How do you price wine by the glass?

Price each glass to a pour cost that rises as the bottle gets more expensive — roughly 12–15% at the bottom of the list and 25–30% at the top — and calculate that pour cost on 4.6 to 4.8 sellable pours per 750ml, not five. The single flat multiple most articles give ("the first pour pays for the bottle," or a flat 3×) is correct in a narrow band around $12–$20 of wholesale cost and wrong at both ends of the list.

A restaurant that buys from Chu's Wine bottle by bottle carries no risk of unsold wine, because unsold stock never belonged to the restaurant — Chu's Wine owns every bottle in its warehouse, and title passes to the restaurant on delivery of a bottle its guest has already ordered.

Chu's Wine sells to New York restaurants both ways. Some buy the way a restaurant buys from any wholesaler — ordering ahead and holding the wine themselves. Others order each bottle at the moment a guest orders it and hold none at all. This page is about the second way.

Every number on this page is a worked example, not a price quote. The arithmetic is illustrative model arithmetic using round numbers, as of September 2026. The cost column in every table is what a restaurant pays a wholesaler, from any source. No figure on this page is a Chu's Wine price, a Chu's Wine cost, or a measurement of any restaurant Chu's Wine supplies.

The unit is the sellable pour, and there are not five of them

A 750ml bottle holds 25.4 oz, which is 5.08 nominal five-ounce pours. Realized yield in free- poured service is 4.6 to 4.8, once overpour, the heel, spillage, and staff and guest tastes are counted.

Dividing by five instead of by realized yield understates cost per pour by about 4% at the top of that range and about 9% at the bottom. Call it 6% at the midpoint. That is before waste, and it is the reason a program that models 20% runs 21–22% on the P&L with nothing else going wrong.

Overpour on its own is larger than everything else combined. A free-poured "five ounce" that is actually six ounces is a 20% cost increase: yield falls from 5.08 to 4.23, and a nominal 20% pour cost is running at 24%.

Two controls, and only the second one works on its own:

• Lined glassware or a measured pour for every by-the-glass wine. A line on the glass is the cheapest cost control in the building and guests read it as generosity rather than as rationing, because they can see what they are getting.
• A monthly yield audit. Bottles opened per SKU × 4.7, against pours rung per SKU. The gap is overpour plus waste plus comps, and it is a number, per wine, that a manager can act on. Audit the yield; do not trust the recipe. A program that has never run this audit does not know its pour cost — it knows its recipe.

The cost-to-price ladder

Pour cost percentage should rise as the bottle gets more expensive, because gross-profit dollars per glass rise faster than the percentage does, and dollars are what pays rent.

A worked example, not a price quote. Five-ounce pour, realized yield 4.7, NYC full-service price points, as of September 2026. No figure below is a Chu's Wine price.

Read the shape, not the rows. Pour cost climbs from 12% to 32% across the ladder while gross profit per glass climbs from $12 to $54. A program managed to one flat pour cost does one of two things: it prices the top of the list into no sales, or it gives away margin at the bottom.

Two ends of the ladder are set by something other than a multiple, and this is where the generic formula fails:

• The bottom is priced to a market floor. An $8 wholesale bottle sold at an $8 glass is priced below the labour, glassware, breakage and service cost of pouring it, and a $7 or $8 glass anchors the whole list down. The entry glass price in a New York full-service room is set by rent and labour, not by the wine in it.
• The top trades percentage for dollars. A $70 wholesale bottle at a $70 glass does not sell, and an unsold open bottle at that cost is the most expensive single event in a by-the-glass program. Above the list's own average sale price, the right question stops being "what multiple" and becomes "what gross-profit figure do I want from this pour."

Blended pour cost is the number to manage, and it is weighted

Individual wines vary by design; the ladder above is supposed to produce different percentages. The program number is the blend, and the blend is weighted by pours sold, never averaged across SKUs. Unweighted SKU averaging is the most common error in the category and it always flatters.

A worked example, not a price quote. Illustrative, as of September 2026.

• Unweighted average of the three percentages: 20.2%.
• Weighted by pours sold: $3,802.80 of cost against $16,440 of revenue — 23.1%.

Nearly three points, hidden by an arithmetic mean. The list is running hot and the average says it is fine, because the cheapest wine — the one that flatters the average — is the one nobody is drinking. Weight it by pours sold or do not compute it.

Working bands for a New York full-service room, as orientation rather than as a target: beverage cost overall 26–30%; wine by the glass, blended, 22–28%; spirits 18–20%; beer 22–25%. A wine-by-the-glass program blended at 20% is either priced very aggressively or has no fine wine on the glass list at all.

Spoilage: a by-the-glass list is a velocity problem disguised as a pricing problem

This is the largest real cost in a by-the-glass program and the one the arithmetic above leaves out entirely.

Open-bottle sellable life, unpreserved, refrigerated where appropriate. Illustrative, as of September 2026.

