Industry conferences have a way of cutting through the noise. After two of the retail world’s most significant gatherings — NRF in the United States and EuroShop in Europe — one theme emerged with particular clarity: the industry’s appetite for new ideas has given way to a harder, more disciplined question. Not what is possible, but what actually works, reliably, across hundreds of stores, day after day.
That shift in conversation points directly at a problem retail has been reluctant to confront head-on. The technology stack has never been more sophisticated. The ambitions have never been more expansive. And yet, at the operational core of the business, the fundamentals keep breaking down in the same old ways — and almost always, the thread leads back to inventory.
The Failure No One Talks About
Consider a scenario that plays out thousands of times every day across retail in the United States. A shopper finds a jacket online — ninety dollars, the right size, available for in-store pickup. The app confirms it. The store system confirms it. The customer makes a twenty-minute drive, walks to the collection point, and is told the item cannot be located.
Somewhere in the gap between a warehouse count, a shop floor transfer, an unprocessed return, or an item quietly misplaced behind something else, the jacket stopped being where the system believed it was. The customer leaves empty-handed. The sale evaporates. A promise made through digital convenience was broken at the moment of physical reality.
On its own, it reads as a minor operational glitch. Multiply it by the scale at which it happens — across thousands of locations, millions of product lines, every single day — and it stops being a glitch. It becomes a structural flaw in the way modern retail functions.
The Store Is Still Carrying Everything
What makes this problem particularly thorny is that it is occurring in a retail environment that looks, on paper, more capable than ever.
U.S. e-commerce continues to expand at a significant pace, with online spending in the fourth quarter of 2025 exceeding $316 billion. Omnichannel capabilities — click-and-collect, ship-from-store, same-day fulfillment — are no longer differentiators; they are the standard expectation. Stores have been redesigned, restaffed, and re-equipped to function simultaneously as sales floors and mini-distribution centers.
And yet, for all of this evolution, one statistic puts the whole picture in perspective: more than 83% of U.S. retail transactions still happen inside physical stores. The store is not a complement to the digital operation. It is the center of gravity around which everything else orbits. It is where online demand and physical experience collide with the messy reality of operational execution — and where inventory accuracy determines whether that collision produces a sale or a breakdown.
When a single store is simultaneously fulfilling online orders, processing returns, restocking shelves, and assisting walk-in shoppers, the margin for error becomes extremely small. A discrepancy that might once have been absorbed quietly now ripples outward — into cancelled orders, disappointed customers, and operational costs that compound quietly over weeks and months.
Spring and Summer: When Problems Take Root
Peak season has a useful characteristic: it exposes problems fast. The pressure is so intense that failures surface immediately, get escalated, and — ideally — get fixed. The spring and summer trading period operates differently, and in many ways more insidiously.
The relative calm of these months doesn’t eliminate inventory problems. It allows them to accumulate undetected. A single size disappears from a key display. A delivery is partially processed and the remainder sits in a stockroom backlog. A returned item spends two weeks in a processing queue before it reappears as available stock. None of these individually triggers an alarm. Together, compounded across an entire store floor over several months, they create a quiet erosion of accuracy that will only become fully visible when the next surge in demand arrives.
The scale of this problem globally is difficult to overstate. Estimates suggest retailers collectively lose in the region of $1.7 trillion annually to inventory distortion — a figure that encompasses both overstock and understock situations, and one that continues to be driven significantly by operational execution gaps rather than purely by external factors.
The Hidden Costs of Being Almost Right
Inventory inaccuracy has a way of generating costs that don’t appear neatly on a single line in any report. They scatter across the business in ways that are often individually explainable but collectively devastating.
A buy-online-pickup-in-store order gets cancelled because the item listed as available cannot actually be located. A customer who drove to collect it doesn’t come back. A shipment gets split because stock levels in the system don’t reflect where inventory physically exists, doubling fulfillment costs and extending delivery times. Store associates spend significant portions of their shifts searching for products that the system says are there but aren’t — time that cannot be recovered. Markdown decisions get made based on inventory counts that misrepresent what is actually on hand, either accelerating discounts unnecessarily or holding prices too long on stock that has effectively already been lost.
With online shopping now representing more than 16% of total U.S. retail sales and growing, these failures are no longer containable in the way they once were. Every fulfillment promise made through a digital channel runs through a store operation that still, in too many cases, cannot verify its own inventory in real time.
