Inventory turnover, and what dead stock is really costing you

Inventory turnover is cost of goods sold divided by average stock at cost, and the invented homeware business worked through below turns its stock 3.15 times a year, or once every 116 days. That whole-business figure hides the only thing worth knowing. Broken down by product, one line turns 8.06 times a year and another turns 0.31 — 3.3 years of stock on the shelf — and that slow line earns £1,387 a year while costing £1,528 a year to hold. It loses £141 annually while every individual sale of it looks profitable. This page works the ratio properly, does it per product, prices what a slow line costs, and compares three ways out of it.

The formula, and the version that is wrong

Inventory turnover = cost of goods sold ÷ average inventory at cost
Average inventory = (opening stock + closing stock) ÷ 2
Days of stock = 365 ÷ turnover

Calder and Roe is an invented homeware seller used to keep the arithmetic concrete; every figure attached to it is illustrative. Over a year they had cost of goods sold of £148,000, opened with £41,000 of stock at cost and closed with £53,000.

Average stock = (£41,000 + £53,000) ÷ 2 = £47,000
Turnover = £148,000 ÷ £47,000 = 3.15 times a year
Days of stock = 365 ÷ 3.15 = 116 days

Every pound of stock they hold sits there for about four months before it becomes a sale. That is the number, and it is an honest one.

The mistake that inflates it

The single most common error in this sum is putting revenue on top instead of cost of goods sold. Calder and Roe turned over £296,000 of revenue:

£296,000 ÷ £47,000 = 6.30 times — which is exactly double the real figure

It is double because their gross margin is 50%, and the relationship is exact rather than approximate:

Revenue-based ratio = true turnover ÷ (1 − gross margin)

At a 50% margin the revenue version doubles the answer. At a 70% margin it more than triples it. The reason it is wrong is simple: the denominator is stated at what the stock cost you, so the numerator has to be as well, or you are dividing a selling price by a cost price and calling the result a frequency. If you ever see a turnover figure that looks surprisingly healthy, check which number is on top before believing it.

The other denominator problem

A two-point average of opening and closing stock is only as good as those two dates. A business that closes its year in January, having just sold through Christmas, will show a low closing stock, a small average and a flattering turnover that describes nothing. Where you have the data, average twelve month-end figures instead:

Average stock = (sum of 12 month-end stock values) ÷ 12

It takes a few minutes and it removes the single largest source of nonsense in this ratio. If your stock value swings by more than about 30% across the year, the two-point version is not worth calculating.

The same ratio per product, which is the only useful version

A whole-business turnover of 3.15 is a summary of things that have nothing in common. It is worth calculating once, to know roughly where you stand, and then never using to make a decision. Every decision is about a product.

Four of Calder and Roe’s lines, over the same year:

Read the last row. At 95 units a year against 310 units held, the lantern carries 3.3 years of stock. And look at what it occupies: the four lines together hold £10,856 of stock and produce £18,365 of cost of goods sold, so the lantern is 64% of the money and 11.6% of the movement.

Blended, the four turn 18,365 ÷ 10,856 = 1.69 times a year. Take the lantern out and the remaining three turn 16,237 ÷ 3,912 = 4.15 times. One product is dragging the whole group down by two and a half turns, and no whole-business average would ever have shown that.

Turnover on its own is still not the answer

Before condemning the lantern, note what turnover cannot see: margin. A line that turns twelve times a year at a 4% margin can be worth less than one that turns twice at 55%. The measure that puts both halves together is the gross profit a line produces per pound of stock it occupies:

Return on stock = annual gross profit ÷ average stock at cost

Which for the same four lines, using the gross profit each actually earns per unit after platform fees and fulfilment:

The lantern has by far the best margin per unit — £14.60 against the planter’s £3.20, more than four times as much — and by far the worst return on the money it occupies. A pound tied up in planters generates £6.29 of gross profit a year. The same pound in lanterns generates 20p. That is a factor of thirty-one, and it is invisible in every report that looks at margin per sale.

The line worth remembering: below 1.0, a product does not generate as much gross profit in a year as it has money sitting inside it.

