ASOS turnaround: the operations behind the profit

ASOS lost around £430m of revenue in FY25. Adjusted EBITDA went up £52m.

That is not a typo, and it is not a fluke. It is what happens when a business stops chasing revenue that was never making money and fixes the operation underneath it instead.

Revenue fell from around £2.9bn in FY24 to £2.46bn in FY25, down about 15% (ASOS FY25 Annual Report). Adjusted EBITDA went the other way, from £80.1m to £131.6m, up 64% (ASOS FY25 Annual Report). Gross margin rose 370 basis points to 47.1% (ASOS final results, November 2025).

Most of the commentary on ASOS is about fashion, competition from Shein and whether the brand still matters to 25 year olds. Fair questions. But they miss what actually moved the P&L. The turnaround has been operational: how much stock the business owns, how quickly it buys, where it stores product, and how much infrastructure sits underneath each pound of revenue.

This is a breakdown of those four things, with the filed numbers.

What was actually wrong

By the end of FY22, ASOS was holding £1.1bn of stock. That figure had doubled during the pandemic, and in the company's own words it came from disruption and "poor commercial practices which led to the build-up of old and aged stock" (ASOS final results, November 2025).

Stock at that level does three things to a fashion business, and none of them is good.

It ties up cash that could be doing something else. It creates markdown risk, because yesterday's fashion only gets cheaper. And it needs somewhere to live, which means warehouse space, and warehouse space is fixed cost you pay whether the product sells or not.

ASOS had built a large warehouse network to hold all of it. It had a buying model that committed cash months before a customer had said what they wanted. And it was discounting heavily to clear what that model produced.

So the business had a lot of revenue, and a lot of complexity built around revenue that was not earning enough to justify the cost of carrying it. The management team's answer was to go backwards first. Clear the stock, shrink the footprint, rebuild the buying model, and accept a smaller top line while doing it.

The stock

The stock reduction is the foundation everything else sits on, so it is worth seeing the full curve rather than just the endpoints.

Source: ASOS FY25 Annual Report and Accounts.

That is a reduction of around 60% in three years. ASOS describes FY23 and FY24 as "peak pain", the period where it cut intake and discounted hard to clear old stock (ASOS final results, November 2025). The revenue decline in those years was partly the cost of doing that.

The cash effect is the part founders should pay attention to. In FY25 alone, inventory reduction released £109.7m of cash (ASOS FY25 Annual Report). That is working capital that had been sitting on shelves, now back in the business, in a year when net debt fell by over £110m to £184.7m and free cash flow turned positive at £14.1m.

The reduction has continued into FY26. Inventory fell another 10% in the first half, taking the three-year reduction past 60% (ASOS interim results, April 2026).

Less stock is not a target in itself. The point is that the stock ASOS still holds is fresher, sells through faster and needs less markdown. That shows up directly in gross margin, which has now improved for eight consecutive quarters (ASOS interim results, April 2026).

The warehouses

When you cut stock by 60%, you no longer need the buildings you bought to hold it.

ASOS has reduced its warehouse footprint by more than 50% since FY21 (ASOS FY25 Annual Report). Two closures did most of the work.

Lichfield, the second UK fulfilment centre, was mothballed in FY24. Atlanta followed in FY25, with US customers now served from the automated UK site in Barnsley plus a smaller, more flexible new site in Dallas (ASOS final results, November 2025).

The Atlanta decision is the clearest example of the logic. ASOS had built a large US distribution centre for a US business that turned out to be smaller than planned. Closing it is expected to save £10m to £20m a year in annualised costs (ASOS final results, November 2025). By the first half of FY26, the company's absolute fixed cost base was down 12% year on year, "primarily reflecting the FY25 mothballing of the Atlanta warehouse" (ASOS interim results, April 2026).

Two other operational numbers sit alongside the footprint change. Supply chain costs fell around 20% year on year in FY25, through warehouse rationalisation, renegotiated distribution contracts and reducing the causes of unnecessary returns (ASOS final results, November 2025). And distribution and warehousing cost-to-serve came down around 3 percentage points over two years (ASOS full year trading update, September 2025).

Warehouses are the most visible form of infrastructure sitting underneath revenue. ASOS had too much of it for the size of business it was becoming, and it removed it.

The buying model

Clearing old stock only works once. What stops it building up again is how you buy.

ASOS has moved a growing share of its own-brand range away from large speculative buys and onto Test & React. The mechanics are simple. Buy a small quantity. Put it on site. See what actually sells. Then chase the winners with larger orders.

In FY25, Test & React reached more than 20% of own-brand sales (ASOS full year trading update, September 2025), with a medium-term target of around 30%. Alongside it, ASOS says it has accelerated time to market by around 30% (ASOS FY25 earnings call, November 2025).

The commercial effect of buying this way is that less cash is committed before the customer has voted. Forecasting risk drops because you are forecasting off real sell-through rather than a guess made months earlier. Markdown drops because you are not clearing stock you should never have bought.

Two supporting changes sit around the buying model. ASOS introduced targeted measures on persistently high return behaviours, which it says reduced returns and strengthened basket-level profitability, while keeping free returns for most customers (ASOS final results, November 2025). And profit per order rose 30% year on year (ASOS final results, November 2025).

