How do I sell more wine in my restaurant?

Mostly by taking the fear out of the order. Restaurant wine is not shrinking because guests stopped liking it; it is shrinking because a wine list is the one place on the menu where a guest can feel judged, and a guest who feels judged orders a beer, a cocktail, or water. The evidence below points to four things, none of which needs a sommelier: a short, plainly described list on the food menu; by-the-glass prices that make a $12 pour an easy yes; servers who ask two questions (what are you eating, and what would you like to spend); and a way for the guest to find out what a wine tastes like without having to ask out loud. The numbers behind each of those follow.

Why is wine harder to sell in a restaurant than it used to be?

Because fewer people drink at all, and the ones who do are choosing other things more often. Gallup's July 2026 poll (1,200 US adults, fielded July 1–19) put the share of Americans who drink alcohol at 54%, unchanged from 2025 and down from 62% in 2023 — a record low for the poll. Among adults aged 18 to 34 the drop is the same eight points, from 58% to 50%. Half of US adults (51%) now say one or two drinks a day is bad for your health.

Wine has taken more than its share of that decline. Terrain, the research arm of American AgCredit, wrote in March 2026 that US wine consumption has fallen four years running and sits about 15% below 2019. Silicon Valley Bank's 2026 State of the US Wine Industry report (published January 15, 2026) put 2025 volume down roughly 2% on 2024 and described Millennial and Gen Z drinkers as "spread across more categories and drinking less overall, particularly under age 29."

Restaurants feel it directly. Southern Glazer's Wine & Spirits, the country's largest distributor, puts on-trade (restaurant and bar) wine sales down about 26% since 2019, as reported by The Drinks Business in May 2026. NielsenIQ's on-premise data in the same report showed wine's share of every dollar spent on alcohol in US bars and restaurants slipping another 0.3 points over the prior 52 weeks, with ready-to-drink cocktails picking up the difference. CGA's on-premise measurement, cited by Bar & Restaurant in December 2024, had wine at 13% of on-premise alcohol dollars and falling 4% year on year.

That is the backdrop. It is not a reason to give up on wine — the National Restaurant Association's beverage research found alcohol is on average 21% of a full-service restaurant's sales, and 69% of wine drinkers say whether a restaurant serves alcohol affects which restaurant they choose. It is a reason to stop assuming the list sells itself.

Why don't guests order wine even when they want to?

Because ordering wine is the one moment at the table where a guest can look ignorant, and most guests would rather not risk it.

The Wine Market Council's 2025 study of 1,500 occasional or "wine-hesitant" drinkers, reported by Wine Enthusiast in December 2025, found that infrequent wine drinkers were more likely than regular ones to be intimidated by shopping for wine; the largest single reason they gave for choosing something else was the occasion — wine felt formal rather than casual (43%) — and 70% named health as a reason they don't drink wine more often. Ben Aneff, quoted by The Drinks Business in March 2026, put the same finding more bluntly: wine is "losing at the moment of choice" — the piece's reading being that wine labels and lists give a guest less direct guidance about taste and style than the drinks wine competes with.

Sommeliers say the same thing from the other side of the table. In a July 2025 Wine Spectator piece that opens with the question "Do you get nervous choosing a bottle off a restaurant wine list?", Luciane Froz of Ci Siamo in New York described her job as drawing "the conversation away from a situation where they might feel judged, away from technical vocabulary." Dan Davis of Commander's Palace: "please, please, please don't hesitate to tell me how much you want to spend." Mike Lee of La Toque said the tool he uses to put people at ease is "finding out what their max budget is and coming in just under it."

Notice what those three are describing: not persuasion, but the removal of two risks — sounding uninformed and overspending. A guest who can find out what a wine tastes like and what it costs without having to ask a person is a guest who orders it. Everything below is a way to give them that.

Guests, for what it's worth, want the help. The National Restaurant Association's beverage-alcohol research found 84% of adults who drink wine, beer, or cocktails consider restaurants a good place to learn about drinks they haven't tried, and 82% trust the staff at their local restaurants to make good recommendations. The trust is there; the asking is what's hard.

What should a small restaurant's wine list look like?

Short, on the food menu, priced without dollar signs, and organized by something other than "style."

