Cash flow

When stock stops being worth what you paid for it

Two decisions hide inside one uncomfortable feeling. One is about a number in your accounts, the other is about a price on a shelf, and doing one does not do the other.

Marcus VolstedCo-founder, Restocio Published 10 October 202615 min read

Every source on this page was opened and read, and each one is dated in the Sources section at the foot.

Write stock down when you no longer believe it will sell, in the ordinary run of business, for more than it cost you landed plus what it will cost to sell. That is a decision about the number in your books, and it moves no cash. Whether to cut the price on the shelf and get some cash back is a separate decision, and you can take either one without the other. The dead stock post on this site is about the shelf: how to age it, bundle it, discount it and get some cash back. This page is about the book value, and about what a number you have stopped believing has been doing to every margin figure you looked at while it sat there.

Two decisions that arrive as one feeling

The feeling is the same for both. A product has slowed, the units are still there, and the line on the valuation report has started to look like a claim rather than a fact. Two different actions follow from that one feeling, and merchants tend to take one and assume the other has been dealt with.

	A write-down	A clearance sale
What it changes	The value the stock is carried at in your accounts	The price a customer pays
Book value of the stock	Falls, to what you now expect to get for it	Unchanged until the units actually sell
Cash	None moves. The money left when you paid the supplier	Some comes back, less than you paid
Gross margin	Takes the loss now, in the period you decide it	Takes the loss as each unit sells below cost
Who sees it	Your accountant, and whoever reads your accounts	Every customer, and every competitor
Can be done without the other	Yes. You can hold the shelf price and still carry the stock at less	Yes. You can discount and still carry the units at full cost until they sell

Read the cash row twice. A write-down does not release a single krona. It changes what you say you own, which is a different thing from having dealt with it. The money went out when you paid the supplier, months before the boat landed, and only the shelf gets any of it back. The reverse holds too. A clearance sale does not fix the books on its own, because until the units sell they sit in your stock figure at full cost, and if the sale clears half of them the other half is still there at a number you have already stopped believing.

The two decisions also sit with different people. The shelf price is yours and your marketing's. The carrying value is yours and your accountant's, inside whatever rulebook your company reports under, and that is where the rest of this page stays. Nothing on it is accounting or tax advice. It describes the decision, so you know what to take to your accountant.

What does a write-down actually test?

Every rulebook we have read states the same test in its own words. The international standard, IAS 2, puts it in one line: "Inventories are measured at the lower of cost and net realisable value." The same page defines net realisable value as "the estimated selling price in the ordinary course of business less the estimated costs of completion and the estimated costs necessary to make the sale". Both read on 27 September 2026.

For a shop, that reads as follows. For each product, take what you paid to get it onto the shelf, and take what you honestly expect to sell it for minus what that sale will cost you. Carry it at whichever is lower. "In the ordinary course of business" is doing real work in that definition. It is the price the stock will fetch as a going business sells things, not a fire-sale figure and not the price you would like to get if the season came back.

The three Nordic rulebooks we already quote for stock valuation say the same thing. Swedish law: "Omsättningstillgångar ska, om inte annat följer av 10, 12, 13 a, 14 a eller 14 e § tas upp till det lägsta av anskaffningsvärdet och nettoförsäljningsvärdet på balansdagen", with nettoförsäljningsvärde defined in the same paragraph as "försäljningsvärdet efter avdrag för beräknad försäljningskostnad". Norwegian law: "Omløpsmidler skal vurderes til laveste av anskaffelseskost og virkelig verdi." Danish tax guidance comes at it from the other end and calls it ukurans, which is "blot et spørgsmål om at ansætte et varelager, som af den ene eller anden grund er faldet i værdi, til en realistisk værdi". All three read on 27 September 2026, and they are quoted so you can see that the principle is common, not so that one country's wording gets applied in another.

What none of them tells you is the date. Swedish law says balance date, the standard says the period in which the write-down occurs, and both leave the question that actually matters to you, which is when a merchant should start believing a lower number, entirely to the merchant. That is a judgement rather than a rule, and it is what the rest of this page is about.

Cost is half the test, and it is usually wrong first

Lower of cost and what it will fetch is a comparison, and on most Shopify stores one side of it was wrong before anyone got round to the other. Shopify's cost per item is, on most importers' files, the supplier price with no freight in it. The ten-minute COGS audit shows you how to check that on one SKU, and stock valuation for your accounts covers why the landed figure is the one the books should carry. For the write-down test, what matters is the direction of that error, and it only ever runs one way.

Take the audit's own numbers. The field says 11.00 and the real landed cost is 12.58. Now say the product has slowed, and your honest expectation is that it clears at 14.00, with a marketplace fee and a free-shipping subsidy of 2.00 per unit on the way out. Net realisable value is 12.00.

