INVENTORY TURNOVER RATIO

Inventory Turnover Ratio: Formula and Benchmarks

By Jason Osajima — former VP of AI at a $250M manufacturer · LinkedIn ·
Quick answer

Inventory turnover ratio formula, DSI, and real benchmarks by industry — plus why a high turnover number can hide a broken supply chain.

Inventory turnover ratio measures how many times you sell through and replace your stock in a year. The standard formula is cost of goods sold divided by average inventory, both stated at cost. A turnover of 6 means you cycle your entire inventory six times a year, or roughly every 61 days.

I ran planning at a $250M furniture manufacturer where the board fixated on this one number. Getting them to read it correctly mattered as much as moving it. A high ratio can mean you're lean — or that you're chronically understocked and bleeding sales on stockouts. A low one can mean dead inventory, or a smart buffer built ahead of a known disruption.

Here's the formula, the benchmarks that actually apply by industry, and how to read the number without fooling yourself.

The inventory turnover ratio formula

There are two common versions. Use the COGS one for operations.

Inventory Turnover = Cost of Goods Sold ÷ Average Inventory

Average inventory = (beginning inventory + ending inventory) ÷ 2, both carried at cost. The companion metric most operators find easier to feel is days of inventory.

Days Sales of Inventory (DSI) = 365 ÷ Turnover

These two are the same fact stated two ways. Turns for the finance crowd, days for the operators. Corporate Finance Institute (2026) defines turnover the same way, and notes you can derive days inventory outstanding either by dividing 365 by turnover or by the inverse ratio directly.

A worked example

Say you run COGS of $48M against average inventory at cost of $8M.

You're cycling your whole inventory six times a year and holding about two months of stock on average. That's a clean read for a general manufacturer. For a grocer it would be a disaster, and for an aerospace supplier it would be borderline reckless. Context is everything, which is why benchmarks matter.

Use COGS, not sales

The single most common error is dividing sales revenue by average inventory. Revenue includes margin; inventory is carried at cost.

Mixing them inflates the ratio and makes you look leaner than you are. This isn't a style preference — under FASB ASC 330 (ASU 2015-11, 2015), inventory is measured at the lower of cost and net realizable value, so the denominator is a cost figure by definition. Match it with a cost numerator.

Some published benchmarks still use sales in the numerator. When you compare against an outside number, confirm both sides use the same basis or you're comparing nonsense. NetSuite (2026) walks through why the COGS-and-average-inventory version is the defensible one for operational analysis.

Benchmarks by industry

There is no universal "good" turnover. It's entirely industry-dependent. A grocer turning 14 and a heavy-equipment maker turning 4 can both be perfectly healthy.

Rough ranges I'd anchor to:

Industry Typical turnover DSI
Grocery / perishables 12–20 18–30 days
Consumer electronics 6–10 36–60 days
Apparel / retail 4–8 45–90 days
General manufacturing 5–9 40–73 days
Industrial / heavy equipment 3–5 73–120 days
Furniture / building products 4–7 52–90 days
Aerospace / long-cycle 2–4 90–180 days

The driver is product shelf life, demand predictability, and production lead time. Perishables must turn fast or they rot. Long-lead capital goods can't turn fast — the build cycle alone is months.

One caution on these ranges: they're directional, not gospel. Two companies in the same NAICS code can land turns apart based on their make-to-stock versus make-to-order mix, how much raw material they pre-buy, and where they sit in the supply chain. A contract manufacturer holding customer-owned consigned stock will read very differently from one that buys everything itself. Treat the table as a starting hypothesis, then verify against peers with a similar operating model.

The macro benchmark hiding in plain sight

For a sanity check on the whole economy, the U.S. Census Bureau publishes an inventories-to-sales ratio every month. In its Manufacturing and Trade Inventories and Sales report (Census Bureau, 2026), the total business inventories-to-sales ratio sat at 1.32 at the end of March 2026, down from 1.38 a year earlier.

That ratio is the inverse-ish cousin of turnover: 1.32 months of inventory on hand implies roughly 9 turns blended across all U.S. trade. It won't replace your own segment benchmark, but it tells you which direction the whole supply chain is leaning. Compare yourself to your own sub-segment and your own trend first, the macro number second.

How to read the number honestly

This is where most teams go wrong. The ratio is a symptom, not a diagnosis. Read it in both directions.

A high turnover ratio — efficient, or starving?

