Nearly six years ago Walmart sent 500 shelf-scanning robots home after concluding that workers picking online orders could see the
Conversations On Retail
July 20, 2026
When radio-frequency identification first emerged in retail, it promised faster and more reliable product identification than barcodes. Walmart’s early pilot programs in the mid-2000s drew industry attention, but widespread adoption lagged.
The technology’s shortcomings were not in the tags themselves but in how they were used. Associates had to manually scan shelves with handheld readers, which was repetitive, time-consuming, and prone to error. Coverage was incomplete, with items easily missed in crowded racks or low-signal areas. Because these scans often happened only once a week or even less, the resulting data was stale and inconsistent.
Without dependable reads, retailers could not use RFID to automate replenishment, support omnichannel fulfillment, or detect shrink in a timely way. Costs for tagging merchandise also remained a barrier, particularly in large-format stores where millions of units needed tagging.
The second act for RFID is unfolding as retailers combine tagging with autonomous systems. Rather than relying on staff to perform scans, robots, drones, and smart carts now handle the work automatically.
Mobile robots can travel store aisles daily, scanning thousands of tags without disrupting shoppers. Decathlon’s use of Simbe Robotics’ Tally system is one example, delivering consistent, automated reads that staff can act on quickly.
Warehouse drones are being deployed by companies such as Maersk and On Running, where they scan overhead racks at speeds of up to 1,000 items per second with accuracy rates above 99 percent.
Hybrid systems that combine fixed RFID readers, handheld devices, and autonomous units are emerging as best practice. According to PAL Robotics, layered deployments give retailers both coverage and flexibility across different store formats.
Because the systems run in the background, they remove labor bottlenecks and keep data fresh. What once required hours of manual work can now be updated in near real time.
Reliable, frequent reads unlock a wide range of operational gains.
Several high-profile moves suggest that RFID’s new phase is scaling.
Walmart has expanded RFID tagging requirements for suppliers beyond apparel into categories such as home goods and consumer electronics, signaling that it expects item-level visibility to become standard.
Old Navy is introducing RADAR, a system that blends RFID, AI, and computer vision across more than 1,200 stores, with the goal of reducing shrink and improving replenishment speed.
Several supermarkets are exploring RFID to monitor perishables and high-theft items, areas where tagging was once considered impractical due to cost but is now seen as feasible.
Although the potential is clear, retailers should approach autonomous RFID carefully.
Integration is complex. Data pipelines, APIs, and system alignment are essential to avoid turning accurate reads into unusable noise.
Piloting in select stores before scaling helps avoid costly missteps.
Tagging strategy matters. Placement, durability, and security of tags influence read accuracy and shrink protection.
Autonomous systems require maintenance, calibration, and oversight. Treating them as mission-critical infrastructure is key to sustaining ROI.
RFID has been “the technology of the future” for two decades, but only now is it achieving widespread, practical impact. By combining item-level tagging with autonomous systems, retailers are finally overcoming the labor and accuracy challenges that kept adoption limited.
What emerges is not just better counting but a live, actionable view of store operations. Accurate replenishment, trusted omnichannel fulfillment, and proactive shrink management are no longer theoretical benefits—they are measurable outcomes.
Autonomous RFID is not a brand-new technology, but it represents a new era of reliability. For retailers, it may prove to be one of the most important operational upgrades of the coming decade.