By Elie Dufeu, CTO & Co-Founder, Metreecs. Published 16 November 2026.
Days of inventory on hand is the number of days your current stock would last at your recent rate of sale: average inventory at cost, divided by cost of goods sold (COGS), multiplied by the days in the period. To reduce it without losing sales, find which product segments hold the days and apply a different fix to each one.
Most teams know their days of inventory on hand as one number on a monthly finance pack. It moves a little, someone asks why, and the answer is usually "we bought ahead" or "sales were soft". That single figure is an average of products that turn in three weeks and products that have not sold since spring, so it says very little about what to do next.
This guide gives the formula with a worked example, shows how to break the number down by segment, and matches a lever to each segment. If you would rather start from your own figures, you can walk through your inventory days with our team.
Key Takeaways
- Days of inventory on hand = average inventory at cost / COGS x days in the period. A retailer with €3.6M of average inventory and €14.6M of annual COGS holds 90 days.
- The 1,000 largest nonfinancial companies in North America held inventory for 56 days on average in fiscal 2025, 0.6 days more than the year before, according to The Hackett Group's 2026 working capital survey as reported by CFO.com.
- The same survey puts the cash that could be freed from inventory at $616 billion, out of a record $1.94 trillion working capital opportunity.
- US retailers held inventory equal to 1.36 months of sales in August 2026, about 41 days measured against sales, per US Census Bureau data published through FRED.
- McKinsey reports that early adopters of AI-enabled supply chain management improved inventory levels by 35% and service levels by 65% compared with slower-moving competitors.
What is days of inventory on hand?
Days of inventory on hand (DOH) is the average number of days a business holds stock before selling it. It is calculated by dividing average inventory at cost by cost of goods sold for a period, then multiplying by the number of days in that period. The lower the figure, the faster stock turns back into cash.
Finance teams usually call the same metric days inventory outstanding (DIO), days sales of inventory (DSI), or days in inventory. The arithmetic is identical. DIO tends to appear in board packs and bank covenants, while planners and warehouse teams say inventory days on hand.
There is one distinction worth keeping. The finance version looks backward, because it divides stock by what was sold over the last month, quarter, or year. Planners often run a forward version that divides stock by forecast demand, usually expressed as weeks of supply. The two can disagree sharply just before a peak, when stock is high against past sales and normal against the sales about to arrive.
What counts as a healthy level depends on what you sell, and a grocer and a furniture retailer will never share a target. We cover sector ranges in a separate guide on what a good DIO looks like in retail, so this article stays on the calculation and on how to bring the number down.
The days of inventory on hand formula, with a worked example
The formula is:
Days of inventory on hand = (average inventory at cost / cost of goods sold) x number of days in the period
Take a home goods retailer with 30 stores (an illustrative example with our own arithmetic). Over the last 12 months its COGS was €14.6M, and the average of its 12 month-end inventory balances, valued at cost, was €3.6M.
- Work out COGS per day: €14.6M / 365 = €40,000.
- Divide average inventory by COGS per day: €3.6M / €40,000 = 90.
- The retailer holds 90 days of inventory on hand.
The same result comes from inventory turnover. Turnover is COGS divided by average inventory (14.6 / 3.6 = 4.06 turns a year), and 365 divided by 4.06 gives 90 days. If your reporting is built on turns, our guide to raising inventory turnover works through that side.
You will meet three versions of the calculation, each suited to a different job.
| Version | Calculation | Best used for |
|---|---|---|
| Financial (value) | Average inventory at cost / COGS x days | Company and category reporting, cash planning |
| Turnover-based | Days in period / inventory turnover | Checking a figure quoted in turns |
| Operational (units) | Units on hand / average units sold per day | A single product in a single location |
The unit version is the one planners live in. A product with 420 units in stock that sells six a day has 70 days on hand. Because it needs no cost data, it works at the level of a product, a variant, or a store.
Three input errors account for most of the wrong numbers we come across:
- Mixing cost and retail value. Inventory at retail price divided by COGS overstates the days by your whole markup. Both sides have to be at cost.
- Averaging two points. Opening plus closing inventory divided by two misses everything that happened in between. Use monthly balances at least, weekly if the business is seasonal.
- Measuring on a flattering date. A year-end count taken after clearance shows a level the business never trades at.
