Sales grew 12% in the year to September 2026; stock grew 31%. For the kitchenware brand this post follows (an example, not a client), the bank shows $210,000 and the balance sheet says it owns $680,000 of stock at cost, up from $520,000 a year ago. Is that extra $160,000 the holiday build, or cash sitting in products that have stopped moving?
Inventory KPIs are the few numbers that answer that each month: how fast stock sells through, what each dollar of stock earns, how much of it is getting old, and whether the count agrees with the accounts. Inventory KPIs are cash questions, so read them in dollars at cost from your accounts, not in units from a warehouse export.
The three inventory KPIs to start with
Start with the number that matches the problem in front of you:
- Cash is tight and stock keeps growing: days of inventory, beside open purchase orders.
- Deciding which line gets the next order's cash: GMROI (gross margin return on inventory) by product line.
- Old stock, or a write-down that surprised you: the share of stock older than 180 days.
- A year-end count that moved the P&L: inventory accuracy, before trusting any of the others.
- A launch that looks slow: sell-through.
- Storage bills rising faster than sales: carrying cost.
With no single problem, start with the first three. Together they say whether cash is going into stock faster than stock sells, which stock earns its keep, and which is turning into a write-down.
The one-page monthly inventory review
Copy this table into the pack you review each month. The figures are the example company's September, worked through in the sections below; the right-hand column is the question each number puts to the CEO or COO.
| KPI | September | Formula | Where the inputs live | The question it answers |
|---|---|---|---|---|
| Days of inventory | 115 days at month end; 101 days for the year | Inventory at cost ÷ daily cost of goods | Balance sheet inventory; P&L cost of goods | Can the next purchase order wait, and what cash is the shelf holding? |
| Inventory turnover | 3.6 turns | Cost of goods ÷ average inventory at cost | P&L cost of goods; 13 month-end inventory balances | Is the pace better or worse than last year? |
| GMROI by line | 4.4 total; cookware 2.2 | Gross profit ÷ average inventory at cost | P&L by product line; inventory by product | Where should the next dollar of stock go? |
| Aged stock share | 14.7% over 180 days; 5.3% over a year | Aged inventory at cost ÷ total | Inventory by receipt date | What needs a markdown or a write-down decision this month? |
| Inventory accuracy | 99.5% by value, before adjustment and excluding transit; 85% by product | Counted value ÷ book value; matched products ÷ products | 3PL file, Amazon reports, inventory adjustments | Can this month's product-level numbers be trusted? |
| Sell-through | 21.8% | Cost of goods ÷ (opening inventory + receipts) | Inventory roll-forward | Is stock arriving faster than it sells? |
| Carrying cost | 27.2% a year; $13,600 a month | Holding costs ÷ average inventory at cost | Storage, fees, insurance, adjustments, interest | What does each extra week of stock cost? |
| Open purchase orders | Deposits paid; balances due by date | Deposits plus balances due | Open purchase orders and the supplier ledger | What cash is already spoken for before the next order? |
| Fill and stockout rate | 96.8%; 5.5% of product-days | Units shipped ÷ units ordered; days out ÷ product-days | Order system and channel reports | How much margin did gaps cost, and was cash the cause? |
Where the inputs come from
A warehouse export tells you how many units you have. It does not tell you what they cost to land, what is on the water, or what went missing. Your accounts do, if they are kept right.
Cost of goods sold is on the P&L, dated to when each sale is recorded. Inventory is on the balance sheet at every month end, valued at landed cost, less any write-down for stock that will sell for less than it cost. Landed cost is the factory price plus the freight, duty and other costs it took to bring the stock to where it sells (the full list is in what landed cost is). Receipts, write-offs and storage fees each have their own account, so every KPI here is arithmetic on figures that reconcile to the bank.
Plenty of brands only true up inventory at year end, with cost of goods estimated in between, often as a flat percentage of sales. If that is you, these numbers are only as good as that estimate, and the year-end true-up can move a single month's margin a long way. The first job is getting inventory at landed cost into the accounts every month; then the KPIs follow. The question to ask your bookkeeper: is cost of goods each month what the units that shipped cost to land, or a percentage until the count?
The example company sells kettles, pour-over sets and enamel cookware on Shopify, Amazon, eBay and to about a dozen wholesale accounts, and carries 40 products. Its twelve months to September 30, 2026:
| Line | Amount |
|---|---|
| Net revenue | $4,800,000 |
| Cost of goods sold, at landed cost | $2,160,000 |
| Gross profit | $2,640,000 (55% gross margin) |
| Inventory at cost, September 30, 2025 | $520,000 |
| Inventory at cost, September 30, 2026 | $680,000 |
| Average inventory at cost, 13 month-end balances | $600,000 |
| Units shipped in the year, average landed cost $18 | 120,000 |
The seven inventory KPIs in detail
Use these as reference, in the order they earn a CEO's attention.
1. Inventory turnover and days of inventory
Inventory turnover is cost of goods sold for a period divided by average inventory at cost. Days of inventory, also called days inventory outstanding, is 365 divided by turnover: how long the average unit sits between arriving and shipping. One measurement, read two ways.
