What a day of inventory really costs you

As long as money cost nothing, carrying inventory was a question of space. Now that money costs something, it's a question of profit. Many businesses crossed that line without ever redoing the calculation — and are still steering their purchasing with a carrying cost estimated when it was half what it is today.

The purchase price isn't the cost#

Why start from the landed price and not the purchase price?#

First correction, and it's a heavy one: the goods you carry aren't worth what you paid the supplier. They're worth what they cost you landed, on the dock, ready to sell.

Depending on geography and routing, goods bought at a hundred routinely cost a hundred and eighteen, a hundred and twenty, sometimes a hundred and twenty-five: main carriage, freight forwarder, duties and import taxes, port handling, insurance. This landed coefficient is the basis of every calculation that follows, and it very often lives in a spreadsheet alongside the system, reconstructed once a year.

If your landed coefficient isn't in your system, everything your system says about your inventory is wrong by about twenty percent.

What it changes on a simple price comparison#

Two suppliers at a hundred and a hundred and three do not compare like that: one ships from a port served twice a week, the other from a platform that imposes a freight forwarder. Landed, they cost a hundred and eighteen and a hundred and sixteen. The cheaper on the catalog is the dearer on the dock — and that is the one that enters your inventory.

The six components#

Annual carrying cost is expressed as a percentage of stock value. You add up:

Typical total: between ten and twenty-five percent a year. In technical distribution, fifteen percent is a reasonable assumption; on short-life products, it's more.

The component everyone underestimates#

The one that surprises most is obsolescence. It isn't only the goods thrown away: it's also those eventually sold at a loss, and those kept while known to be unsellable because nobody wants to sign the write-down.

A worked example, for the order of magnitude#

How do you calculate what one day of inventory costs?#

Take a business carrying forty million euros of stock, with a carrying cost of fifteen percent a year and average cover of a hundred and twenty days.

A million euros of profit for twenty days of cover. That's the figure that changes conversations, because it translates a logistics indicator into a line of the P&L.

Twenty days, on paper, impresses nobody. A million, everybody understands.

And you have to add to that million the cash released, which is a balance sheet effect rather than a profit one: forty million at a hundred and twenty days, brought back to a hundred, is a little over six million euros returning to cash, once and for all.

What the figure should change#

A carrying cost serves three trade-offs, all of them daily:

Is a volume discount always worth taking?#

A supplier offers three percent more to double the order quantity. If that adds sixty days of cover at fifteen percent a year, the discount costs two and a half percent in carrying. Half a point is left. That's a very different proposition from "three percent off".

The clearance threshold#

A dormant item at two hundred days costs you eight percent of its value a year. Losing twenty percent today therefore equals two and a half years of carrying. If it won't sell within two years, clearing is the profitable choice — and that can be demonstrated rather than argued.

Economic order quantity#

The classic trade-off between ordering cost and carrying cost only makes sense with a correct carrying cost. With a five percent figure inherited from 2019, your quantities are systematically too large.

Putting it in the tool rather than in a spreadsheet#

The calculation above works perfectly well in a spreadsheet — once. The problem is that it's useless once a year: it's useful at the moment the buyer approves an order.

What we put into the applications we deliver is one line shown at that moment: this quantity represents N days of cover and will tie up X euros for N days, i.e. Y euros of carrying cost. Nothing more. The buyer adjusts, or doesn't, but they decide knowing.

That's the difference between an indicator and a tool: an indicator is consulted, a tool shows up by itself at the moment the decision is made.

How we make that cost visible#

The calculation above is worthless if someone has to redo it by hand every quarter. What we put in place is the permanent counter: receipts, issues and thresholds per SKU, with the holding cost updating itself as inventory moves.

The output is a list, not a dashboard: anything falling below its limit triggers a supplier order already drafted, and anything sitting above its normal cover surfaces with its annual cost beside it. The page describing this module shows what that looks like in use.

Cost of stock: what to remember#

Start from landed value, coefficient included, not the catalog purchase price. It's the most frequent and the largest error.

Add up the six components rather than reusing a percentage you heard somewhere. Yours has probably risen since you last worked it out.

Then convert your cover into euros per day. One day of inventory, in your business, is worth a precise amount. Until that's written down somewhere, nobody in the company is really making the trade-off.