Why the company-wide number hides where the days sit
Ninety days sounds like every product waits three months to sell. In practice, a handful of segments sit far above the average and one or two sit far below it. Here is the same retailer split into five segments (illustrative figures that add up to the totals above).
| Segment | Average inventory at cost | Annual COGS | Days on hand |
|---|---|---|---|
| Core fast movers | €0.9M | €7.30M | 45 |
| Seasonal lines | €1.2M | €4.38M | 100 |
| Slow tail | €1.0M | €2.19M | 167 |
| New launches | €0.3M | €0.73M | 150 |
| Aged and discontinued | €0.2M | none | no sales to measure against |
| Total | €3.6M | €14.60M | 90 |
Core fast movers generate half of COGS with a quarter of the stock. The slow tail and the aged products together hold €1.2M, a third of all inventory, for 15% of COGS. Any plan that trims every segment by the same percentage takes most of its cut from the 45-day products the business depends on.
Leila, the finance director at this retailer, had asked the buying team for a 10% reduction in stock across the board for two years running. Each time, availability on core tableware dropped and the total barely moved by year end.
When her team built the table above, the request changed: leave the fast movers alone, clear the aged stock, and bring the slow tail down by €300k. That takes average inventory to €3.1M and days of inventory on hand to about 78, with €500k released. (Leila is an illustrative example.)
Across Metreecs' work with multi-store retailers, the slow tail and aged stock usually hold more of the days than the team expects before the breakdown is done. The total gets reviewed every month. The split by segment often has never been produced, because it requires stock and sales at product level in the same place.
Statistic callout: Inventory accounts for $616 billion of the $1.94 trillion working capital opportunity among North America's 1,000 largest nonfinancial companies, per The Hackett Group's 2026 North American Working Capital Survey.
How to reduce days of inventory on hand, segment by segment
Each segment carries its days for a different reason, so each one responds to a different lever.
| Segment | What drives the days | Lever that fits |
|---|---|---|
| Core fast movers | Order size and safety stock | Smaller, more frequent orders, with safety stock set by variability |
| Seasonal lines | Size of the initial buy | Smaller first commitment, in-season reorders, earlier exit decisions |
| Slow tail | Minimum order quantities and too many locations | Hold centrally, lengthen review cycles, prune the range |
| New launches | No sales history | Forecast from comparable products, stage the buy |
| Aged and discontinued | Nobody owns the exit | Aging thresholds, a named owner, early markdown or liquidation |
Core fast movers
These products already turn quickly, and the stock they hold is mostly cycle stock (what arrives with each order) plus safety stock. Reducing days here means ordering less at a time and sizing the buffer to each product's actual demand variability instead of a flat number of weeks. The saving per product is small, and the risk is the highest in the range, so changes should go in gradually with availability tracked alongside.
Seasonal lines
Seasonal days on hand are decided on the day the initial buy is signed. A first commitment sized to the uncertain part of the forecast, followed by reorders once real sales come in, holds far less stock in the back half of the season. It depends on suppliers who can deliver a second drop in time. Where they can't, the remaining lever is to make the markdown decision in week six instead of week twelve.
Slow tail
Slow products are often overstocked by construction. A minimum order quantity of 48 units on a product that sells one a week puts nearly a year of stock in the building on delivery day. Spreading a slow product across 30 stores multiplies the problem, because every store needs at least one unit on display plus a spare. Holding the tail in one central location and shipping to stores or customers on demand cuts the days without removing the product from the range.
New launches
A launch with no history tends to be bought on optimism, and 150 days on hand is the result. Forecasting from comparable products (same category, price band, and launch season) gives a starting quantity that can be defended. Staging the buy in two or three deliveries lets the second one follow actual sales.
Aged and discontinued stock
This segment has no sales to divide by, so every euro in it adds to the days until it leaves. Set an aging threshold by category, review everything past it each month, and give one person the authority to mark down, bundle, or liquidate. Our guide to reducing excess inventory and dead stock covers the exit routes and how to pick between them.
The supply-side drivers: order frequency, lead time, and forecast error
Three inputs set how much stock a product needs before anyone makes a buying decision. Each can be translated into days.
Order frequency. Average cycle stock is half of what you order each time. A product ordered in eight-week quantities carries four weeks of cycle stock on average, or 28 days. Ordered every two weeks, it carries seven days. The demand is the same, and the product now holds 21 fewer days.
Lead time. Safety stock grows with the square root of lead time. Cutting a supplier's lead time from 12 weeks to six reduces the safety stock needed for the same service level by about 29%, because the square root of one half is 0.71. Shorter lead times also reduce stock in transit, which sits in the inventory figure once you own it.
Forecast error. Safety stock is proportional to the size of the forecast error it has to absorb. A 20% reduction in error at product and location level allows a 20% smaller buffer for the same availability. This is where forecasting quality shows up on the balance sheet.
Tomás buys core styles for a footwear retailer. His main supplier shipped full containers every eight weeks, which made the freight invoice look good and left his best sellers with close to two months of stock after each delivery.
He negotiated mixed containers every three weeks for the 40 core styles, kept the eight-week rhythm for everything else, and accepted a slightly higher freight cost per pair. Stock on those styles fell by roughly half while availability held. (Tomás is an illustrative example.)
