CPG Terms Explained, a series by Cyril Ovely

What Is On Shelf Availability (OSA)? The Retail Metric That Determines Whether Shoppers Can Actually Buy Your Product

On Shelf Availability (OSA) measures whether a product is physically present on the retail shelf and available for a shopper to purchase at the time of their visit. It is the positive framing of Out of Stock, and the metric retailers and brands use to measure shelf execution quality.

The short answer

On Shelf Availability (OSA) answers one deceptively simple question: when a shopper walks down the aisle, is the product they want actually there? Not in the warehouse, not in the backroom, but on the shelf where a customer can pick it up.

OSA is expressed as a percentage. A key SKU at 96% OSA means the product was found on the shelf 96% of the time across measurement periods. The industry target for key SKUs is usually 95% or higher. Anything below that signals execution problems in the supply chain or the store.

Think of OSA as the positive mirror of Out of Stock (OOS). Where OOS tells you what went wrong, OSA tells you how often things went right.

Why it matters in CPG

A product not on the shelf cannot be sold. Industry studies consistently show that 8 to 12% of SKUs shoppers want are unavailable when they visit the store. For promotional items, the number can be even worse.

The consequences cascade. A shopper who cannot find their preferred brand will buy a competitor's product, and research shows a significant portion will switch permanently. You are not just losing one sale; you are losing a customer.

The business decisions OSA informs:

  • Sales effectiveness: "We negotiated the listing, but is the product actually available for purchase?"
  • Supply chain health: "Where in the pipeline are products getting stuck before they reach the shelf?"
  • Store execution: "Is the problem that products are not delivered, or that they are delivered but not put on the shelf?"
  • Revenue forecasting: "How much sales volume are we losing to avoidable out of stocks?"

Without OSA measurement, brands and retailers are guessing at shelf execution. And guessing is expensive.

How it works in practice

OSA is measured by checking whether specific SKUs are physically present on the shelf during a store visit or audit. The measurement can happen through several methods, each with different tradeoffs in cost, accuracy, and frequency.

MethodHow It WorksAccuracyFrequencyBest For
Manual AuditA field rep or auditor walks the aisle and records which SKUs are present or missingGood, but depends on auditor diligencePeriodic (weekly or monthly)Small stores, detailed planogram compliance checks
Image RecognitionA rep photographs the shelf, and AI detects product presence and gaps automaticallyHigh, with trained modelsEvery visit (daily or weekly)Scalable coverage across large store networks
IoT SensorsWeight sensors or smart shelves detect when a product position is emptyVery high for real time detectionContinuousHigh value categories, pilot programs
EPOS DataSales data is analyzed to infer availability. Zero sales for a listed SKU suggests it was not on the shelfIndirect, can miss partial availabilityContinuousIdentifying problem stores at scale
Combined ApproachImage recognition for direct measurement, EPOS for validation and trend analysisHighestContinuousLeading CPG teams

Scenario: A beverage brand tracks OSA across 500 stores. A Monday audit shows 95% OSA. But when the brand looks at availability by time of day, OSA drops to 78% by 4 PM on Saturdays because stores sell through faster than they restock. The headline number hides a critical pattern.

For the technically minded: OSA measurement is being transformed by computer vision. A field rep takes a photo of the shelf, and an AI model detects which products are present and which positions are empty, generating an OSA score in seconds. Under the hood, this is an object detection pipeline (typically a CNN or vision transformer) trained on product packaging, running inference on shelf images and outputting a presence matrix that maps directly to the planogram. The result integrates with replenishment systems to trigger automatic restock orders when OSA drops below threshold.

Key metrics & related concepts

  • OSA %: the percentage of measurement instances where a specific SKU was found on the shelf. The headline number.
  • Availability by time of day: tracks how OSA changes throughout the day. Reveals restocking gaps and peak demand shortfalls.
  • Availability by store tier: segments OSA by store size or importance. A 95% average might mask 70% OSA in your highest volume stores.
  • Phantom inventory (theoretical vs. actual gap): when the system says a product should be in stock but it is not on the shelf. This gap between recorded inventory and physical reality is one of the most costly problems in retail.
  • Out of Stock (OOS): the inverse of OSA. If OSA is 95%, OOS is 5%. Teams use both framings depending on the audience.

Common mistakes & misconceptions

Mistake #1: Confusing OSA with numeric distribution.
Numeric distribution tells you what percentage of stores carry your product. OSA tells you whether that product is actually on the shelf. A SKU can have 90% numeric distribution but only 70% OSA, meaning it is "listed" in most stores but frequently absent where it matters.

Mistake #2: Measuring OSA only during scheduled visits.
If your reps visit on Tuesdays and you measure only then, you have no idea what happens on Saturdays. Continuous measurement is essential to understand true availability patterns.

Mistake #3: Not distinguishing between "not on shelf" and "not in backroom."
A product missing from the shelf but sitting in the stockroom is a store execution problem. If it is not in the backroom either, the issue is upstream. Blurring these scenarios leads to wrong fixes.

Mistake #4: Treating all SKUs equally.
Key SKUs driving most sales should be held to 95% or higher. Slow movers may have lower targets. A blanket OSA score across your full range hides problems with the products that matter most.

Regional variations

OSA is a universal concern, but how it is measured and managed varies significantly by market:

  • US: NielsenIQ and Circana (formerly IRI) track OSA through store audit panels. Major retailers like Walmart and Kroger run their own OSA scorecards with strict supplier compliance targets.
  • UK: Retailer scorecards set aggressive OSA targets, often 97% or above for key lines. Failure to meet targets can result in financial penalties or delisting.
  • India: OSA tends to be lower due to supply chain fragmentation across thousands of distributors and millions of kirana stores. Brands often rely on distributor stock data as a proxy for direct shelf measurement.
  • NZ/AU: Chain level OSA reporting is standard, with Coles and Woolworths requiring detailed availability data from suppliers.

How leading teams use OSA

Top performing CPG teams have moved beyond periodic audits to build continuous OSA monitoring into their operating rhythm. Image recognition at every store visit generates a steady stream of shelf presence data, replacing the old model of sampling a few stores per month.

The best teams layer in predictive alerts. By analyzing historical OSA patterns alongside current inventory levels and sales velocity, the system flags SKUs at risk of going out of stock before it happens. A rep gets a notification: "Store 247, SKU 8912, predicted OSA risk, recommend priority restock."

The most advanced organizations close the loop by integrating OSA data with ordering systems. When shelf availability drops below threshold, the system generates an automatic replenishment trigger based on real consumption data. The result is a self-correcting shelf that stays stocked without waiting for a human to notice a problem.

OSA is no longer just a metric you report. It is a signal that drives action.

Sources and further reading


Lighthouse for CPG

See how Lighthouse turns these terms into retail execution

Lighthouse connects distribution, execution, trade, and supply into one system your commercial teams act on at the store and SKU level.

Request a demo
Cyril Ovely
Co-Founder and CTO, Vxceed

Cyril is the Co-Founder and CTO at Vxceed. With over two decades of experience in engineering and entrepreneurship, he focuses on building scalable SaaS solutions that transform demand chain execution and help businesses operate with greater agility in evolving markets.