In 2003, Walmart told its top 100 suppliers to put RFID tags on every pallet shipped to its distribution centres. The mandate was supposed to start a revolution. Instead, it became a cautionary tale. The technology worked, but the cost was too high, the standards too fragmented, and the business case too narrow. Within a few years, RFID was quietly moved from the “disruptive” column to the “niche” column, useful for pharmaceutical serialisation and airline baggage tracking but not much else.

That was then. RFID is making a comeback, and this time the difference is not the technology. It is the world that has changed.

1. RFID Feeds AI: But Only If the Data Is Real-Time

AI-driven demand forecasting is only as good as the data it trains on. Most supply chain data is backwards looking. A weekly inventory snapshot tells you what the stock position was three days ago. A daily update tells you what it was yesterday morning. Neither gives an AI model what it actually needs: current, item-level visibility across every location in the network.

RFID changes that. When every item carries a tag and every exit point has a reader, the data stream is continuous. The AI sees a product move from the back room to the shelf, from the shelf to the checkout, from the checkout to the returns bin, and it adjusts its forecast accordingly in near real time. Companies using RFID-enabled inventory data have improved forecast accuracy by 15 to 30 percent compared to barcode-only operations. The difference is not the algorithm. It is the data velocity.

For supply chains adopting “China Plus One” diversification strategies, real-time RFID data becomes even more critical. A multi-source strategy creates exponentially more nodes to track. Each new source is a new inventory pool, a new transit leg, a new handoff point. RFID provides the data layer that makes diversification manageable rather than merely aspirational.

2. Omnichannel Retail Cannot Afford to Guess

Barcodes tell you what entered the building. They do not tell you what is still there. An omnichannel retailer processing orders from stores, warehouses, and drop-ship locations simultaneously needs to know whether a specific item exists anywhere in the network. Barcode scans only capture snapshots at receiving and shipping. Everything in between is a blind spot.

That blind spot is expensive. The National Retail Federation estimates that shrinkage costs U.S. retailers $112 billion annually. RFID cuts that number dramatically. Retailers who have deployed item-level RFID in omnichannel operations report shrink reduction of 40 to 60 percent. When you can see every item at every point in its journey, the gap between what your system says you have and what you actually have narrows to near zero.

3. The Cost Curve Has Finally Crossed the Threshold

When Walmart issued its 2003 mandate, a single passive UHF tag cost roughly 50 cents. For a retailer processing billions of items, that made item-level tagging financially impossible. Today, passive UHF RFID tags cost between 3 and 7 cents in volume. Readers have dropped from thousands of dollars to hundreds. The total cost of a deployment has declined by roughly 70 percent over the past decade. Meanwhile, labour costs have risen. A warehouse worker scanning barcodes costs $18 to $25 per hour. The combination of cheaper tags and more expensive people has crossed the economic threshold that killed the first wave.

4. RFID Is Becoming the Data Layer for Autonomous Operations

As warehouses deploy autonomous mobile robots and automated sortation systems, they need a data layer that works without human intervention. Barcodes require a person to present the label to a scanner. RFID does not. A robot driving through an aisle can read every tag within its range in under a second, without stopping and without line of sight.

This is the difference between automation that operates at human speed and automation that can run at machine speed. A robot with RFID does not just know where the bin is. It knows what is inside the bin. It can verify contents without opening. It can identify misplaced items without human review. It can perform cycle counts continuously, not weekly, because there is no scanning labour cost to count.

The Difference Between Watching and Knowing

Barcodes give you a transaction record. They tell you what happened at the point of scan. RFID gives you a state of being. It tells you what is happening right now. The difference is not incremental. It is the difference between a history lesson and a live feed.

For supply chains operating in a nonlinear, accelerated, volatile, and interconnected world, a live feed is no longer a luxury. Companies that know their inventory position in real time will respond to disruptions faster, serve omnichannel customers more reliably, feed their AI models with fresher data, and execute multi-source strategies with confidence. Companies that still rely on yesterday’s snapshot will be making decisions about a world that no longer exists.

RFID’s comeback is not about the technology maturing, though it has. It is about the world finally catching up to what the technology always promised. The question is no longer whether RFID works. It is whether your supply chain can still afford to work without it.