Inventory Shrinkage: Causes and How to Track It

Most inventory shrinkage advice is written for a retail store with a sales floor, a shoplifter problem, and a loss-prevention budget. If you run a warehouse and sell to other businesses, most of that doesn’t apply to you. Nobody is walking out your front door with a jacket under their coat. Your shrinkage is coming from somewhere else entirely, and the articles built around shoplifting stats won’t help you find it.

What is inventory shrinkage?

Inventory shrinkage is the gap between what your records say you have and what you actually have on the shelf, measured in units or dollar value. If your system says 500 units of a SKU and a physical count turns up 480, you’ve shrunk 20 units. For a wholesale distributor, that gap usually comes from four places: theft (employee, rarely customer, since customers aren’t walking your aisles), damage that never got logged, administrative or counting error, and supplier shortages on receiving, meaning a purchase order for 500 units that arrived as 480 but got checked into the system as a full 500 anyway.

How to calculate your shrinkage rate

The formula is simple:

Shrinkage rate = (recorded inventory value − actual inventory value) ÷ recorded inventory value × 100

Worked example: your system shows $120,000 of inventory value for a given warehouse. A physical count values what’s actually on the shelf at $114,000. The gap is $6,000, and $6,000 divided by $120,000 is 0.05, or a 5% shrinkage rate. Run this per SKU, not just at the warehouse level, and you’ll usually find the total is hiding a lot of variance: most of your catalog shrinks at 1% or less, and two or three SKUs are doing most of the damage.

What realistic benchmarks look like

General retail benchmarks put average shrinkage around 1.5% of sales, with anything from 0.5% to 2% considered normal and anything past 5% a real problem. Those numbers come from stores with foot traffic and shoplifting exposure, which most B2B distributors don’t have. If you’re a warehouse operation with no retail floor, no walk-in customers, and controlled access, your realistic target is meaningfully lower, closer to 0.2% to 1%. If a warehouse-only operation is running shrinkage above 1.5%, that’s not a theft problem in the way a retail benchmark would suggest. It’s almost always a receiving or counting process problem, and it’s worth treating it as one before assuming someone on staff is stealing.

One distributor we’ve talked to found this the hard way: a 2.8% shrinkage rate that looked like a theft issue on paper turned out to be almost entirely a receiving problem on one high-volume SKU. A supplier had switched to shrink-wrapped pallets of 48 units instead of individually-boxed cases of 50, and the receiving team had kept checking in pallets at the old count out of habit for three months before anyone noticed the math no longer matched. That’s 96 units a month, on one SKU, from a process change nobody flagged, not a person walking off with product.

The four real causes, and how each one actually shows up

Theft. In a warehouse setting this is almost always internal, an employee with access to stock and the opportunity to move it without anyone noticing a specific unit is gone. It tends to concentrate on high-value, easy-to-move SKUs rather than spreading evenly across the catalog, which is itself a useful signal: if your shrinkage is one or two expensive SKUs consistently underperforming their count while everything else tracks fine, that pattern points toward theft in a way that even, catalog-wide shrinkage doesn’t.

Damage. A pallet gets dropped, a case gets crushed in a forklift pass, a product gets damaged sitting too long in a corner nobody was watching. The actual loss isn’t the damage itself, it’s the damage that never gets written up. A warehouse where damaged units get logged and pulled from sellable inventory the same day shows up in your books honestly. One where a damaged case gets shoved behind a shelf and dealt with “later” shows up as a mystery gap during the next count instead, weeks after anyone remembers what happened to it.

Administrative and counting error. This is the least dramatic cause and, for most B2B distributors, the biggest one: a picker grabs the wrong SKU off an adjacent shelf, a unit gets logged into the wrong bin location and effectively disappears from the count that matters, a return gets processed back into stock at the wrong quantity. None of this is theft or damage, it’s just human counting error compounding quietly across thousands of transactions a month. If you’ve never audited how often your own bin-to-SKU accuracy is actually right, this is very likely where a chunk of your shrinkage is hiding.

Supplier shortage on receipt. This is the cause retail-focused shrinkage articles almost never mention, because a store buying from a distributor rarely deals with it directly, but it’s a real and common source for anyone receiving pallet or case-pack freight. A purchase order for 500 units shows up short at 480, and if receiving checks the shipment in against the PO quantity instead of a physical count, the missing 20 units get absorbed into your system as if they arrived. You never had them, but your records say you did, and the “loss” only surfaces months later at your next cycle count, disconnected from the shipment that actually caused it. It’s also the cause most tied to the supplier relationship itself: a vendor who’s reliable about lead times tends to ship accurate counts too, and a short shipment is often the first sign a supplier relationship is starting to slip before the lead time itself does.

