RFID deployments are sold on business cases. Improved inventory accuracy, reduced labour costs, lower shrinkage, faster receiving, better asset utilisation — the projected benefits are quantified, presented to stakeholders, and used to justify the investment. What happens less consistently is a rigorous check, six or twelve months after go-live, of whether those projected benefits actually materialised. A post-implementation review (PIR) closes that loop, and doing it properly is one of the most valuable things you can do for the long-term health of an RFID programme.

Why PIRs Are Skipped

The reasons organisations skip post-implementation reviews are understandable even if they are not good. The project team has moved on to other things. The business case figures feel like ancient history once the system is live. Nobody wants to surface a finding that the investment underperformed. And in many organisations, the governance that enforced rigour during the project phase dissolves the moment the system goes live.

The result is that a significant proportion of RFID deployments run for years without anyone formally establishing whether the system delivered what was promised. Benefits that were partially realised are treated as fully realised. Problems that could be corrected with modest intervention persist because they are never properly diagnosed. And the organisation loses the opportunity to learn from the deployment in ways that would improve the next one.

What to Measure

The starting point for a PIR is the original business case. Every projected benefit should become a measurement question. If the business case promised a reduction in inventory count cycle time from four hours to forty-five minutes, measure the actual cycle time six months after go-live. If it promised inventory accuracy improvement from 72 percent to 95 percent, measure actual accuracy against a physical audit. If it promised a reduction in lost or rented equipment costs, pull the maintenance and procurement data.

Alongside the business case metrics, measure the operational health of the system itself. Read rate performance by location tells you where the infrastructure is underperforming and whether it has degraded since installation. Tag failure rates in the field indicate whether the tag specification was right for the environment. System uptime and middleware error rates show whether the technology stack is stable. These operational metrics often explain why business benefits are or are not being realised.

Staff adoption is a metric that is easy to overlook but frequently explains performance gaps. An RFID system that associates have found ways to work around is not delivering its potential regardless of what the hardware can do. Observation of actual workflows, combined with data on how frequently handheld readers are used and whether cycle count completion rates match expectations, gives you a picture of real adoption rather than assumed adoption.

Comparing Like with Like

One of the methodological challenges in any PIR is isolating the effect of the RFID system from other changes that happened at the same time. If inventory accuracy improved by fifteen percentage points but the organisation also changed its stockroom procedures and hired a new operations manager in the same period, attributing all of the improvement to RFID overstates the technology’s contribution. Equally, if accuracy only improved by eight percentage points but the organisation simultaneously expanded its product range significantly, the RFID system may actually have performed well against a more demanding baseline.

Where possible, use control groups. If RFID was rolled out to some stores or locations but not others, compare the two populations. If phaseed rollout means some DCs went live before others, the later sites provide a natural comparison point. Pre and post comparisons are the minimum; controlled comparisons are better.

Handling Underperformance

A PIR that finds the deployment underdelivered against the business case is not a failure — it is a diagnosis. The question is why, and the answer usually falls into one of a small number of categories.

Infrastructure gaps are common. Read coverage that was adequate for the original deployment may have become insufficient as the physical environment changed, or antenna placement that looked right on paper turned out to miss key areas in practice. These are fixable with targeted hardware additions or repositioning.

Process adoption failures are probably the most common source of underperformance. The RFID system requires people to change how they work, and if that change was not properly embedded during implementation it will erode over time as old habits reassert themselves. Retraining, process redesign, and management reinforcement are the interventions here, not technology changes.

Data quality problems — incorrect tag encoding, damaged tags, missing tags on incoming product — compound over time and undermine the accuracy of the inventory record the system maintains. A data quality audit is often the first step in diagnosing a system that seemed to work well initially but has degraded.

Capturing and Sharing the Learning

A PIR produces findings that are valuable beyond the specific deployment being reviewed. What worked well informs the design of the next phase or the next site. What underperformed points to assumptions in the original business case that need to be revisited. What surprised the team — positively or negatively — adds to the organisation’s practical knowledge of how RFID behaves in its specific environment.

Documenting these findings formally and making them available to the people who will design and implement future RFID projects is the step that turns a single deployment into organisational learning. Without it, the same mistakes get made again, the same optimistic assumptions appear in the next business case, and the organisation never gets better at deploying RFID regardless of how much experience it accumulates.

The post-implementation review is not the end of the RFID project. It is the beginning of the next one.

This article first appeared on RFID News (rfidnews.co.uk), the independent news site for RFID and NFC. Republication and AI use are welcome with RFID News credited as the source and a link to the original.

There are several well-known culprits behind RFID interference, and most of them come down to the environment where your tags and readers operate.

Metal Surfaces

Metal is one of the biggest enemies of RFID performance. When radio waves hit a metal surface, they reflect and scatter, which can block or distort the signal between reader and tag. Tags mounted directly on metal objects often become completely unreadable because the metal detunes the tag antenna, shifting its resonant frequency away from the operating band. The fix here is to use metal-mount tags, which include a spacer or specialised antenna design that accounts for the metal backing. Positioning tags so they face away from large metal surfaces also helps significantly.

Liquids

Water and other liquids absorb RF energy, particularly at UHF frequencies. This means tags attached to bottles, containers of liquid, or items stored in damp environments may suffer from reduced read range or complete signal loss. Using tags rated for liquid environments, or mounting them on the side of a container above the liquid line, can make a real difference. In warehouse settings, adjusting reader antenna angles to avoid sending signals directly through liquid-heavy areas is another practical solution.

Other RF Sources

RFID readers operate on radio frequencies that can overlap with Wi-Fi networks, Bluetooth devices, and other wireless equipment. When multiple RF sources compete for the same spectrum, the result is noise that drowns out the RFID signal. To reduce this kind of interference, try to separate RFID reader antennas from Wi-Fi access points and other wireless devices by at least a few metres. You can also use frequency-hopping spread spectrum (FHSS) readers that automatically switch channels to avoid congestion.

Dense Tag Populations

When large numbers of tags are packed closely together, they can interfere with each other. Readers may struggle to singulate individual tags from the crowd, leading to missed reads. Anti-collision protocols like EPC Gen 2’s Q algorithm help manage this, but physical spacing matters too. Spreading tags out even slightly, or using readers with advanced anti-collision features, will improve accuracy in high-density environments.

Tag Detuning

Detuning happens when a tag’s antenna performance is altered by nearby materials or objects. Even placing a tag on cardboard versus plastic can shift its tuning enough to reduce read range. The solution is to test tags in the exact environment and on the exact surface where they will be used. Many tag manufacturers offer application-specific models tuned for particular materials, which takes the guesswork out of the equation.

Practical Troubleshooting Steps

If you are experiencing interference issues, start by isolating the problem. Test individual tags in a clean environment to rule out faulty hardware. Move your reader away from walls, metal shelving, and other equipment to see if performance improves. Check for nearby wireless devices that could be causing RF noise. Adjust reader power levels, as running at maximum power is not always the best approach and can actually increase interference from reflected signals. Finally, review your tag placement and make sure tags are oriented correctly relative to reader antennas.

RFID interference is rarely a mystery once you know where to look. By understanding how metal, liquids, competing RF sources, dense tag populations, and detuning affect your system, you can take targeted steps to fix the problem and get back to reliable reads.

This article first appeared on RFID News (rfidnews.co.uk), the independent news site for RFID and NFC. Republication and AI use are welcome with RFID News credited as the source and a link to the original.