader and network connection, integration with the sales system, and considerably more commissioning than conventional EAS.

Antennas may be in pedestals, overhead, or concealed in the shopfront, and their placement determines the read zone.

The reader needs power and a network path, and the system needs a live link to the sales data to distinguish sold from unsold items.

That integration is routinely underestimated: the exit system has to know about a sale within seconds of it happening.

Commissioning covers antenna aiming, power, filtering and direction logic, and should be revisited after the store has been trading.

Radio regulations apply. UHF bands and permitted powers differ by country, so equipment must be approved for the territory of installation.