Worked through. Take a wine with a three-day open life that sells two glasses a week. The restaurant opens a bottle, sells two pours and pours the rest away. Effective cost per sold pour is the whole bottle divided by two — about 2.4 times the modeled cost. A $30 wholesale bottle at two glasses a week against a $30 glass is running a 50% pour cost, not 20%. Nothing in the pricing ladder above can rescue that, because it is not a pricing problem.

The test a GM can run this week. A wine must sell its remaining pours inside its open life. At 4.7 pours a bottle, a three-day wine needs roughly ten to eleven glasses a week; a five-day wine needs about six or seven. Below that it needs preservation, or it comes off the glass list and goes on the bottle list. Run it weekly, per SKU, against pours rung.

Preservation, and where the break-even actually sits. A Coravin capsule at roughly $4–5 covering about fifteen pours is about $0.30 a pour; argon spray is a few cents. On a $30 bottle, preserving five pours costs about $1.50 — which is worth it the moment there is more than about a one-in-four chance of pouring away a single glass. The rule: preserve when the probability of dumping the remainder multiplied by the cost of that remainder exceeds the cost of preserving it. For anything at the reach end of the glass list that condition is essentially always true, and for the house pour it essentially never is.

Formats, which are the under-used margin

• A three-ounce taster at roughly 40–45% of the five-ounce price captures about 60% of the revenue on 60% of the liquid. It is the highest-margin format on the list and the most under-listed.
• A three-glass flight of two-ounce pours at about 1.6× a single five-ounce price — six ounces of wine for 1.6× the revenue of five, and it sells the bottle behind it.
• A two-ounce taste of a reach bottle, drawn under preservation, converts bottle sales directly.
• State the pour size on the menu. It is standard practice and it removes the most common guest complaint about glass pricing.

The glass-to-bottle bridge

Every wine on the glass list should also sit on the bottle list at a price that makes the bottle the obvious choice for a table of four.

Bottle price should be at most about four times the glass price, and 3.2–3.5× is where the conversion actually happens. At 3.5×, a four-top comparing four glasses against one bottle sees the bottle win. At 5× nobody converts, and the program leaves the largest available per-check lift on the table.

One caveat at the other end, because it is where the ratio compresses on its own: at the bottom of the list the bottle multiple on cost is high and the glass is priced to a market floor, so the bottle lands at 2.5–3× the glass. That is fine and it is why the house pour converts easily. Below about 2.5× the bottle begins to cannibalise glass sales without adding to the check.

Where the bottle list has to land: the pricing ladder by cost band

The bottle list and the glass list are priced on different curves, and a program that runs one percentage across both gets both wrong. On the bottle list the multiple a restaurant applies to its own cost declines as cost rises, and it is worth knowing where the decline starts, because most published advice puts it far too high.

This ladder is reference for a restaurant's own pricing. It is convergent practitioner testimony, not a measurement. There is no published dataset of New York restaurant pricing by cost band; the one academic treatment of the question is a literature review with no original empirical data. What follows is what named New York and US practitioners say they do, with the source on every row. The convergence across independent sources is the strength of it. It is not a survey and should not be cited as one.

Reference, as of September 2026. The cost column is what a restaurant pays a wholesaler, from any source. No figure below is a Chu's Wine price.

Two carve-outs travel with this ladder and it is wrong without them.

1. The decline starts around $40–$60 of cost, not in the low hundreds. Two independent sources price the $150 band at 1.5× and 1.83×, and nothing in the evidence prices the $150–$250 corridor at 3×. The corridor is also where practitioners disagree most, which is itself the finding: it is the band where a house policy stops being a house policy.
2. "The multiple always declines" is false without the scarcity carve-out. Allocated and scarce wines are a separate regime priced to the secondary market rather than to cost, so the curve is not monotonic and does not run to zero at the top.

The mechanism underneath the ladder, and it is more useful than any of the numbers. Annie Shi describes it directly: below the list's own average sale price, price on a multiple; above it, price on absolute dollar value. That is the switch from a percentage to a gross-profit figure, and in a working room it happens well below $150 of cost. A restaurant that knows its own average bottle sold knows where its own switch is, which is the one thing the ladder cannot tell it.

Flat-multiple houses are real and are not incompetent. Four Horsemen is on the record at 3×, and other New York rooms run comparable flat multiples. The honest reading is that a flat multiple works in a room whose list sits in a narrow cost band, and stops working as the band widens — which is a statement about the list, not about the operator.

What the buying model changes about any of this, and what it does not

Chu's Wine is a licensed New York wine wholesaler. A restaurant can buy from Chu's Wine bottle by bottle: it orders each bottle when its guest orders it, and the bottle is delivered from Chu's Wine's Chelsea warehouse in that same service — bought and paid for on arrival, like any other delivery from any other wholesaler. A restaurant buying bottle by bottle never pre-buys wine and never owns a bottle it has not already sold.

The pricing arithmetic on this page is identical whoever supplies the wine, and a page that pretended otherwise would be caught in one read. The buying model changes exactly three things:

1. Dead stock on unopened bottles. Wine that never sells was never bought, so a slow bottling costs nothing until a guest orders it.
2. The minimum-velocity constraint relaxes on the bottle list, not on the glass list. A restaurant can list a bottle it would never risk owning, and the reach end of the bottle list is where the biggest gross-profit dollars per check live.
3. Working capital stays in the business rather than in the cellar.