The finance leaders we work with often underestimate how much of the number is fixed at the purchase order, months before it reaches the balance sheet. By the time a high figure shows in the monthly pack, the only remaining options are promotions and markdowns. Product-level forecasting and AI-driven inventory optimization move the decision back to the point where the order quantity is set.
Statistic callout: Early adopters of AI-enabled supply chain management improved inventory levels by 35% and service levels by 65% compared with slower-moving competitors, per McKinsey's April 2021 research on AI in supply chains.
Where reducing inventory days goes wrong
A lower number is only an improvement if the business sells as much as before. Four patterns produce a better figure and a worse result.
Even cuts. One mistake we repeatedly see is a single days-of-cover target applied to every product. Fast movers hit the target by running out, and slow movers miss it because their minimum order already exceeds it.
Year-end timing. Delaying receipts into the next period lowers the closing balance and does nothing for the average. If the 12-month average of month-end balances hasn't moved, the reduction is cosmetic.
Ignoring margin and price changes. A deep clearance lowers inventory and also lowers the COGS value of what remains if stock is written down. Check that the days fell because units left, and at what margin.
Buffers nobody revisits. The Hackett Group notes that companies kept carrying buffer stock in 2025 to manage tariff exposure and supply uncertainty. Extra cover taken for a specific risk is reasonable, provided it has a named reason and a review date.
Ingrid runs operations for a pet supplies retailer. Under pressure to hit a days target by December, her team cut every open order by 15%.
The figure landed on target, and in January the best-selling dry food lines were out of stock in a third of stores while aquarium accessories still had months of cover. The next year she set targets by segment and reported days on hand next to availability on the top 200 products. (Ingrid is an illustrative example.)
To keep a reduction honest, track three things together each month: days on hand by segment, availability on the products that drive sales, and the value of stock past its aging threshold. Our article on working capital tied up in inventory shows how to turn each day saved into a cash figure for the finance team.
FAQ
What does days of inventory on hand mean in simple terms?
It is how many days your stock would last if you kept selling at your recent pace and received nothing new. Ninety days on hand means the business holds about three months of sales in stock, valued at cost.
How do you calculate days of inventory on hand?
Divide average inventory at cost by cost of goods sold for the period, then multiply by the number of days in the period. For a single product, divide units on hand by average units sold per day. Use monthly or weekly balances for the average, since a single year-end figure can mislead.
Is days of inventory on hand the same as DIO or DSI?
Yes. Days inventory outstanding (DIO), days sales of inventory (DSI), and days in inventory all describe the same calculation. Weeks of supply is different, because it divides stock by forecast demand and looks forward.
What is a good number of days of inventory on hand?
It depends on the sector, the margin, and the season. Food retailers run on a few weeks, while furniture and jewelry hold several months. Judge your own figure by its trend and by availability on your best sellers instead of a universal target.
How fast can a retailer reduce days of inventory on hand?
Aged and discontinued stock can leave within weeks once someone owns the decision. Changes to order frequency and safety stock show up over two or three replenishment cycles. Seasonal buys only change at the next buying window, so the full effect often takes two seasons.
Does lowering inventory days cause stockouts?
It does when the cut is applied evenly. Fast movers have little excess to give, so a flat reduction empties their shelves first. Reductions aimed at the slow tail, aged stock, and oversized orders lower the average while leaving availability on core products intact.
Conclusion
Days of inventory on hand is simple to calculate and easy to misread, because the company-wide figure averages segments that behave nothing alike. Calculate it once on a clean cost basis, split it by segment, and aim each lever at the stock that carries the days. To see that breakdown on your own products and stores, book a demo of Metreecs' inventory planning.
Sources
- The Hackett Group, 2026 North American Working Capital Survey (FY2025 results of the 1,000 largest nonfinancial companies headquartered in North America), Business Wire release and CFO.com coverage: days inventory outstanding up 0.6 days to an average of 56 days, $616 billion inventory opportunity within a record $1.94 trillion, buffer stock carried for tariff exposure and supply uncertainty.
- US Census Bureau, Retailers: Inventories to Sales Ratio, seasonally adjusted, via FRED, Federal Reserve Bank of St. Louis: 1.36 in August 2026, or about 41 days of sales (1.36 x 30.4).
- McKinsey & Company, article on succeeding with AI in the supply chain (April 2021): early adopters improved logistics costs by 15%, inventory levels by 35%, and service levels by 65% compared with slower-moving competitors.








































































