Editorial Article
Why a USD 528 Billion Market Can't Sell Direct
The one channel that pays the producer directly moved 0.28 per cent of the world’s wine last year — and it shrank 15 per cent
World wine exports in 2025 were 94.8 million hectolitres for EUR 33.8 billion, according to the OIV. That is EUR 3.57 a litre at the winery gate. The market this site sizes is USD 528.4 billion over 208.0 million hectolitres — USD 25.41 a litre at the shelf.
Put those two side by side without converting the euro into dollars and the traded value of all the wine that crosses a border is 6.40 per cent of the market we size. Freight, duty, distributor margin, retailer margin and the glass are all outside the export number and all inside the 528.4. Roughly ninety-three per cent of what a consumer pays for wine is not paid to a producer.
That is the structure the category lives inside. The obvious question is why nobody bypasses it. The answer, in twelve months of data from the only market that measures the attempt properly, is that somebody does — and it is smaller than anyone outside the trade expects.
What the direct channel actually moved
The 2026 Direct-to-Consumer Wine Shipping Report, published on 27 January 2026 by Sovos ShipCompliant and WineBusiness Analytics, compiles anonymised shipment data from more than 1,300 US wineries covering more than 27 million shipments in 2025, modelled against a database of 11,165 US wineries. It found the most difficult year since the report began in 2010:
| US winery DtC shipping, 2025 | Change |
|---|---|
| Volume | −15% (−967,000 cases) |
| Value | −6% (more than −USD 230m) |
| Average price per bottle shipped | +11% |
Those are the reported percentages and the loss amounts. Working back from them, the channel shipped roughly 6.45 million nine-litre cases worth roughly USD 3.83 billion in 2025. Those two totals are ours, derived from the published deltas, and they carry the imprecision that derivation implies.
Now put them against the world series.
| US DtC winery shipping, 2025 | World wine market, 2025 | Share | |
|---|---|---|---|
| Volume | 0.58 m hl | 208.0 m hl | 0.28% |
| Value | USD 3.83 bn | USD 528.4 bn | 0.73% |
Under three-quarters of one per cent of world wine value, and barely a quarter of one per cent of its volume. In the largest premium wine market in the world, in the channel that exists specifically to let a winery sell to a consumer, on a data set covering more than a thousand producers and twenty-seven million parcels.
The caveats are real and they cut both ways. The report excludes purchases picked up or carried out at the winery, so it understates direct sales. It covers one country. And our denominator is every channel in every country, which is why the share is small. Even allowing generously for all three, the order of magnitude does not move: direct wine in the United States is a sub-one-per-cent channel.
USD 49.55 a bottle against USD 19.06
USD 3.83 billion over roughly 77.4 million bottles is USD 49.55 a bottle. The world average, USD 528.4 billion over 208.0 million hectolitres, is USD 19.06.
The direct channel’s average bottle is 2.60 times the world average. And that is the whole explanation. Direct-to-consumer is not expensive to serve because parcel carriers charge a lot. It is expensive per bottle because direct is where the top of the market sells — the wine that can carry a shipping cost and a story at the same time. Nobody ships a commodity.
Which produces the finding that matters, and it is not a happy one for anyone who thought distribution was a technology problem. The channel with the best economics for a producer is the channel with the worst scale, and in 2025 it contracted fastest of all. Value fell 6 per cent while volume fell 15 per cent, which means the price per bottle rose 11 per cent — the same value-over-volume mechanism the whole market runs, executed in a channel with a tenth of the volume.
Why the structure is not an accident
In the United States, wine normally reaches a consumer through a three-tier system: producer to distributor to retailer, with the distributor licensed by the state and the state’s alcohol rules deciding what a producer may do directly. A producer therefore sells to a wholesaler at a wholesale price and the wholesaler decides what reaches a shelf. Everything the producer spends on brand — the vineyard story, the critic score, the label — has to be paid for out of the first of those three margins.
Direct shipping changes exactly one thing: it converts the distributor’s margin into a producer’s margin. That is why it is worth 2.60 times the world average per bottle despite the freight. It is also why it cannot scale. A distributor carries a portfolio across thousands of accounts and needs throughput; a winery shipping direct needs a consumer who already knows what it wants. Direct is a channel for brands that have been built by something else.
Europe’s version of the same constraint is grocery concentration rather than a statutory three-tier. The mechanism is identical: the party that owns the shelf owns the margin, and the number of parties that own a shelf is small.
The consequence for a market that grows on price
Wine grows 3.81 per cent on volume falling 1.19 per cent, because realisation rises 5.06 per cent a year. Every point of that growth is delivered at the moment of sale. A category that does not own the sale is a category that negotiates for its own price increase with somebody who would rather it did not happen.
That is the mechanism behind the finding we published in What a Wine Company Actually Earns: the largest listed pure-play raised revenue per case 10.2 per cent, and it did so on brands strong enough to be worth a distributor’s shelf space. Producers without that leverage do not get the 10.2. They get whatever the shelf will pay, which in a falling-volume market is a negotiation going the wrong way.
Which is why the consolidation in The Sale That Made a Consolidator Out of a Debt Stack is buying scale rather than margin. At A$46.88 a case, the consolidator is not trying to win the shelf argument. It is trying to be large enough that the shelf has to call it.
What we are not saying
We are not saying direct is the future of wine. The evidence from 2025 is the opposite: the channel shrank 15 per cent in volume in the largest market that measures it, and it has now fallen for several consecutive years. We are not saying USD 528.4 billion is unreachable — it is reached, by retailers.
We are saying that a market of this size, growing on price, hands most of its value to the party that owns the point of sale, and that the one channel built to change that is worth under three-quarters of one per cent of the total. Any strategy for a wine business that begins with “cut out the middleman” is beginning with a channel that is 0.73 per cent of the market and getting smaller.
More in this series: What a wine company actually earns · The sale that made a consolidator out of a debt stack · The wine market grows because the bottle got dearer.