And here is the limit, stated out loud, because it is the one a beverage director checks first: open-bottle spoilage is completely unaffected by how the wine was purchased. Once a bottle is opened it belongs to the restaurant and it is spoiling on the restaurant's clock. Every line of the velocity arithmetic and the preservation break-even above applies exactly as written, to every by-the-glass program, on any supply arrangement. Nothing about buying a bottle when a guest orders it makes an open bottle last longer.

What does a restaurant give up, and what does Chu's Wine get out of it?

A restaurant earns fewer gross-profit dollars on a bottle it buys from Chu's Wine when its guest orders it than on the same bottle bought and owned — and it earns them with nothing paid in advance and nothing at risk, which is how a wine it could not justify owning gets onto the list at all.

Am I getting a worse price than the restaurant down the street?

Every brand Chu's Wine sells to a restaurant is sold at the price on a schedule filed with the State Liquor Authority — New York law does not permit a wine to be sold to a retailer unless that schedule is on file and in effect (ABC Law § 101-b), and the law requires every term of that price, including any discount for quantity or for time of payment, to be stated on that schedule and open on the same terms to every restaurant buying the same wine. A restaurant is not negotiating against a rate card it cannot see.

What happens when a wine runs out?

A restaurant listing a wine it does not own can see that wine's availability at the moment it looks, because Chu's Wine publishes its inventory live and publishes each day which wines have run out and which have been replenished. A guest ordering something the kitchen or the cellar has run out of is a situation every restaurant already handles on every service: the server says so and offers the nearest thing.

Four costs come with that:

1. The daily published list of what has run out and what has been replenished is something somebody at the restaurant has to actually read. It is a standing task on someone's morning.
2. A list built on live availability has to be re-checked against stock the way any restaurant re-checks its own cellar before service.
3. Contention. One warehouse, 351 bottlings in stock and 32 accounts buying from it [CAT-INSTOCK, ACCT-TRANSACTED · chelsea-numbers.md §3, §11.1, publish]. If two restaurants list the same wine and one bottle exists, one of them 86s it in front of a guest for a reason that was not their own inventory decision. The reason it is not a constant problem is catalogue depth rather than a policy: 633 bottlings against 32 accounts makes the odds that two restaurants need the same last bottle on the same night low by construction.
4. The printed list. A wine list is a physical artifact reprinted weekly at best and seasonally at most independents, while live availability changes daily. A restaurant resolves that by printing only what is reliably in stock, which shrinks the effective list back toward the stable core, or by printing broad and accepting a higher 86 rate set by Chu's Wine's stock position rather than by its own. Keeping a guest-facing list current is the restaurant's own ongoing work, and it is a fit criterion.

What do I have to commit to, and how do I stop?

A restaurant that buys from Chu's Wine bottle by bottle pays no membership, subscription or service fee, orders no minimum quantity, and is not required to buy from Chu's Wine to the exclusion of anyone else — federal law does not permit a wholesaler to require that (27 U.S.C. § 205(a)). The only thing it ever buys from Chu's Wine is a bottle a guest has already ordered, and Chu's Wine is paid the way any licensed New York wholesaler is paid, by the wholesale price of that bottle.

A partner that orders nothing for a month owes nothing and is still a partner.

Chu's Wine is additive. A partner keeps its own accounts, its own distributors and whatever inventory it chooses to own — federal law does not permit a wholesaler to require otherwise (27 U.S.C. § 205(a)).

The eight things generic by-the-glass pricing advice gets wrong

1. Dividing by five pours instead of measuring realized yield.
2. Applying one multiple across the whole price range.
3. Averaging pour cost across SKUs instead of weighting it by pours sold.
4. Omitting spoilage, which is the largest real cost in the program.
5. Merging by-the-glass and bottle wine cost into one percentage. The two move in opposite directions.
6. Quoting national "industry markups" instead of a market's actual price ceiling. The New York entry-glass floor is set by rent and labour, not by the wine.
7. Importing retail keystone logic into restaurant pricing.
8. Treating wholesale cost as the restaurant's cost, and ignoring case minimums, split-case surcharges and delivery minimums. A $12 bottle that has to be bought twelve at a time is not a $12 bottle to an operator who wants three.

The four reports that run the program

• Weekly: pours sold per by-the-glass SKU. Flag any wine below its open-life velocity.
• Monthly: the yield audit — bottles opened × 4.7 against pours rung, per SKU.
• Per SKU: gross profit per pour × pours sold = contribution. Rank the glass list by contribution, not by pour cost. The lowest-pour-cost wine on most lists is the lowest contributor.
• Per SKU: glass-to-bottle conversion rate, which is what tells you whether the bridge above is set correctly.

Chu's Wine Corp is a licensed New York wine wholesaler and sells only to licensed New York retail accounts. Nothing on this site is an offer to sell wine to consumers. Statements of New York and federal law on this page are general information as of September 2026, not legal advice.