Cost you compare against	Net realisable value	Lower of the two	Write-down per unit
The Shopify field, 11.00	12.00	11.00, the cost	None. The stock looks fine
The landed cost, 12.58	12.00	12.00, the value	0.58, and the stock is already worth less than you paid

An understated cost sits below net realisable value for longer, so the write-down arrives later than it should, or never. On 800 units that is 464 of value already gone that no report on the store has mentioned. The error is not noise, either. Every missing cost line pushes the same way, towards a stock figure healthier than the shelf, and the further the goods travelled the harder the push.

When should you write down inventory?

There is no calendar for this, and anyone selling you one is guessing. What there is instead is a short list of signals that each say the same thing: at the price the stock will really fetch, by the date it has to be gone, the cost is no longer the lower number.

Signal	What it says about value	Shelf decision	Book decision
Days of cover longer than the time the product has left	Part of the stock will not sell at the ordinary price before the season ends or the model year turns	Stop reordering, start bundling	Estimate what the surplus will fetch after the deadline, and carry that part at it
The price that would clear it by the deadline is below landed cost	Net realisable value is already under cost, whether or not you have cut the price yet	Decide whether to take the cut now or later	The write-down is due now, not when the sale runs
The supplier has a newer version, or you do	The old version's ordinary price no longer exists	Clear it before the replacement lands, if you can	Value it against the price it will fetch next to the new one
One variant has not sold in months while its siblings do	That size or colour has a lower ordinary price than the product as a whole	Check at variant level, which is where dead stock hides	Write down the variant, not the product
Damaged or returned units you would not sell as new	Their ordinary price is the seconds price, or nothing	Sell as seconds, or dispose of them	Carry at the seconds price, and at zero if you will not sell them at all

The first row is the one that catches importers, because the days-of-cover number is already in front of you on any replenishment view and nobody reads it as a valuation signal. A product with 400 days of cover and a season that ends in 90 has roughly three quarters of its units on the wrong side of the deadline. That does not make all of them worth less. The units that will sell before the deadline are worth their ordinary price, the ones after it are worth whatever a post-season price is, and the honest book value is the sum of the two. Days of stock works the first number, and the aging bands on the dead stock post decide what to do on the shelf about each band. The book question sits underneath both.

Whether a tax authority also expects the value to fall, and how fast, is a separate question that only your own country answers. Denmark's answer is the most concrete we have found, and it is a tax rule for Danish companies rather than a general one. Den juridiske vejledning says that where no other and more reliable basis for the valuation exists, a company may, under a ruling from the former Statsskattedirektorat, value technically or economically obsolete goods at 50 per cent in the first year, 25 per cent in the second and 0 per cent in the third of the value they were carried at when the first year began, or the lower value the varelagerlov allows, on condition that the goods are teknisk/økonomisk ukurante and that their turnover rate has fallen substantially against the normal rate for an equivalent product. Read on 27 September 2026, version 3.17 of the guidance. Whatever your jurisdiction, the shape of it is worth noticing. An authority's own fallback for a forgotten product is that half the value is gone inside a year.

What does a write-down do to the margins you have been reading?

Here is the part that surprises people, and it should not. IAS 2, on the same page, says that the amount of any write-down "and all losses of inventories are recognised as an expense in the period the write-down or loss occurs". The loss goes through your results in the period you decide it, in one lump.

Follow what that means for the months before. While 800 units that were worth 12.00 sat at 12.58, every report you looked at carried them at full cost. Gross margin on the products that did sell was reported without any of that loss in it. The stock figure said you owned 10 064 of that product when you owned 9 600. The year's profit was, quietly, 464 too high, and it stayed too high until the day you wrote the stock down, at which point the whole 464 arrived at once, usually at year end, usually in a conversation that opens with why the margin is lower than the dashboard said.

A write-down is not a bad month. It is several flattering months being corrected late, which is why the decision is worth making early and per product rather than once a year for the whole warehouse: the sooner a slow product is carried at what it will fetch, the closer the margin you read each month is to the margin you actually earned.

Shopify's own reports never see any of this. Its month-end inventory value is, per the Help Center, read 27 September 2026, "the cost per item multiplied by the ending quantity amount for each product variant option", and the report has one cost column, which is that field. There is nowhere in it for a lower figure. So after a write-down the store's stock value and the accounts differ by the amount you wrote down, on top of the freight gap they already had, and that difference is correct rather than a mistake. The two numbers answer different questions.

