The Champagne Pyramid Is a Spike, Not a Triangle

Market reports usually describe a category as a ladder, and a ladder suggests rungs of similar width. Champagne’s is not a ladder. Measured by what each tier contributes, it is a very broad base with one narrow, extremely valuable spike standing on top of it.

The Comité publishes shipments by quality in both volume and value. Divide one by the other and you get the only structural map of the appellation that exists.

The spike, priced

The per-bottle column is our arithmetic, not a published series: it is each tier’s value-to-volume ratio multiplied by the EUR 24.00 average export price the Comité reports. The method is worth stating because the result is the point. Prestige cuvées clear at roughly EUR 81.91 a bottle — 3.41 times the export average — on 4.6 per cent of the bottles and 15.7 per cent of the money.

And the base is not merely large. It is below average. Brut non-vintage is 77.9 per cent of export volume and 65.1 per cent of value: a ratio of 0.84, or about EUR 20.06 a bottle. Three-quarters of everything Champagne exports clears under the appellation’s own average price, and the average is dragged up by a tier that accounts for one bottle in twenty-two.

Two shapes of the same spike

The Comité splits the export data three ways — the world outside France, the European Union, and third countries — and the pyramid has a different geometry in each.

Inside the European Union, prestige cuvées are 3.4 per cent of volume and 12.5 per cent of value: a multiple of 3.68, the steepest ratio anywhere in the published data. In third countries, prestige is 5.2 per cent of volume and 17.0 per cent of value: a lower multiple of 3.27, but on a much wider base.

That inversion is worth sitting with. The EU buys proportionally less prestige Champagne than the rest of the world, but each bottle it does buy sits higher above its own market average — a small, expensive top on a market that otherwise buys the entry tier. Third countries buy more of the top tier in volume terms and price it, relative to their own average, less aggressively. The premium is not a property of the wine. It is a property of the market that is buying it, and it is thinnest exactly where the volume is growing.

What a spike cannot do

Three constraints follow from the geometry, and they are the reason the category’s forecasts keep disappointing on the value line.

One: the top is too small to move the base. The entire prestige category is 7 million bottles — export shipments, after a 17 per cent fall. Recovering the whole of that loss adds about 1.4 million bottles against a stock surplus of 162.3 million. The top tier is 15.7 per cent of the value and 2.6 per cent of the bottles, and there is no version of “premiumise our way out” that runs through 7 million bottles.

Two: the base sets the price, because the base is where the discount lands. At EUR 20.06 a bottle, brut non-vintage is the tier that competes with Crémant and Prosecco at EUR 5 to 9 and with a spirits serve at EUR 12 to 15. When the average price fell 10.56 per cent in 2025, the fall was not in the tier that costs EUR 81.91. It was in the tier 77.9 per cent of the bottles sit in.

Three: the mix lever is worth more than the growth lever. Move one percentage point of export volume — about 1.5 million bottles — from brut non-vintage to prestige, and the value effect is roughly EUR 94 million, or 1.65 per cent of the appellation’s entire revenue. Nothing else available to Champagne is that leveraged: not a percentage point of volume growth, which at EUR 21.42 is worth about EUR 57 million, and not the yield.
Which is exactly why 2025 hurt. The mix moved the wrong way by 0.9 of a point at the top.

The strategy the shape implies

A triangle implies you can trade a customer down and keep him. A spike implies something harder: the category’s revenue is one product plus a small ornament, and the ornament is losing share while the product is losing price.

That leaves two defensible moves, and they are not “premiumise”.

The first is to protect the base price, because that is where 65.1 per cent of the money is, and a 10.56 per cent fall there is worth more than any plausible gain at the top. That is a channel problem rather than a marketing one: the house sets an ex-cellar price and a three-tier chain sets a shelf price, and the shelf price is what the consumer’s substitute — a EUR 7 Crémant or a EUR 14 cocktail — is measured against.

The second is to make the middle legible, which is what actually moved in 2025. Low dosage and vintage grew because a buyer could check something on the label: 16.1 per cent and 4.8 per cent respectively, against a 17 per cent decline in the tier with no shared definition at all. If the appellation wants a wider top, the instrument is a checkable specification rather than a house’s own claim about its own range.

The romantic picture of Champagne is a pyramid in which value rises as volume narrows, and every rung supports the one above. The published numbers describe a different structure: three-quarters of the bottles sold under average, a tenth of the money concentrated in one bottle in twenty-two, and a competitive set that gets harder the lower you go. Understanding which shape you are actually pricing is the difference between a forecast and a wish.

Companion analysis: Global Champagne Market Report, Size & Forecast 2026–2033.