What a slow line actually costs to keep

So far the lantern looks inefficient. Costed properly it is worse than that: it loses money.

Holding stock is not free, and the cost is usually stated as an annual percentage of the stock value. Build your own from your own figures. Calder and Roe use an illustrative 22%, made up of:

Applied to the lantern’s £6,944 of average stock:

£6,944 × 0.22 = £1,528 a year to keep it on the shelf

Against the £1,387 a year of gross profit it produces. The lantern loses £141 a year, and it does so while every single sale of it shows a healthy £14.60 of gross profit. Nothing in a product report, a margin report or a bestseller list would ever surface that. Only this sum does.

The same arithmetic applied to the planter: £738 × 0.22 = £162 a year to hold, against £4,640 of gross profit. It earns 28 times what it costs to keep.

The threshold, in one line

Because the holding cost and the return on stock are both annual percentages of the same number, they compare directly:

A line is losing money on its stock when its return on stock is below the holding cost rate

At a 22% holding cost, that threshold is a return on stock of 0.22. The planter is at 6.29, the napkins at 2.77, the jug at 1.79 and the lantern at 0.20. Three lines clear the bar comfortably and one sits just under it. That is a complete review of a product range, done with two columns you already have.

It also explains something that confuses a lot of sellers: why a high-margin product can be the worst thing in the business. Margin is measured per sale. Holding cost is measured per year. A product that sells rarely enough gets charged the annual cost many times over between sales, and the per-sale margin never sees it.

The dead stock decision: hold, discount or clear

Calder and Roe have 310 lanterns on hand at £22.40 each, which is £6,944 of stock, selling at 95 a year. Three options, all of them costed. The lantern sells at £44.00, the channel takes 10.5%, and fulfilment costs £2.38, which is where the £14.60 of gross profit comes from: £44.00 − £22.40 − £4.62 − £2.38.

The holding cost over the run uses the average stock across the run-down, which is roughly half of 310 units, or £3,472, times 22%, times the number of years. Option C is a single job-lot sale at £21.00 a unit with no channel fee but £0.80 a unit of cost to shift them, which brings in 310 × £21.00 − £248 = £6,262 of cash immediately.

On those three bottom-line numbers, keeping the lanterns is obviously best and clearing them is obviously worst. That reading is wrong, and the reason it is wrong is the whole point of this section: the table prices the lanterns and ignores entirely what the money could be doing instead.

The comparison that matters

Take option C against option A, because C is the clean case: it turns the shelf into £6,262 of cash today. Option A gets the same money back too, but gradually across 3.26 years, so on average it arrives about 1.63 years from now. The difference between the two is therefore 1.63 years of having £6,262 available to buy something else.

Calder and Roe’s other three lines return £11,596 of gross profit on £3,912 of stock, which is a return on stock of 2.96. Take a deliberately pessimistic 1.0 instead — one pound of gross profit per pound of stock per year, a third of what those lines actually manage:

£6,262 × 1.0 × 1.63 years = £10,207 of gross profit

Against that, option A earns £2,718 more on the lanterns themselves than option C does (£2,036 against −£682). It is not close, and at the business’s real return of 2.96 it is not in the same universe.

This arithmetic holds only if the released cash is actually redeployed. If the £6,262 sits in a bank account, the comparison collapses and option A wins on its own numbers. So the honest question is not “should I clear this stock” but “do I have something better to buy with the money, and will I actually buy it”. If the answer is no, the lanterns are as good a place for the cash as anywhere, and the £141 a year they lose is the price of not having a better idea.

A practical note on option C. Selling at £21.00 against a £22.40 cost is taking a loss, and it feels like one. Financially the £22.40 is gone whatever happens next; it was spent, and no decision made today can unspend it. The only live question is which future is better, and a future with £6,262 of cash and an empty shelf is usually better than a future with 310 lanterns and three years of holding costs. The discomfort is real and it is not an argument.

