The most powerful diagnostic for understanding grocery spoilage shrink¹—and preventing it at the root cause level—is surprisingly simple: When inventory days of cover exceeds shelf life, shrink happens. That single ratio—days of cover divided by shelf life at a SKU/store level—can unlock millions in savings hiding in the data you already have.

Reducing retail spoilage shrink at the root cause level may be the single best investment grocery retailers can make. It’s a direct lift to profit while also improving quality and instock rates² at the same time. But using data systems to diagnose the root causes of shrink can be a daunting task, and often the interconnectedness of supply chain, pricing, merchandising, supplier management, and more can cause analysis paralysis. The right inventory health analysis can cut through all that complexity.

For a $500 million retailer operating on 3% net margins, cutting shrink by just 0.5 percentage points adds $2-3 million to the bottom line. But the payoff goes beyond dollars: Reducing shrink means fresher products reach customers, building the quality reputation that drives loyalty in an increasingly competitive market.

The catch? Many retailers never run this analysis because it seems daunting, and they often cite concerns such as: "Our data isn't clean enough." "We're worried reducing shrink means more empty shelves." "We're investing millions in AI ordering systems - isn't that enough?"

These are common concerns, but the good news is that none are roadblocks. The analysis I'm describing also improves instock rates and quality, complements your software investments, and works with messy, imperfect data. While your AI systems optimize ordering, this approach identifies root causes that those systems often can't see: supplier shelf life problems, merchandising issues, promotional hangovers.

The Ratio That Predicts Shrink

Based on my grocery supply chain experience, including building Amazon's produce, meat, and fish supply chain and developing grocery inventory optimization software, I’ve consistently seen that the most actionable way to analyze shrink is to break it down to its components of days of cover and shelf life, and look for when days of cover is exceeding shelf life (ratio > 1).

Let’s break it down:

• The analysis must be run at a SKU/store/day level.

• Days of cover is the total number of inventory units of a SKU that you have in a store, divided by the daily sales of that item. You can use your daily forecast if forward looking, or sales if backward looking.

• Shelf life is the number of days the item will last in store before it is pulled from shelves (e.g. for spoilage reasons). For products with expiration dates, this is best calculated using the actual data of how much shelf life you’re receiving from suppliers. If your customers aren’t shopping First Expire First Out (they aren’t), then the shelf life you can expect is actually less than the number of days you receive.

The best way to start is to pick a few known problem SKU/store combinations to get comfortable with the data, then start scaling up the analysis to your entire network (every single perishable SKU/store combination) to see the real valuable insights. The math is quite simple, but the implications are profound for understanding your inventory.

With this formula as your starting point, you can diagnose what's driving stubborn shrink.

• Is the days of cover numerator too high (over-ordering, too much shelf space, excessive minimum order quantities)?

• Or is the shelf life denominator too low (supplier issues, product sitting in the distribution center too long)?

Each answer points to different solutions—and different departments that need to act.

European retail researchers independently validated this approach. The ECR organization analyzed over 17,000 item-store combinations across 27 supermarkets and found that this ratio explained 42% of waste variation and helped one retailer save €1 million annually. This isn't theory. It's proven math.

From Diagnosis to Root Cause Prevention

The real power isn't just knowing the formula, it's systematically applying it to solve root causes and predict problems before they escalate. By using this formula to focus on the components of what’s actually causing shrink, you can easily decide on actionable steps so it’s not a blame game or merely a pilot with good intentions that doesn’t move the needle. It’s actionable, and you can easily measure how any changes improve the data. And, by monitoring the days-of-cover ratio in near real-time, you can spot emerging problems before they result in waste. If shelf life declines for a supplier, or if inventory builds after a promotion, the ratio trends upward—giving you an early warning to adjust before product expires. And you can use the same analysis looking backwards and forwards in time.

Using this analysis as a leading indicator of shrink also protects your margins. When you catch problems early you can adjust orders, work with suppliers, or shift inventory, which preserves full margin. Compare that to discovering the problem when the product is two days from expiration and your only option is a 50% emergency markdown. One approach fixes root causes. The other is a recurring tax on profitability.

Working With Imperfect Data

"Our data isn't good enough," is the objection I hear most. The truth? You probably have more useful data than you realize, scattered across systems.

• POS systems → sales velocity

• Inventory systems → stock levels

• Supplier EDI feeds → delivery timing, quantities, and often expiration dates

• Waste tracking → where shrink spikes already are

The key is triangulating these imperfect sources to calculate days-of-cover ratios by item and store. You don't need hyper-precision; you're looking for systematic issues. The biggest shrink problems show up as trends affecting multiple SKUs, stores, or time periods.

Your Next Step

With grocery retailers typically experiencing 3+% shrink rates, and perishables accounting for nearly two-thirds of that loss, even small improvements mean millions in bottom line savings. If you’re an executive, take this back to your team and see how it can help reduce shrink. If you’re an analyst, try this analysis and see what specific opportunities you can find. The analytical framework exists. Your data exists. So go ahead, connect the dots and see how much you can reduce shrink.

About the Author: Arthur Peterson leads Food Waste Lab, a food waste data consultancy that partners with grocery retailers to develop data-driven shrink reduction strategies through deep data analysis and supply chain optimization. Arthur is also a Principal Consultant with Good Food Advisors. Previously, Arthur led software product development for Grocery Inventory Health at Amazon, and helped build, scale, then manage Amazon Fresh’s Produce, Meat, and Fish supply chain and inventory control team.

¹Spoilage shrink is when product sits in store until its pull date (either expiration date or due to quality issues). This is the type of shrink that most directly leads to food waste.

²Quality refers to the freshness or remaining shelf life that customers can expect to have with perishable products, and instock rate is the measurement of fullness levels and product availability.

The views and opinions expressed in this guest blog are those of the author and do not necessarily reflect those of ReFED.