The best-known study of what wine-list design correlates with sales is Sybil Yang and Michael Lynn's 2009 Cornell Center for Hospitality Research report, which coded 46 attributes across 270 restaurant wine lists in major US metropolitan areas and matched them to wine sales. Four things were associated with higher sales: including the wine list on the food menu rather than as a separate document; listing prices without a dollar sign; naming more wines from well-known wineries; and carrying a "Reserve" or similar special section. One thing was associated with lower sales: organizing the list by wine style ("bold," "crisp," "sweet"). In casual restaurants, sales rose with list length up to roughly 150 wines and were higher where the list carried more low-priced wines; in fine dining, neither length nor price level showed any relationship. The authors were explicit that these are correlations, not proven causes, and the study is now seventeen years old — but its first two findings cost nothing to try.

For a one-location restaurant the practical reading is: put the list on the same sheet as the food, so a guest sees it while deciding what to eat rather than having to ask for it; drop the dollar signs and the cents; give the list a shape a guest recognizes (by color, then by grape or region — not by adjectives); and make sure at least a third of the list is at a price a guest won't feel they have to justify. The "reserve" finding is likely about signalling that the cheaper bottles are a choice, not the ceiling — even three bottles under a heading of their own does that.

Do wine descriptions on the menu actually sell wine?

In the most recent controlled restaurant trial to measure it, yes — and the interesting part is who bought.

Scott Sheridan and Marc Dressler ran a 111-day experiment at a single 26-seat upscale restaurant in Stuttgart, Germany, published in the journal Businesses in May 2025. The same list was printed three ways across the period — no descriptions, short descriptions (a few aroma and flavor words), and long descriptions (origin, the producer, fuller tasting notes) — and a separate set of weeks added a recommended wine beside each dish. Short descriptions raised wine sales about 7% on average; long descriptions about 13%, with white wines up 20% by unit. Pairing recommendations printed on the menu raised the number of bills carrying wine by 30–50% in the weeks they ran (the authors' normalized estimate of the sales lift is up to 10%). Two findings matter more than the percentages for a small restaurant. First, the authors report that more than 70% of the extra wine sold came from guests who would otherwise have ordered something other than wine — the descriptions were not upselling wine drinkers, they were converting the hesitant guests the Wine Market Council describes. Second, the restaurant's staff reported that descriptions "reduced the frequency of requests for recommendations," while guests less familiar with wine actually talked to servers more once the menu had given them something to ask about.

One honest caveat: this is one restaurant, in Germany, and the authors themselves note their statistical test on order composition was not significant. A 2023 experiment in the Journal of Wine Economics by Nadia Streletskaya and colleagues, run in a tasting-room setting rather than a restaurant, found that adding sensory notes and pairing suggestions made no measurable difference to whether people chose to taste a wine. The fair reading is that descriptions help most where the guest is deciding under social pressure and can't easily ask — which is a dining room, not a tasting bar.

What to write: one line per wine, in words a guest already owns. "Dry, light, tastes like green apple; good with the fish" does more than a grape, a region and a vintage. If a wine needs a sentence of story — the producer, why you carry it — that is the "long description" that sold 13% more.

How should I price wine by the glass and by the bottle?

Price the glass so it's an easy yes, and stop marking bottles up so far that the guest switches to a cocktail.

The conventions, as stated by Toast in September 2025 and by Provi in 2022: a bottle is listed at 200–300% over its retail price, with the multiple falling as the wholesale cost rises (Toast's tiers run three to four times wholesale on a $10–20 bottle down to two to two-and-a-half times above $50); a glass is priced at roughly what the whole bottle cost you wholesale, on the assumption of four to five pours (Toast) or four to six (Provi) per 750 ml bottle. Toast puts the average margin on restaurant wine around 70%. Provi's reading of consumer data is that guests on a by-the-glass list of $8–$15 wines "strongly preferred" the $11–$15 range — the cheapest glass is rarely the one that sells.