What turnover cannot tell you

Turnover is one of the most useful ratios a stockholding business has, and one of the easiest to over-read. Six things it does not know:

• It cannot see seasonality. A product that sells its whole year in six weeks has a perfectly good turnover figure and spends ten months looking like dead stock. Calculate seasonal lines over their season, and never judge one on an annual average taken in March.
• It cannot see why the stock is there. Some slow stock is a deliberate buffer against a long or unreliable lead time, and cutting it produces stockouts rather than efficiency. Check the reorder point before cutting the level.
• It cannot see attachment. A slow, expensive line that brings buyers in who then fill a basket with fast lines is earning more than its own row shows. If your order data can tell you what else is in the basket when a lantern sells, look before you cut. If it cannot, be cautious.
• It cannot see minimum order quantities. Sometimes stock is slow because the supplier will not sell fewer than 500, and the real decision is about the supplier rather than the product.
• It is distorted by how you value stock. Turnover compares cost of goods sold to stock at cost, and if your stock is carried at a cost that is out of date — an old landed cost, a pre-freight-increase price — the ratio inherits that error. The figure is only as good as the cost per unit behind it.
• It cannot tell you anything about stock you do not hold. Dropshipped and made-to-order lines have no turnover figure at all, and comparing them to stocked lines on this measure is meaningless. They have a different problem, which is margin per sale, and it is the one to look at instead.

The three numbers this page runs on all come from elsewhere on this site: the unit cost, which is landed cost per unit; the gross profit per unit, which is how to price a product so it actually makes money; and how much to hold in the first place, which is the reorder point formula.

Arithmetic and general information only — not financial, tax, legal or investment advice. Your own figures, and your own accountant, decide what any of this means for you.

Frequently asked questions

What is the inventory turnover ratio formula?

Cost of goods sold divided by average inventory at cost, where average inventory is opening stock plus closing stock divided by two. Days of stock is 365 divided by the turnover. In the worked example on this page, 148,000 pounds of cost of goods sold over 47,000 pounds of average stock gives 3.15 turns a year, or 116 days of stock.

Should I use revenue or cost of goods sold in the turnover formula?

Cost of goods sold. The denominator is your stock valued at what it cost you, so the numerator has to be a cost too, or you are dividing selling prices by cost prices. Using revenue inflates the answer by exactly one divided by one minus the gross margin, so at a 50 percent margin it doubles the ratio and at 70 percent it more than triples it.

What counts as dead stock?

There is no universal threshold, but there is a test that works: a line is losing money on its stock when the gross profit it produces per pound of stock in a year is below your holding cost rate. In the worked example the holding cost is 22 percent, so the threshold is 0.22, and the brass lantern at 0.20 sits under it. It earns 1,387 pounds a year and costs 1,528 pounds a year to hold.

How much does it cost to hold stock for a year?

Build the rate from your own figures rather than borrowing one. The illustrative 22 percent used on this page is 9 percent cost of money, 6 percent storage, 2 percent insurance and shrinkage and 5 percent obsolescence. Applied to 6,944 pounds of average stock that is 1,528 pounds a year, which is the number to compare against what the line earns.

Is it better to discount dead stock or wait for it to sell?

It depends entirely on whether you will actually redeploy the cash. In the worked example, holding earns 2,036 pounds over 3.26 years while clearing the lot at a loss earns minus 682 pounds, so holding looks better on its own numbers. But clearing turns the shelf into 6,262 pounds of cash today rather than on average 1.63 years from now, and even at a pessimistic return of one pound of gross profit per pound of stock a year that extra time is worth 10,207 pounds. If the freed cash would sit in a bank account instead, that argument disappears.

Why does a high-margin product show up as my worst line?

Because margin is measured per sale and holding cost is measured per year. A product that sells rarely gets charged the annual holding cost several times over between sales, and the per-sale margin never sees it. In the example the brass lantern makes 14.60 pounds a unit against the planter 3.20 pounds, but a pound tied up in planters generates 6.29 pounds of gross profit a year against the lantern 20 pence.

Other guides

• How much does Etsy take from a sale?
• How to build a cash flow forecast
• Agency utilisation rate: the formula, and the denominator that decides it