That is the number to hold onto. Fewer orders, but each one worth more.

Flexible fulfilment

The fourth change is about what ASOS chooses not to own at all.

Under its flexible fulfilment models, partner brands hold and fulfil more of their own stock while ASOS still sells it through the platform. Some of that is Partner Fulfils, where the brand ships direct. Some of it is the AFS model, where ASOS handles the customer side but the inventory sits with the brand.

By the end of FY25 these models had scaled to more than 10% of third-party GMV, including the transition of Inditex onto AFS, and around 100 new partner brands launched during the year (ASOS final results, November 2025).

The trade is more assortment for the customer with less working capital for ASOS. The business gets breadth without buying the stock to support it.

There is an honest wrinkle here that ASOS reports rather than hides. Because the variable cost to serve stays with ASOS under AFS, growing it adds structurally to cost-to-serve, around 60 basis points in H1 FY26 (ASOS interim results, April 2026). ASOS's position is that the net EBITDA effect is still positive because of the margin and availability benefits. That is a reasonable read, but it means flexible fulfilment improves the balance sheet more cleanly than it improves the cost line.

What it added up to

Put the four changes together and the FY25 numbers make sense.

Revenue down around 15%. Gross margin up 370 basis points to 47.1%. Adjusted EBITDA up 64% to £131.6m, a margin of 5.3%. Stock down to £402m. Net debt down over £110m. Free cash flow positive for the first time in the turnaround. Profit per order up 30% (ASOS FY25 Annual Report and final results, November 2025).

And the direction has held into FY26. In the 26 weeks to 1 March 2026, revenue fell another 14% on an adjusted basis to £1.11bn, but gross margin reached 48.5%, up 330 basis points, and adjusted EBITDA rose 50% to £64.0m from £42.5m (ASOS interim results, April 2026). Over three years, adjusted EBITDA is up almost 14 times.

The first half also carried something worth noting for anyone selling into the US. ASOS recognised £7m of IEEPA tariffs in the period and has started the process of claiming refunds following the court rulings (ASOS interim results, April 2026). Even at ASOS's scale, tariff exposure is now a line item to manage, not a footnote.

Guidance for the full year is unchanged: gross margin of 48% to 50% and adjusted EBITDA of £150m to £180m (ASOS interim results, April 2026).

What it cost

None of this is free, and a proper read of the turnaround includes the other side.

Active customers fell 8% to 6.5 million in FY25 (ASOS FY25 Annual Report). Revenue has now declined for several consecutive years. Statutory losses remain large because of property and impairment charges from the very warehouses being closed. Net debt rose again in H1 FY26 to £294.9m, mainly from non-cash interest on the convertible bonds (ASOS interim results, April 2026). And cost-to-serve as a percentage of revenue rose 90 basis points in the first half, because a smaller revenue base has less to spread fixed costs across (ASOS interim results, April 2026).

There are green shoots. New customer growth in the top four markets turned positive, and March 2026 was the first month of new customer growth since September 2021 (ASOS interim results, April 2026). But the business that comes out of this is smaller than the one that went in, and it will need to grow again for the model to prove itself fully.

The honest summary is this. ASOS has not fixed the business by selling more. It has fixed how much stock it owns, how quickly it buys, where it stores product and how much infrastructure sits underneath each pound of revenue. The growth question is still open. The operational one is largely answered.

What this means if you run a scaling brand

You are not ASOS. But the four problems ASOS had are the four problems most brands have once they pass a few million in revenue, just at a smaller scale and usually with nobody looking.

Your stock is cash. Every pound of inventory that sits for longer than it should is a pound you cannot spend on the products that actually sell. ASOS released £109.7m in one year by getting this right. At your size the number is smaller, but the ratio is often worse, because founder-led buying tends to over-order the range and under-order the winners. If you do not know your weeks of cover by SKU, start there. We wrote a guide on how to calculate weeks of cover.

Your warehouses are a choice, not a given. Footprint is fixed cost that you decide to carry. Most brands scale into more space, more 3PL locations and more contracts without ever asking whether the volume still justifies them. ASOS took out over half its footprint. You may only need to renegotiate one contract, but the question is the same. If you suspect your 3PL setup has outgrown the business, here are the signs it is time to leave your 3PL.

Buy smaller and chase what sells. Test & React is just a disciplined version of what every good buyer knows. Commit less before the customer votes, then move fast on the evidence. It needs a planning process that can actually see sell-through by SKU, which is the part most brands are missing. Our note on inventory forecasting methods covers the practical options.

Do not own everything you sell. Flexible fulfilment is ASOS's version of a question every brand should ask: which parts of the range should I hold stock for, and which should I let someone else carry? At a smaller scale that might be dropship for the tail, consignment for a category, or simply not stocking the slow lines at all.

None of this needs a £2bn revenue base to work. It needs someone to own the operation, look at the numbers, and make the unglamorous decisions about stock, space and buying that founders rarely have time for once the business gets past the point where they can hold it in their head.

That is usually when we come in. If you want a second pair of eyes on your stock, your warehouses and how you buy, see how we work.