The pressure now is at the top of that range, not the bottom. The Drinks Business reported in July 2026 that some US restaurants have pushed bottle markups to six times cost, up from a four-times convention, and quoted Wine Market Council data that 75% of wine drinkers will not pay more than $16 for a glass. Gino Colangelo of Colangelo & Partners, quoted in the same piece: "If people stop drinking in your restaurant because your prices are prohibitive, then it's a very short-sighted way." Kevin Bratt, Lettuce Entertain You's national wine and spirits director, was quoted by The Drinks Business in May 2026 saying purchases of $1,000-plus bottles were down 5% while a Napa Cabernet around $100 was "moving briskly." The guests who are still buying are buying in the middle.

A study worth knowing when you set the list: David de Meza (LSE) and Vikram Pathania (Sussex) analysed 470 wine lists from 249 London restaurants — 6,335 wines — in a 2021 American Association of Wine Economists working paper. Restaurant prices averaged more than three times retail, the highest percentage markups sat around the middle of the list, and the second-cheapest wine — the one folklore says is the rip-off — carried a markup significantly below the third, fourth and fifth cheapest. Guests who order "second cheapest" to avoid looking cheap are not being fleeced, and a restaurant that prices its second-cheapest bottle fairly is pricing the wine most nervous guests will pick. The same study found a large 250 ml glass costs less than 7% more per millilitre than the bottle — by-the-glass, priced honestly, is not the raw deal guests assume.

How do I get servers to sell wine without a sommelier?

Give them two questions and one tasting, and take the vocabulary away.

The Wine Spectator sommeliers above all converge on the same script, and none of it requires certification: ask what the table is eating, ask what they'd like to spend (and mean it), then suggest one bottle or two glasses in plain words. A server who has tasted every wine on a twelve-wine list and can say "this one's the lighter red, it's what I'd have with the pork" is doing what 82% of guests say they trust their local restaurant's staff to do. The training that gets a server there is a staff tasting when the list changes — a bottle of each, ten minutes, the owner or the distributor's rep talking — not a course.

If you do want a credential on the floor, the entry points are modest. The Court of Master Sommeliers Americas' online Introductory Sommelier Course and Examination is $649 for 180 days of access (plus $250 for the optional tasting kit), with the exam at 70 multiple-choice questions; the WSET Level 1 Award in Wines is a six-hour course, with prices set by each local provider. Either is a reasonable investment in one motivated server. Neither is a substitute for a list a guest can read on their own — hiring is the harder problem: the National Restaurant Association's February 2026 State of the Industry found nearly three-quarters of operators plan to hire this year and expect difficulty finding experienced people, and 42% of operators were not profitable last year. Whatever wine knowledge your restaurant has should live somewhere that survives a server's last shift.

What should an independent restaurant actually do this month?

Five things, in rising order of effort:

1. Put the wine on the food menu, without dollar signs. The two list changes with the best evidence behind them cost a reprint and nothing else.
2. Write one plain line per wine. What it tastes like, what it goes with. If you can't write the line, you can't expect a server to say it — and a guest who can read it doesn't have to ask.
3. Price a glass a guest won't have to think about. Under $16, on the wholesale-bottle rule, four to five pours. Put the second-cheapest bottle at an honest markup; it's the one hesitant guests choose.
4. Run a staff tasting every time the list changes. Two questions — what are you eating, what do you want to spend — and one sentence per wine, in the server's own words.
5. Give the guest a way to find out without asking. A line on the menu, a card on the table, a page on your phone — whatever lets the guest who is afraid of sounding uninformed get the answer privately. More than 70% of the extra wine the Stuttgart restaurant sold came from exactly that guest.

None of this needs a sommelier, a wine program, or software. It needs a list that answers the guest's question before they have to ask it, and a floor that asks the budget question without flinching.

That last item is the one our own product was built around, for what it's worth. Brief First's free page is your menu as a live page a guest reaches from a small card on the table, and a wine row is just a menu row: the name, the price as text, and the line you wrote about it — so the "what does this taste like" a guest won't ask a person is already answered on their own phone. The paid tier is an AI host that answers the follow-up ("which of these goes with the lamb?", "is the Vermentino dry?") in the guest's own language, only from the words you wrote — and it never claims a bottle is in stock; that answer goes to your server. The page and the private line are free for as long as you run them; the host is from $30 a month. Either way, the guest who is nervous at the list is worth reaching, whether you use us or a server who has tasted every bottle and asks what they'd like to spend.