Formula. Turnover = cost of goods sold ÷ average inventory at cost. Days of inventory = 365 ÷ turnover.
Worked figure. $2,160,000 ÷ $600,000 = 3.6 turns, or 101 days. Using sales instead of cost of goods would put the markup in the numerator and show 8 turns, more than twice the real pace. Category comparisons, with sources, are in inventory turnover benchmarks.
At month end the opening question gets its answer. The year's cost of goods averages $5,918 a day, so the closing $680,000 is 115 days of stock against 101 for the year. Those 13 to 14 extra days are about $80,000 of cash, the gap between the closing balance and the $600,000 average.
A bad reading. Days rising while sales are flat: cash is going into stock faster than stock sells. Days measure how long stock lasts, not when an order pays back; cash returns as net receipts on each channel's payment dates, worked through in from factory deposit to payout. Days rising ahead of a known peak, in the products that peak will sell, is a plan rather than a problem.
The decision. Whether the next purchase order is affordable now or should wait for stock already paid for to sell. Set the cash on the shelf beside the bank balance and a 13-week cash flow forecast; weeks of supply reads the same position forward, week by week.
2. Gross margin return on inventory (GMROI)
GMROI is gross profit divided by average inventory at cost: the gross profit each dollar of stock earned in a year, which turnover alone cannot show.
Formula. GMROI = gross profit ÷ average inventory at cost.
Worked figure. $2,640,000 ÷ $600,000 = 4.4 overall, with a wide spread underneath:
| Product line | Gross profit | Average inventory at cost | GMROI |
|---|---|---|---|
| Kettles | $1,100,000 | $180,000 | 6.1 |
| Pour-over and accessories | $1,120,000 | $230,000 | 4.9 |
| Enamel cookware | $420,000 | $190,000 | 2.2 |
| Total | $2,640,000 | $600,000 | 4.4 |
Enamel cookware holds almost a third of the cash in stock and earns about a sixth of the gross profit.
A bad reading. A line falling below the company's own average while its stock grows. Judge each line against the total and against last year, not a borrowed rule of thumb. If Amazon's fees and ad spend sit below gross margin in your accounts, run a second version after both, because a line can earn well before them and poorly after.
The decision. Where the next dollar of stock goes. With limited cash for purchase orders this quarter, the kettle order earns more per dollar than the cookware order.
3. Aged stock as a share of inventory
Aged stock is inventory at cost grouped by how long ago it was received. The share over 180 days, and over a year, is the early warning for dead stock.
Formula. Aged share = inventory at cost received more than N days ago ÷ total inventory at cost.
Worked figure. At September 30:
| Age since received | Inventory at cost | Share |
|---|---|---|
| 0 to 90 days | $412,000 | 60.6% |
| 91 to 180 days | $168,000 | 24.7% |
| 181 to 365 days | $64,000 | 9.4% |
| Over 365 days | $36,000 | 5.3% |
| Total | $680,000 | 100% |
So $100,000, 14.7% of stock, is more than six months old. All $36,000 over a year is one discontinued enamel kettle: 1,500 units that landed at $24 each.
A bad reading. A rising share over 180 days. A stable share with a few named products in it is a decision waiting to be made.
Stock is carried at cost only while it will sell for at least that much. Say that in October the company tests prices and finds $29 is the best the discontinued kettle can fetch. It takes $9 of marketplace fees and shipping to sell each one, so it nets $20, which is $4 below cost. If $29 is the best price the kettle can fetch, the $6,000 write-down belongs in October's accounts whether the company clears it now or holds it, not spread across the months it sat there.
The decision. Hold or clear. Clearing turns the remaining $30,000 into cash; holding keeps that cash on the shelf at about $8,200 a year to carry at the company's carrying cost rate (rank 6), and the price rarely rises with age.
4. Inventory accuracy: the count against the accounts
Inventory accuracy compares what is on hand, from the 3PL's stock file, Amazon's inventory reports and your own locations, with what the accounts say you own. It is the check on every other KPI here, since each one divides by the inventory balance.
Formula. Accuracy by value = counted value at cost ÷ book value at cost. Accuracy by product = products whose count matches ÷ products counted.
Worked figure. Before the September adjustment the accounts showed $683,300 and the locations reported $674,800, an $8,500 difference: $5,200 shipped to Amazon and not yet checked in (still owned, in transit); $1,900 that Amazon reported lost or damaged, which comes off inventory at cost in the month the loss happened, with any reimbursement booked separately once the claim is approved or payment is probable; and $1,400 that could not be traced after a recount, written off as shrink through an inventory adjustment in the month it was found. Nobody will ever know where those units went.
Leaving out the $5,200 in transit, the count was 99.5% of the books before the adjustment. After it, the books show $680,000. By product the picture is weaker: 34 of 40 products matched, 85%, because overs and unders on different products cancel out in the dollar total. Why counts drift, and how to work a mismatch location by location, is in inventory accuracy and ecommerce inventory management.