Shrink Is Part of the Same Problem
At the same time, retail theft has surged sharply. Data from the National Retail Federation shows shoplifting incidents rising by more than 90% between 2019 and 2023, with associated losses growing by a similar magnitude. The two trends — inventory inaccuracy and theft — are increasingly difficult to disentangle.
When stock data is unreliable, a retailer cannot clearly determine whether a missing item was sold legitimately, placed somewhere incorrectly, or taken. The three outcomes carry entirely different implications for response: replenishment, relocation, or security intervention. Without the ability to distinguish between them in anything close to real time, the default becomes a delayed, generalized response that addresses none of the three causes effectively.
Inventory visibility, seen through this lens, is not just a merchandising concern. It is a loss prevention tool. Accurate, continuous awareness of stock location and status is what makes it possible to respond to anomalies before they compound — and to tell the difference between a gap that needs a restocking order and one that needs a security review.
The Store Has Outgrown Its Own Infrastructure
Underlying all of this is a fundamental structural mismatch that the industry has been slow to address directly. The physical store has been asked to do dramatically more than it was originally conceived to do, in a timeframe that has not allowed the operational model to catch up.
Today’s store is expected to function as a retail floor, an order fulfillment hub, a returns processing center, and in some cases a local distribution node — all within the same footprint, staffed by teams that are simultaneously under greater pressure and operating with fewer resources than a decade ago. The systems supporting these functions — periodic stock counts, manual interventions, partial visibility tools — were designed for a simpler operating environment.
The gap between what stores are now expected to execute and what their underlying infrastructure reliably supports is where the jacket-that-isn’t-there problem lives. And it is a gap that grows wider with every new fulfillment commitment made to a customer online.
Continuous Visibility as the New Standard
At EuroShop in particular, the conversation around technology had a noticeably different quality than in previous years. The focus was less on what AI and automation could theoretically do and more on how these capabilities could be embedded into the daily flow of retail operations in a durable, scalable way — working in the background, always on, rather than requiring constant manual activation.
That framing is directly relevant to the inventory problem. The traditional approach — periodic counting, batch updates, manual correction — produces accuracy as a snapshot. By the time the snapshot is acted upon, it is already outdated. What the most forward-thinking retailers are moving toward is treating inventory as a live, continuously updated data set rather than a periodic report.
Fixed, always-on RFID infrastructure is one of the technologies accelerating this shift. Rather than requiring staff to initiate a count or scan a location, always-on systems passively track item-level movement across the store environment in real time. The result is an accuracy baseline that reflects what is actually happening on the floor at any given moment — not what the system last confirmed during a scheduled check three days ago.
The downstream effects are substantial. Fulfillment reliability improves because the system has genuine confidence in what it is promising. Availability data presented to online shoppers reflects reality rather than an approximation. Loss events become identifiable earlier, when response is still effective. Store associates can be directed toward productive activity rather than spent hunting for items that may or may not be where the system suggests.
What the Next Season Will Reveal
The spring and summer period will not immediately expose the full scale of the inventory accuracy problem for most retailers. It will, however, determine the conditions under which the next peak season arrives. Retailers who address the underlying issue during these quieter months — investing in the infrastructure and processes that produce reliable, continuous stock visibility — will enter the next high-pressure period on stable ground. Those who do not will find, when the volume returns, that the small failures have become large ones.
This is, ultimately, a question about the relationship between promise and delivery. Every digital channel, every omnichannel service, every fulfillment commitment made to a shopper rests on the assumption that what the system shows as available actually is. When that assumption fails, the entire model built on top of it becomes unreliable — regardless of how sophisticated its architecture.
The retailers who will perform best over the next twelve months are not necessarily those with the most advanced technology or the broadest channel presence. They are the ones who have done the disciplined work of making their existing operations genuinely trustworthy — removing the friction, closing the gaps, and building the kind of day-to-day confidence that turns a promise made into a promise kept.
In retail, that reliability is not a given. It is earned, maintained, and lost through the accumulation of a thousand small decisions. And it nearly always comes back to whether the inventory is where you think it is.
Ideas drawn from an article by Peter Oram, CEO of PervasID, originally published on Retail Customer Experience.
