There's no shortage of inventory numbers retailers can track. Between sales, on-hand quantities, replenishment, transfers, receiving, and forecasting, it's easy to end up with more reports than anyone has time to look at.
Instead of getting bogged down with reports, it’s best to focus on a handful of inventory KPIs that tell you what's actually happening with your business.
A weekly review should give you a sense of whether your inventory is accurate, whether customers can find what they want, how quickly products are moving, and whether your purchasing and replenishment decisions are keeping up with demand.
Here are seven worth keeping an eye on.
Inventory accuracy is one of those numbers that can affect almost everything else. If your system says you have 12 units of something, but there are only three sitting in the store, your replenishment decisions are already working with bad information. The same goes for eCommerce availability, transfers, purchasing, and fulfillment.
Calculate inventory accuracy by comparing how many of your counted SKUs actually match what the system says you have — count 500 SKUs, find 485 match, and you're at 97%. Most retailers would probably guess they're closer to that than they actually are. The Association for Supply Chain Management, citing Auburn University's RFID Lab, puts the average without RFID tracking at around 63%, rising to roughly 95% once retailers adopt it.
If your own number starts slipping, don't stop at the percentage.
Find out where the discrepancies are happening. One location may have a much larger problem than the rest of the business, or a particular category may consistently be off. Receiving, returns, transfers, shrink, and counting processes can all play a role.
Stockouts are the flip side of accuracy: how often is a product simply not there when a customer wants to buy it?
Divide the SKUs currently out of stock by your total assortment and multiply by 100, and you've got a rate you can track daily, weekly, or by location.
The interesting part is figuring out what's behind the number. A company-wide stockout rate can look fine even when individual stores are struggling — you might have plenty of a product across your network, just not at the location where a customer is actually standing. And the consequences aren't trivial: NielsenIQ has estimated that out-of-stock issues cost the U.S. retail industry around $48 billion a year.
Pay attention to which products are stocking out, where, and whether the same ones keep showing up — a recurring stockout deserves more of your attention than a one-off shortage. It also helps to check stockouts against inventory accuracy. If the system says you have inventory available but customers still can't buy it, accuracy may be the real problem hiding underneath.
An overall in-stock rate doesn't tell you whether your most important products are actually the ones staying on the shelf. This metric isolates that: in-stock top-seller SKUs divided by total top-seller SKUs, ideally broken out by location rather than rolled into one company-wide figure.
For a sense of where the bar generally sits, FMI (The Food Industry Association) has tracked retail out-of-stock rates at a typical historical level of around 8%, improving to 6.5% in 2023 as supply chains recovered from pandemic-era disruption. That puts a healthy store-wide in-stock rate in the low-to-mid 90s, with top sellers specifically tracked well above that.
A product might have plenty of inventory somewhere in the business while one store keeps running out — a sign of a transfer, replenishment, or allocation issue rather than a true shortage. For multi-location retailers, this number is particularly useful once you can see it store by store instead of averaged across the operation.
Inventory adjustments aren't automatically a bad thing.
Sometimes you find a discrepancy during a count or need to account for damaged or missing merchandise. The bigger question is how often that's happening — track this as adjusted units or dollar value against total inventory on hand for the same period.
If a store is constantly adjusting inventory, there's probably something worth looking into: a receiving issue, shrink, misplaced merchandise, transfer problems, or a process that's creating errors along the way. Shrink is a significant piece of that picture industry-wide — the National Retail Federation's 2023 Retail Security Survey put the average shrink rate at 1.6% of sales in fiscal 2022, about $112.1 billion across the industry, up from 1.4% the year before.
Rather than treating every adjustment as a problem, watch for patterns. A sudden increase at one location or within one category tells you a lot more than the total count across the business. It's especially useful alongside inventory accuracy — adjustments climbing while accuracy falls is a pretty clear signal something's worth digging into.
Your sell-through rate is one of the cleanest signals for how much of your inventory is actually moving versus sitting around. Say a store receives 1,000 units of a jacket and sells 750 of them by the end of the month — that's a 75% sell-through rate (units sold divided by units received, times 100).
The number needs context, and "good" varies a lot by category and price point, so your own category history is usually a better benchmark than any single industry number. A high sell-through can mean you've got a hit on your hands — or that you're about to run out. A low one might mean overbuying, though seasonal or newer items naturally take longer to move. Looking at sell-through by product, category, and location gives a much better picture than one number for the whole business.
Units on hand won't tell you much without knowing how quickly the units are selling. For example, let's say you have 100 units of a product and typically sell 20 per week — this means you have about five weeks of supply (units on hand divided by average weekly units sold).
This metric makes it easier to spot products heading toward a stockout as well as ones you're overstocked on — and the second case has a real cost. The Institute for Supply Management notes that most companies aim for inventory carrying costs in the 20–30% range of inventory value per year, once storage, insurance, and tied-up capital are factored in. A few extra weeks of supply sitting on a shelf is actively costing you, not just a number that looks a little high.
Just remember that recent sales aren't always a perfect picture of future demand. Promotions, seasonality, and changes in customer behavior can all affect how quickly inventory moves. That's where weeks of supply becomes particularly useful alongside forecasting.
Demand forecasting is supposed to help you make better inventory decisions. Forecast accuracy tells you how closely those expectations line up with what actually happens.
The standard way to measure it is Mean Absolute Percentage Error, or MAPE — averaging how far off your forecast was from actual demand, as a percentage. A lower MAPE means a tighter forecast; some retailers simply report accuracy as 100% minus MAPE.
What counts as "good" depends heavily on what you're forecasting. A staple grocery item with steady demand is a much easier problem than a seasonal fashion item with a few months of shelf life and no sales history, so compare performance within categories rather than across your whole assortment. A brand-new product in any category will also run a higher error rate simply because there's less history to work from.
You don't need to panic over one bad forecasting week — look for patterns over time. Maybe a category is regularly under-forecasted, or one location behaves differently from the rest of the business. Those patterns are what let you adjust purchasing and replenishment before they turn into bigger problems.
A high sell-through rate sounds great until you notice that weeks of supply has dropped to almost nothing. A stockout might look like a replenishment problem until you realize your inventory records are consistently inaccurate. This is where a weekly KPI review beats simply checking whether each number went up or down.
Start by comparing the current week to the previous few. When you see a meaningful change, drill into the products and locations behind it, reviewing the numbers weekly and flagging the biggest changes.
From there, compare related metrics. If stockouts are going up then check accuracy, weeks of supply, and forecast performance for those same products. If inventory is building up, look at sell-through and weeks of supply to see what's really sitting.
Inventory KPIs are only as useful as the information behind them. Sales, receiving, transfers, eCommerce orders, inventory counts, and replenishment all affect these numbers. When those processes live in disconnected systems, getting a clear picture can mean spending more time reconciling data than acting on it.
That's where unified commerce makes a difference. When your retail data is connected across locations and channels, you get a clearer picture of what's happening throughout the business — making it easier to spot availability issues, understand product movement, and make replenishment decisions using current information.
FieldStack brings POS, eCommerce, inventory, purchasing, forecasting, and other retail operations together on one unified commerce platform. Retailers can see what's happening across their business without piecing together data from separate systems.