High turns usually mean tight working capital and fresh inventory. Good. But push it too far and the number starts lying to you:

A rising turnover ratio with a falling fill rate isn't a win. It's a supply chain getting squeezed. If you don't already separate those two, our piece on service level vs fill rate untangles why a great-looking turns number can mask a service problem.

A low turnover ratio — bloated, or strategic?

Low turns usually flag excess, slow movers, and trapped cash. But sometimes it's deliberate:

The question is never "is the number high or low." It's "is the inventory I'm holding the right inventory." Which brings us to the real limitation.

Where turnover fits in the working-capital picture

Turnover doesn't live alone. It's one leg of the cash conversion cycle.

The ASCM SCOR Digital Standard (2020) frames cash-to-cash cycle time as inventory days of supply plus days sales outstanding minus days payable outstanding. Shorter inventory days pull the whole cycle down and free cash. That's why finance cares about turnover even when operations only feels it as days-on-hand.

Holding that inventory isn't free, either. According to the CSCMP State of Logistics Report (2026), inventory carrying costs — capital, storage, service, and risk — run a meaningful share of inventory value every year. Every turn you add is carrying cost you stop paying.

Put rough numbers on it. If carrying cost runs 20% of inventory value and you hold $8M in average inventory, you're spending about $1.6M a year just to keep it on the shelf. Lift turnover from 6 to 8 and average inventory drops to roughly $6M for the same COGS — about $400K of annual carrying cost gone, plus $2M of cash freed off the balance sheet. That's the math that makes a CFO care about a number operations thinks of as housekeeping.

The aggregate number hides everything that matters

A single company-wide turnover ratio is an average that conceals the truth. You can hit a healthy blended 6.0 turns while:

The aggregate looks healthy and you're still sitting on a pile of stranded cash. The fix is to calculate turnover by SKU segment — run it across your ABC classes, or per SKU, and the picture changes completely. A structured ABC-XYZ inventory analysis is the cleanest way to slice it: value on one axis, demand variability on the other.

When we segmented turnover at the furniture plant, the blended 5.5 hid a C-class turning under 2. That C-class was where the excess lived. The aggregate metric had been telling everyone things were acceptable for two years.

The pattern isn't unique to furniture. McKinsey (2024) found that some aerospace and defense manufacturers carried so much unneeded stock that the industry risked burning an extra $60 billion in cash — excess invisible at the aggregate level until someone analyzed inventory entitlement part by part.

Improving turnover the right way

Don't chase the ratio directly. That's how you cut stock into a stockout. Improve the inputs instead.

McKinsey's work on slow-moving inventory (2024) makes the same point from the cash side: predicting demand signals and constraining unnecessary inflows frees working capital while cutting holding and obsolescence costs. Turnover improves as a result of those moves, not as a target you slash toward.

Frequently asked questions

What is a good inventory turnover ratio?

It depends entirely on your industry. Grocers and perishables run 12–20 turns, general manufacturers land around 5–9, and long-cycle aerospace suppliers may sit at 2–4. Compare yourself to your own sub-segment and your own trend rather than a single universal target.

Should I use COGS or sales in the inventory turnover formula?

Use cost of goods sold. Inventory is carried at cost under FASB ASC 330, so the denominator is already a cost figure, and the numerator must match. Using sales revenue includes profit margin and inflates the ratio, making you look leaner than you actually are.

How do I convert inventory turnover to days of inventory?

Divide 365 by your turnover ratio. A turnover of 6 equals about 61 days of inventory on hand, and a turnover of 10 equals roughly 37 days. Turns and days are the same measurement expressed two different ways.

Why does my healthy turnover ratio still leave me with excess cash tied up?

Because the company-wide number is an average that hides SKU-level problems. Fast-moving A-items can mask slow C-items and obsolete stock that net out to a "fine" blended figure. Segment turnover by ABC class or by SKU to find where the stranded cash actually lives.

How is inventory turnover related to working capital?

Turnover drives the inventory leg of the cash conversion cycle. The ASCM SCOR model defines cash-to-cash time as inventory days plus receivable days minus payable days, so fewer inventory days free cash directly. Faster turns also cut the carrying costs you pay to hold stock.

Let's see what's worth building first.

A 15-minute call: tell me where your AI or planning is stuck, and I'll tell you the one thing worth building first — and whether it's worth doing at all.

More field notes

Multi-Echelon Inventory Optimization Explained SimplyService Level vs Fill Rate: Definitions and Trade-offsAI Demand Forecasting: How It Works in 2026Machine Learning for Demand Forecasting: A Primer