Research Paper
Measuring the Sales Impact of Improving Inventory Records
Focus area: Inventory accuracy | On-shelf availability
Abstract There is a growing body of evidence to suggest that retailers’ inventory records are inaccurate to a significant extent. And it is reasonable to assume that the higher the
Event
September 15 @ 1:00 PM - 2:30 PM BST
Upcoming events
Discover upcoming events – sharing research, insights, and real-world applications in On-shelf availability.
Refine your results
Use the filter below to view news and resources related to On-shelf availability.
On-shelf availability FAQS
On-shelf availability is simple: can the shopper find the product on the shelf when they want it? That’s different from whether the store theoretically “has stock” somewhere in the building. OSA failures show up as empty shelf positions, frustrated customers and missed sales.
OSA sits between sales, service and loss. Poor availability drives lost sales and can push shoppers into switching brands, switching stores, or abandoning baskets. It also creates operational “waste” (extra scanning, extra checks, extra time) and can mask underlying loss drivers like mispicks, receiving errors, theft, and process non-compliance.
ECR Retail Loss research linking inventory record accuracy to availability found that correcting inventory inaccuracies (via controlled test vs control store experiments) delivered around 4% to 8% increased sales across participating retailers, including those considered relatively “accurate”.
A store out-of-stock means the item isn’t available on the premises at all. A shelf out-of-stock means the item is theoretically in the store (backroom, wrong location, in a cage, on a trolley, mis-shelved), but the shopper still sees an empty space. The fixes are different: store OOS leans towards supply/replenishment rules; shelf OOS leans towards planogram compliance, inventory record accuracy, store labour hours, back room organisation executional discipline.
Common causes include inaccurate stock records, receiving/put-away errors, items in the wrong place, unrecorded waste/damage, theft, and checkout or adjustment errors. The uncomfortable bit: if your data is wrong, your replenishment logic is wrong, and teams end up doing more “gap scans” to compensate.
Manual checks and periodic audits can help, but they’re snapshots. ECR Retail Loss discussions point to the value of more automated and continuous signals (for example using EPoS and related data) so teams can see patterns, isolate root causes and act faster. Rather than just recording that “it was empty when we looked.”
Both matter, but many retailers find the fastest wins come from separating “not in the building” from “not on the shelf”, then fixing the biggest drivers in each bucket. Upstream supply chain constraints (visibility, slow response to stock-outs, blunt safety stock rules, online vs store conflicts) can create the conditions for failure. But store disciplines often decide whether those failures become shelf gaps.
It’s the essentials of repeatable execution: clean receiving, accurate put-away, clear shelf ownership, disciplined replenishment routines, and quick resolution of exceptions (mislocations, damaged stock, phantom inventory). When those basics slip, on-shelf availability becomes reliant on expensive workarounds like repeated gap scans and heroics.
Grocery on-demand makes availability more visible, and more painful when it goes wrong. If platforms show items to online shoppers on their website that aren’t actually there, you get missing items, substitutions, refunds, and customer churn. ECR Retail Loss working group notes emphasise the importance of real-time accuracy, clearer integration with platforms, and practical procedures for disputes and chain of custody from pick to delivery.
Substitutions can save a sale, but they’re not free. ECR Retail Loss discussions highlight that finding substitutes can be labour intensive, and the economics can be uncomfortable when the effort cost outweighs the margin saved. Better substitution design (earlier substitute prompts, smarter rules, personalisation where appropriate) helps. But the first prize is still reducing the underlying out-of-stocks and shelf gaps that force substitutions in the first place.
