In the food and grocery sector, a persistent “data accuracy gap” creates significant daily challenges. This gap shows up as shrinking profit margins due to food waste, increased labor strain from manual processes, and rising losses from theft.
With new regulations like FSMA Rule 204 demanding greater traceability, grocers face mounting pressure to gain control over their supply chains. To that end, Impinj recently published its 2025 State of Supply Chain Integrity for Food & Grocery report that highlighted RAIN RFID as a key solution for achieving the real-time, item-level visibility that traditional barcodes and QR codes cannot provide
We spoke with George Dyche, Vice President of Endpoint IC Product Management at Impinj, to discuss the how technology is closing the information gap, it is a crucial component for FSMA 204 compliance and is an essential foundation for advanced applications like AI and machine learning, unlocking a new level of operational intelligence for grocers.
RFIDJournal.com: George, thanks for taking time to talk with us. What is the “data accuracy gap” in grocery supply chains, and how does it show up day-to-day in stores and DCs?
George Dyche: Glad to join you. The gap refers to the lack of insights, visibility and accuracy required to drive confidence in supply chain integrity and use said data to respond quickly to market changes. Impinj’s survey found that this gap is most pronounced within the food and grocery sector— 9 out of 10 respondents noted their organization is equipped to drive accurate supply chain visibility, yet only one-third are consistently driving 360-degree, real-time visibility.
For grocers, this gap manifests in two key areas: workforce availability and food waste. Dependence on traditional barcodes or QR codes leads to challenges such as manual scanning requirements, human error, and label damage— all of which consume valuable labor and create blind spots in inventory tracking. At the same time, unsold perishables that expire on shelves eat into profit margins.
RFIDJournal.com: How does RAIN RFID improve real-time, item-level visibility compared to barcodes and QR codes? What are the typical accuracy and read-rate benchmarks grocery teams should expect?
Dyche: Traditional barcodes and QR codes require manual scanning and a direct line of sight, which limits both accuracy and efficiency. Though barcodes have established a strong baseline for data capture, they can simply fail in complex and high-throughput environments such as food supply chains. Print quality issues, smudging, scanning angle and speed, human error, label placement, and environmental factors such as condensation or frost can exacerbate data accuracy gaps.
In contrast, RAIN RFID endpoint ICs can be attached to food packaging directly and allow employees to count thousands of items in a matter of seconds, followed by an immediate report on inventory status.
In terms of read rates, RAIN RFID can scan up to 1,000 items per second at distances of up to 10 meters, whereas QR codes and barcodes can only read fewer than 25 items per second and up to three meters.
RFIDJournal.com: Which perishables (bakery, meat, deli, dairy, produce) see the fastest ROI from RAIN RFID—and why?
Dyche: Bakery products are often the best candidates for an initial RAIN RFID program. They are typically confined to a single department and a single supply chain, making implementation simpler. At the same time, bakery items are highly perishable, and ensuring shelves are consistently stocked with fresh products is a labor-intensive, manual process. By automating inventory tracking and freshness checks with RAIN RFID, grocers can reduce waste and improve shelf availability.
RFIDJournal.com: What are best practices for launching a pilot—where to start, success metrics, and timeline expectations? What lessons can we learn from bakery rollouts to scale across fresh departments and beyond?
Dyche: A good starting point for launching a pilot is to start in the fresh departments where items already carry in-store labels, such as bakery, deli, and meat. The bakery is an ideal “sweet spot” for RFID because these products already use on-site data or price labels, making the transition to RFID tags straightforward and delivering rapid returns.
However, other departments where items are sold by weight (like deli and meat), are strong opportunities to roll out RAIN RFID without requiring upstream supply chain changes. Applying RAIN RFID in these areas can generate fast benefits, including labor savings, reduced waste, improved inventory tracking, and more associate time for customer engagement. Near real-time item-level data on freshness enables smarter production planning and markdowns, helping to keep shelves stocked with the freshest items.
Generally, starting with bakery and comparable fresh departments offers a low-risk, high-reward pilot that can quickly demonstrate RAIN RFID’s value.
RFIDJournal.com: What measurable impacts can RAIN RFID have on shrink from spoilage and theft?
Dyche: In our survey, 32% of respondents reported the value of their product lost through food waste is $49-150 million annually, and 75% said they lose at least $1 million to waste every year. Shoplifting is another concern for 45% of survey respondents in the food and grocery sector, with 33% citing employee theft as a growing threat.
RFIDJournal.com: What role can RAIN RFID play in loss prevention when paired with gates/cameras or exit analytics?
Dyche: RAIN RFID enables better visibility and data accuracy to help alleviate these losses by improving inventory and expiration management, as well as loss prevention.
RAIN RFID allows item-level tracking through store entries and exits, and it can be layered with other technologies like security cameras and self-checkout machines. As a result, RFID, in combination with other security measures can help pinpoint exactly which items were stolen and by whom.
RFIDJournal.com: In what ways does RAIN RFID reduce labor strain and reallocate associate time to higher-value tasks?
Dyche: RAIN RFID enable automation of time-consuming tasks such as inventory counts, freshness checks, and markdowns. Unlike traditional barcodes and QR codes, which require manual scanning and a direct line of sight, RAIN RFID endpoint ICs can be attached to or embedded in packaging to allow employees to count thousands of items in seconds and instantly receive inventory reports. This automation not only improves accuracy and efficiency but frees associates from repetitive tasks, allowing them to focus on higher-value activities, such as engaging with customers and enhancing the in-store experience.
RFIDJournal.com: How does RAIN RFID data improve demand forecasting, replenishment accuracy, and on-shelf availability?
Dyche: RAIN RFID provides granular inventory data that enhances demand forecasting and replenishment decisions. By automating inventory tracking, organizations gain greater visibility into stock levels, enabling more accurate forecasts, timely replenishment, and improved on-shelf availability.
RFIDJournal.com: How can RAIN RFID support fast, surgical recalls and compliance with FSMA Rule 204 traceability?
Dyche: Data blind spots can exacerbate serious issues like food recalls if employees cannot quickly remove affected products from store shelves. However, when highly accurate data on impacted products is readily available, suppliers can then quickly intercept shipments before they arrive at stores and surgically remove only the affected products while leaving ones verified as “safe” on the shelf.
RFIDJournal.com: How does RAIN RFID set the foundation for AI/ML—what future use cases become possible with clean data?
Dyche: By accurately tracking fresh perishable foods throughout the supply chain, RFID serves as a foundation for innovations like AI/ML. Accurate, near-100% item-level data is essential for these systems, and without it, AI models receive incomplete or inaccurate inputs, and the “garbage in, garbage out” problem becomes particularly critical in the food system. Integrating data from RAIN RFID into AI/ML systems can make predictive inventory management, dynamic pricing, automated freshness monitoring, and a wide range of other advanced use cases possible.
The successful implementation of AI/ML systems depends on strong data and AI/ML technology is only as good as the data it is learning from.