Tracking shrinkage without an ERP

You don’t need a full ERP system to catch most of this. A few practices, done consistently, catch the majority of shrinkage a small distributor deals with:

Count against the physical shipment, not the purchase order. Every inbound delivery gets counted by hand against what actually arrived, not rubber-stamped against what the PO says should have arrived. This one habit closes off the supplier-shortage cause almost entirely, and it’s the single most commonly skipped step we hear about from distributors who later find a shrinkage problem was really a receiving problem.

Run cycle counts on a rotation, not one big annual count. Counting your entire warehouse once a year means a problem that started in February doesn’t surface until December, ten months of compounding error with no chance to catch it early. Splitting your catalog into sections and counting a different section every week means every SKU gets a real count every couple of months, and a gap that opens up gets caught within weeks instead of most of a year.

Log damage the moment it happens, not at the next count. A simple damage log, even a shared spreadsheet by the receiving dock, that gets a line the same day a pallet gets dropped keeps damage from becoming an unexplained gap later. The goal isn’t a formal process, it’s just making sure damage gets written down before everyone forgets it happened.

Track shrinkage per SKU, not just warehouse-wide. A single blended shrinkage number tells you almost nothing about where to look. Per-SKU tracking is what turns “we’re off by 5%” into “we’re off by 5% because of these three SKUs,” which is an actual, fixable finding instead of a vague worry.

Reconcile discrepancies against a real record, not memory. The practices above only work if there’s something to compare a physical count against in the first place, and that’s exactly where a spreadsheet maintained by memory starts to break down at any real volume. This is closer to what real-time inventory tracking is supposed to solve: a record that reflects what should be on hand at any moment, so a discrepancy shows up as a discrepancy instead of getting explained away as “we probably miscounted.”

Shrinkage isn’t the same problem as dead stock

It’s worth being precise about what shrinkage is not. A SKU that’s genuinely unsellable, discontinued, seasonal, or simply not moving anymore, isn’t shrinkage, it’s dead or slow-moving stock, a different problem with a different fix. Shrinkage is specifically inventory that your records claim exists and physically doesn’t, whether that’s from theft, damage, counting error, or a short shipment. Mixing the two together in your reporting makes both harder to act on: a shrinkage number inflated by dead stock you haven’t written off yet looks worse than your real loss problem, and a dead stock number that’s actually shrinkage in disguise means you’re trying to discount your way out of what’s really a receiving or counting gap.

Where this becomes a system problem, not a counting problem

Manual cycle counts and a damage spreadsheet catch a lot, but they all depend on someone remembering to compare the record against reality regularly and consistently, across every SKU, every week. That’s the part that quietly stops happening once a warehouse gets busy or a key person is out for two weeks. Inventory Tracking is built around exactly this gap: it keeps a live record of what’s on hand and flags a discrepancy between recorded and actual counts as it appears, rather than waiting for the next scheduled count to surface a gap that’s already months old. Paired with accurate reorder point math, which only works if the inventory number it’s built on is trustworthy in the first place, catching shrinkage early stops being a once-a-year fire drill and becomes a normal part of how the warehouse runs.

FAQ

What is a normal inventory shrinkage rate? General retail benchmarks put average shrinkage around 1.5% of sales, with 0.5% to 2% considered normal. A warehouse-only distributor with no retail floor and controlled access should expect to run lower, closer to 0.2% to 1%, since the shoplifting exposure that drives up retail benchmarks doesn’t apply the same way to a B2B operation using inventory tracking to keep counts current.

How do you calculate inventory shrinkage rate? Subtract your actual, physically-counted inventory value from your recorded inventory value, then divide that gap by the recorded value and multiply by 100. $6,000 of shrinkage on $120,000 of recorded inventory is a 5% shrinkage rate. Running the same calculation per SKU rather than warehouse-wide usually shows the total is driven by a small number of problem SKUs, not spread evenly across the catalog.

What’s the most common cause of shrinkage for a distributor, as opposed to a retailer? Retail shrinkage skews heavily toward shoplifting. A warehouse distributor with no retail floor usually sees more shrinkage from administrative and counting error, plus supplier shortages that arrive short of the purchase order quantity but get checked in as a full count. Both are process gaps, not theft, and both show up as a mystery discrepancy at the next count rather than as an obvious event, the same discrepancy reorder point calculations quietly inherit if nobody catches it first.

Is shrinkage the same thing as dead stock? No. Shrinkage is inventory your records say exists but physically doesn’t, from theft, damage, error, or a short shipment. Dead stock is inventory that’s still physically there but isn’t selling anymore. Treating the two as the same problem makes both numbers harder to trust and harder to act on.

How often should a small distributor do physical inventory counts? Rotating cycle counts, a different section of the warehouse every week so the full catalog gets counted every couple of months, catch problems faster than one big annual count. An annual count means a discrepancy that started early in the year doesn’t surface until the count finally happens, by which point it’s much harder to trace back to whatever caused it.