Perfect wine pairing: traditional method vs the Masterestaurant method

What is a perfect pairing for a restaurant?

A perfect pairing is the dish-and-drink combination the table accepts, that lifts the average check and that leaves more contribution margin than the dish alone, and everything else is tasting-note literature.

Diego F. Parra defines it that way at Masterestaurant because the definition decides what you measure. Judge by harmony on the palate and the sommelier wins every argument with reasons nobody can audit; judge by the register and each combination carries an acceptance rate and a dollar contribution you review every week at close. Guests are not generous with approval either.

Vinetur (2023) found that 40 % of North American respondents felt pairings at wine and food events were not up to par, and a bar that low is an OPPORTUNITY for whoever works it with numbers instead of adjectives.

Step 1: audit the real cost of every glass and every dish

Without a sheet showing every dish and every label at its cost per served portion, open-bottle waste included, no pairing gets designed, so that sheet is the first deliverable. Wines by the glass come first, since that is where the loss hides: a bottle opened Tuesday and poured out Thursday costs about as much as two glasses sold, and no spec sheet records it.

Those costs then get crossed with each dish's food cost, kept under the method's 32 % ceiling as a MAXIMUM and with no payroll or rent on the plate. If a glass costs you 2 USD with its waste and sells for 8, for example, it contributes 6 USD, and that figure competes with dessert or coffee at the same table.

The sheet is ready when anyone on the team can find the cost of a glass in under a minute.

Step 2: choose combinations by margin first and taste second

Register first, palate second: candidates come from crossing the sales mix with each drink's contribution, and tasting opens only among the ones that win there. A dish that does not sell drags no glass along, however well it harmonizes on Monday, because the menu rules over the bar.

Rank your star dishes by volume, test several drinks for each with the chef and the bar lead (wine, beer, a cocktail or something non-alcoholic) and drop whatever contributes less than the dessert it displaces. Beer gets less attention than it deserves. The National Restaurant Association (2024) puts at 83 % the share of US restaurants serving alcohol to go that offer beer with pickup or delivery orders, which means pairing travels beyond the dining room too. What you end with is a short list, one pairing per star dish with its contribution calculated.

Step 3: design the by-the-glass and half-glass format

Glass or half glass per course, rather than the bottle, is the format that works for most pairings: the table tries without committing and you charge for every step of the menu. Guests have already moved there, since in the Coravin report published by Wine Industry Advisor (2025), 54 % of wine drinkers in the US, the UK and Australia say they order wine by the glass more often than two years ago.

Still, every open glass is potential waste, and that is why this step gets decided with the previous sheet in hand. Price the half glass so its percentage contribution matches or beats the full glass. By-the-glass service stays for labels that turn over weekly, while slow movers are sold by the bottle only. You are finished when each pairing carries a glass price, a half-glass price and a cap on open bottles per shift.

Step 4: train the floor and measure the acceptance rate

If the server cannot sell it, the pairing does not exist, which makes the fourth step a short script per dish plus a weekly count of how many tables say yes. Forget recited tasting notes. Two sentences are enough, why that drink suits that dish and what the half glass costs, delivered while the table picks its main course and not once forks are already moving.

Italy mirrors what happens when wine is habit: according to Federvini and TradeLab (2026), 41 % of wine-drinking diners say they often drink it when eating out, and in markets where the habit is weaker the script has to work harder. Your POS shows how many star dishes leave the kitchen with their suggested drink, a number the team reviews every Monday.

After four weak weeks you swap the drink or the script, but a combination is NEVER left sitting on the menu.

What are the most common mistakes when building a pairing?

Designing from the tasting sheet instead of from cost is the most repeated mistake, followed by leaving bottles open with no rotation cap. Confusing trend with demand comes third:

the National Restaurant Association (2024) reports that 54 % of full-service operators with alcohol planned to add new cocktails, and plenty copy that wave without asking whether their own tables want it. For years I defended long five-course pairings because they looked great when the menu was presented, and I was wrong, since hardly anyone finished them and waste ate the contribution. Picture a premium pairing launched with no bottle cap.

The table orders a half glass of the expensive label, the bottle stays open, the next shift cannot move it and that glass ends up costing double, invisible until inventory. So the cap gets set BEFORE launch.

How do you know the pairing was done right?

A pairing is done right when every combination has a verified cost per glass, a glass and half-glass price, a contribution calculated against the dish alone and an acceptance rate tracked in the POS for four weeks.

Go over that list with the chef and the bar lead at one table, because many failures live on the border between kitchen and bar, where each side assumes the other measured. Confirm too that no dish in the pairing breaks the 32 % food cost ceiling and that open bottles are capped per shift.

In Spain, apparent wine consumption stood at 9.78 million hectoliters as of November 2024, up 0.7 % year over year (INFOVI and Nielsen IQ, via Interempresas): demand that holds steady and gets split rather than grows. One task this week: pull the mix of your five best-selling dishes from the POS and calculate the contribution of the best-paying drink next to each.

The differences that decide whether the pairing sells

The first thing I correct when a chef shows me a pairing list: it is NOT a list of textbook harmonies, it is a product with a price, a cost and a take rate, like any dish.

At Masterestaurant, Diego F. Parra sets a sequencing rule that is not up for debate, margin first and taste second, because out of ten combinations that work on the palate there are always three or four that also work at the register, and those are enough. Wine by the glass changed pairing math.

In the by-the-glass consumer report covered by Wine Industry Advisor (2025), 34% of wine drinkers in the US, UK and Australia choose different glasses to match each course, which rewards the restaurant that designs costed half-pour pairings and punishes the one that can only suggest full bottles a table of two will not risk. There is a real tension between the sommelier and the register: the best gastronomic match is often the label with the worst margin.

The differences that decide whether the pairing sells — in practice

You resolve it without picking sides, filtering first by contribution margin and letting the sommelier choose freely inside what passed, so expertise rules where it adds value and the register rules where it decides. What happens if you pair your signature dish with the priciest bottle in the cellar?

The take rate drops, because menu price psychology makes guests add dish and glass in their heads; the opened bottle then ages at the bar; by day three the waste eats the margin of the glasses you did sell, and at month-end the sales mix shows a signature dish

The differences that decide whether the pairing sells — key points

selling as before next to a wine nobody orders. The wine was never the problem, the order of the decision was. The suggestion lives on the printed menu, next to the dish, where servers use it to pace service and upsell with hospitality; the QR code complements it with tasting notes, the alcohol-free version and current prices. We always recommend BOTH, each with its role.

And I got this wrong for years: I thought a good pairing sold itself, when what sells it is a server offering it at the right moment with a line they rehearsed.