Wayfair Never Touches the Couch. The Problem Is Everything It Has to Touch Anyway.

You order a sofa from Wayfair, and Wayfair never sees it. No warehouse of theirs holds it, no truck of theirs moves it, no employee of theirs lifts it. The order pings a supplier you've never heard of, that supplier boxes the couch and hands it to a carrier, and the thing arrives at your door with a Wayfair label on a shipment Wayfair never physically touched. For the trouble of standing in the middle — setting the price, owning the website, eating the returns — Wayfair keeps roughly thirty cents on every retail dollar.⁶ On $11.9 billion of revenue, that's $3.6 billion of gross profit.¹ And then it loses money anyway.

The official story is that Wayfair is an asset-light marketplace that has simply never figured out how to make a profit. Both halves are wrong. It isn't a marketplace — it's a first-party drop-ship retailer that buys wholesale and sets its own prices.⁶ And the model itself isn't the thing bleeding. The drop-ship engine throws off cash. What doesn't pencil out is everything Wayfair chose to bolt onto it.

Where the thirty cents actually comes from

Call something a marketplace and you imagine a commission — eBay takes a cut, the seller does the rest. Wayfair is a different animal entirely. It purchases inventory wholesale from about 20,000 suppliers, marks it up to a retail price it sets itself, and pockets the spread when the supplier ships.⁶ The spread is the business. It runs around 30% of revenue and has for years: 28.0% in the bruising inflation year of 2022, 30.6% in 2023, 30.2% in 2024, and roughly 30.2% in 2025 (30.3% in Q4).³⁴¹²¹⁰ That band barely moves, which tells you the core machine is sturdy. The thing people call a 'collapse' in 2022 was a two-point dip from a pandemic peak — not a structural break.

Read that table left to right and the story tells itself. The margin is flat. Adjusted EBITDA went from deeply negative to $743 million in three years.² Free cash flow turned positive in 2023 — improving by more than $1 billion versus 2022 — and stayed there, reaching $329 million in 2025.¹¹² By every measure of the operating engine, Wayfair got better. And the GAAP net loss shrank, but never crossed zero. The gap between 'the engine works' and 'the company is profitable' is the whole story — and it lives below the gross-profit line.

The ad bill that can't be cut

Here is the trap drop-ship retailers walk into, and Wayfair walked straight in. The model has no inventory moat — anyone can list the same supplier's couch — so the only durable asset is the customer who comes back. To get that customer, you advertise. Wayfair spent $179 million on marketing the year it went public on $1.3 billion of revenue; by 2018 it was spending $774 million on $6.8 billion.⁹ By 2024 the advertising line was $1.47 billion — 12.4% of revenue, right inside the 12-14% band it has lived in since the IPO.⁷ That spend doesn't capitalize into anything. It buys this quarter's traffic and then it's gone. Stop spending and the flywheel stops turning — but keep spending and you're handing back nearly an eighth of every dollar before a single other cost is paid. It is a toll you pay to rent demand you never get to own.

And the ad spend only justifies itself if the customer base grows. It hasn't. Active customers sat at 22.4 million at the end of 2023 and 21.3 million at the end of 2025 — flat to slightly down for three straight years.⁴² What's kept revenue from sliding is that each customer spends more: LTM revenue per active customer rose to $586, up 5.6%.² That's a real lever, but a finite one. You can squeeze more from the same people only so far. Eventually the model needs the headcount to grow again, or the ad bill is buying replacement, not expansion.

Turning the cost center into a revenue line

The interesting move isn't on the demand side — it's in the warehouse. CastleGate, Wayfair's logistics network since 2015, now spans 22 million square feet across 60 buildings on several continents.⁵ Conventional wisdom says this contradicts the whole drop-ship thesis: why build warehouses if your suppliers ship for you? Because Wayfair figured out it could charge for them. Suppliers who stash inventory in CastleGate get faster delivery promises — with a growing share of U.S. items carrying speed badges — and Wayfair collects fees for forwarding, storage, and fulfillment.⁵ CastleGate Forwarding volume grew 40% year over year, and long-term inbound supplier commitments are up more than 30% since the start of 2025.⁵ What was a pure cost is becoming a supplier-fee revenue stream and even a third-party logistics service for other retailers. Crucially, it's still a minority: about 25% of revenue ships through CastleGate, while roughly 75% still flows the old drop-ship way, straight from the supplier's dock.⁵ So this isn't Wayfair abandoning drop-ship. It's Wayfair selling the logistics layer back to the suppliers it depends on.

Isn't a model that never earns a GAAP dollar just broken?

The fair objection is blunt: a company that has posted a GAAP profit exactly once since 2014 — briefly, at the pandemic peak — doesn't have a transformation story, it has an excuse. Adjusted EBITDA strips out exactly the costs that are killing the company, and a $3.2 billion long-term debt load¹² isn't a footnote, it's a verdict on a decade of cash-burning. That's a serious read, and half of it is right. The debt is real and the interest is permanent until it's refinanced or repaid. But notice what the engine is now doing despite that weight: $743 million of Adjusted EBITDA and $329 million of free cash flow in 2025 means the operating business is generating cash to service the debt rather than borrowing more to survive.² In 2022 it was burning $1.1 billion of free cash flow.³ That is the difference between a model that's broken and a model that's leveraged. A broken model can't pay its bills. Wayfair's can — it just can't pay its bills and grow its customer base at the same time yet. The whole thesis rests on that 'yet.' If the 21 million customers start climbing again, the fixed costs spread over more revenue and the GAAP line crosses zero. If they stay flat forever, the cash that services the debt is the high-water mark, and the company is permanently stuck just below profitable.

Wayfair built a clever thing: a retailer that sells furniture without ever holding furniture, keeping a thirty-cent spread on a couch it never touches. The cleverness was never the problem. The problem is that to make a stranger choose your couch over an identical one elsewhere, you have to pay to be remembered — every quarter, forever — and to grow fast enough to justify that, you borrowed. The drop-ship model isn't under pressure. The company built on top of it is. And the test isn't whether the engine runs. It does. The test is whether 21 million people who already buy from Wayfair are ever joined by a 22nd million — because that, and not the spread, is the only number that turns the cash flow into a profit.