Where the stock that needs writing down comes from

Almost always from an ordering habit rather than a demand surprise, and the habit is visible months before the write-down is. A trailing sales rate keeps recommending a full reorder into a falling curve, so every cycle buys a little more of something that is selling a little less, which is its own post. A minimum order quantity accepted without the arithmetic creates the surplus on the day the order is placed. Both show up first as days of cover climbing while sales stay flat, and the share of your stock value sitting above your cover threshold is the earliest number that predicts a write-down. Watching it is the replenishment job, and it is the one part of this page where the fix costs nothing. The cheapest write-down is the order you did not place.

What Restocio gives you for this, and what it does not

Restocio is purchasing software, so it holds the cost side of the comparison and none of the value side. Its Stock valuation report carries every product at a moving average of what you actually paid, landed, freight included, blended at each delivery and traceable to the receipt that produced it. Stock on a date tells you what the shelf was worth at the end of any recorded day at that cost, which is the figure the write-down test starts from. Cost of goods sold for a period values every unit sold at the cost that applied on its sale day, and its Left without a sale count names the units that left the shelf with no sale behind them, matched against what was logged as lost, damaged, found or recounted. That is where a physical write-off of units shows up, and units counted in as damaged stay in the shelf count, and in the value, until you write them off. Days of cover per product is on the Replenishment page, in the column labelled Days left.

Now the limit, plainly. There is no place in the app to record that a unit is worth less than you paid for it. No net realisable value, no write-down, no obsolescence flag, no aging bands, no markdown engine. A product you have written down in your accounts still shows in Restocio at its landed cost, and from that day the app's stock value and your accounts differ by exactly your write-down, on purpose, because the app's number is what you paid and stays traceable to the delivery. The comparison, and the judgement about which price the stock will really fetch, are yours and your accountant's. Duty is not in the cost yet either, and both reports say so on screen.

And the obvious one. We build purchasing software, not accounting software, and nothing on this page is accounting or tax advice. The rules above are quoted with their country and read date attached so you can check them, and they are not each other's rules. A Danish tax rule of thumb is not a Swedish annual-accounts rule, and neither is yours unless you report there. Whether a product needs writing down, by how much, and whether your tax return follows your accounts on it, are questions for your accountant in your own jurisdiction.

Common questions

When should I write down inventory?

When you no longer expect it to sell, in the ordinary run of business, for at least what it cost you landed plus what the sale will cost. The test in IAS 2, and in the Swedish, Norwegian and Danish rules this site quotes, is the lower of cost and what the stock will fetch net of selling costs. The date is a judgement. The signals are days of cover longer than the product's remaining life, a clearing price under landed cost, a replacement version arriving, and a variant that has stopped while its siblings sell.

Is writing down stock the same as putting it on sale?

No. A write-down lowers the value the stock is carried at in your accounts and moves no cash. A clearance sale lowers the price a customer pays and brings some cash back. You can do either without the other: carry stock at less while holding the shelf price, or discount it while the units sit in the books at full cost until they sell. Both usually need doing, and they are decided by different people.

Does a write-down free up cash?

No. The cash left when you paid the supplier, months before the goods landed. A write-down changes what you say you own. The only things that get money back are selling the units or sending them back to the supplier, and the only thing that stops more of it going out is not placing the next order. Treating a write-down as having dealt with slow stock is the most common way it gets forgotten a second time.

Why does a write-down hit my margin all at once?

Because the loss is recognised as an expense in the period the write-down occurs, which is how IAS 2 puts it. The stock was losing value for months while every report carried it at full cost, so the margins you read in those months were flattered, and the correction lands in one lump on the day it is admitted. Writing slow products down early and per product spreads that correction over the months it actually belongs to.

What is net realisable value?

The estimated selling price in the ordinary course of business, less the estimated costs of completion and the costs necessary to make the sale, in the words of IAS 2. For a shop, that is the price you honestly expect the stock to fetch as a going business sells things, minus the fees and the shipping subsidy it takes to move it. Swedish law calls it nettoförsäljningsvärde and defines it the same way.

Does Restocio write down stock?

No. Restocio carries every product at the moving average of what you paid, landed, and it has no field for a value below that. Its Stock on a date and Cost of goods sold reports give you the cost side of the test at any recorded date, and its Replenishment page gives you the days of cover that signal a problem. The value side, and the write-down itself, live in your accounts, so after one the app's stock value and your books will differ by that amount.

Sources

Every source below was opened and read on the date shown against it. Rules, rates and schedules change, so check the current page before you act on a number that matters. If anything here is out of date or wrong, tell us and we will correct it.

Marcus Volsted

Co-founder, Restocio

Marcus co-founded Restocio and works on it daily with a Swedish importer who plans their purchasing in it every working day. Restocio is built in Sweden by two founders, one Swedish and one Danish, and Marcus is the Danish one. Why we are building it.