A bad reading. The same products missing month after month, which usually points to receiving or listing problems rather than theft.
The decision. Whether to trust product-level GMROI and aged stock this month, and whether receiving, the 3PL or a cycle count needs attention before year end.
5. Sell-through rate
Sell-through rate is the share of the stock available in a period that sold in that period.
Formula. Sell-through = cost of goods sold for the period ÷ (opening inventory at cost + receipts at cost).
Worked figure. September:
| Line | Amount |
|---|---|
| Opening inventory at cost | $653,300 |
| Receipts at landed cost | $220,000 |
| Available | $873,300 |
| Cost of goods sold | $190,000 |
| Lost, damaged and shrink written off | $3,300 |
| Closing inventory at cost | $680,000 |
$190,000 ÷ $873,300 = 21.8% for the month. If opening plus receipts less cost of goods, less write-offs and write-downs, does not equal the closing balance, something is booked wrong and every ratio this month is off.
A wholesale buyer's sell-through report is usually in units at their stores; this is the company-wide version in dollars, so do not compare the two directly. For a single product, the ratio in units is a close stand-in, because units of one product usually carry similar costs: a new kettle colorway received 3,000 units in early July and had sold 1,350 by September 30, 45% in thirteen weeks.
A bad reading. Sell-through falling while receipts rise, the same signal as days of inventory climbing, seen a month earlier.
The decision. Whether a slow launch gets a price change or more marketing, and whether the follow-on order should be smaller or later. A demand forecast sets the expected pace; sell-through shows whether the launch is keeping it.
6. Carrying cost
Carrying cost is what it costs in a year to hold stock: storage, insurance, shrink and write-offs, and the cost of the money tied up in it. The full method is in inventory carrying cost.
Formula. Carrying cost rate = annual holding costs ÷ average inventory at cost.
Worked figure.
| Holding cost, twelve months | Where it sits | Amount |
|---|---|---|
| 3PL storage | Storage expense, from the 3PL's invoices | $54,000 |
| Amazon storage fees | Fees on the Amazon settlements | $31,000 |
| Inventory insurance | Insurance expense | $6,000 |
| Shrink, lost units and write-downs | Inventory adjustments | $18,000 |
| Cost of money, 9% on $600,000 | Interest on the credit line, or your own rate | $54,000 |
| Total | $163,000 |
$163,000 ÷ $600,000 = 27.2% a year, about $13,600 a month.
A bad reading. Storage fees rising faster than average inventory, which often means slow stock sitting in the most expensive place.
The decision. What extra stock really costs. The 13 to 14 extra days from rank 1, about $80,000 of stock, cost about $21,800 a year to carry at this rate. Set that beside any discount a supplier offers for a bigger order.
7. Fill rate and stockout rate
Fill rate is the share of units ordered that shipped from stock on the first attempt. Stockout rate is the share of product-days a product could not be sold. The counts come from the order system and the channels, and the finance view is narrow: what the gaps cost in margin.
Formula. Fill rate = units shipped from stock ÷ units ordered. Stockout rate = product-days out of stock ÷ (products × days).
Worked figure. In September customers ordered 10,900 units and 10,550 shipped from stock, a 96.8% fill rate. Of the 350 short, 200 shipped later and 150 were cancelled, about $3,300 of gross profit at the company's average $22 a unit. A listing that is out of stock takes no orders, so fill rate misses it: across 40 products and 30 days, the company was out for 66 product-days, 5.5%. Nine were the best-selling kettle on Amazon, which sells about 40 a day at $89 and lands at $28, so about $22,000 of gross profit before Amazon's fees was at stake. How much of that came back later as delayed sales is a judgement; assume some did.
A bad reading. A falling fill rate or rising stockouts on the products with the highest GMROI.
The decision. When to buy and how much belongs to the planning tools the operations team runs. The finance question is whether cash caused the gap: if the kettle balance payment waited for a wholesale buyer to pay, the fix is in the cash forecast and in how fast orders turn into cash. The wider set of ecommerce numbers is in the ecommerce KPI dashboard guide.
Reading the review as cash
Each line of the monthly review is a dollar figure at cost, and each one ends in a cash decision: whether to place the next order, where its dollars go, and what to clear. In the example, the 31% stock growth comes down to about $80,000 of extra days, $36,000 of kettles worth less than they cost, and a cookware line earning half the company's average on its stock.
FAQ
How do you calculate inventory turnover?
Divide cost of goods sold for the period by average inventory at cost for the same period. A company with $2,160,000 of cost of goods in a year and $600,000 of average inventory turns its stock 3.6 times. Use cost of goods, not sales, because sales include the markup and inflate the ratio. Divide 365 by the turnover figure to get days of inventory.
What is a good GMROI for an ecommerce brand?
The useful comparison is internal: each product line against the brand's own total and against last year. A brand earning $2,640,000 of gross profit on $600,000 of average stock has a GMROI of 4.4. A line well below that while its stock grows is the first place to look before the next purchase order. If marketplace fees and ad spend sit below gross margin in your accounts, run a second version after both.



