How AI evaluates restaurant wine — and why most people guess wrong.
Most people guess when they order wine at a restaurant. Somm-AI doesn't.
When you stare down a 40-bottle wine list at dinner, you're not just choosing a flavor — you're trying to figure out which bottle gives you the most quality for your money. That's a surprisingly complex problem, and it's one that even experienced wine drinkers get wrong.
Somm-AI solves it using a proprietary 5-Dimension Wine Value Model: a structured framework that evaluates every bottle on a restaurant wine list across five independent quality and value signals, then combines them into a single ranked score. Here's exactly how it works.
A $90 bottle of wine on a restaurant list could be a magnificent deal — or a significant overpay. The number alone tells you nothing about whether that wine is actually good, whether you're paying a fair price for it, or whether a $60 bottle two rows up would outperform it in every meaningful way.
Restaurant wine pricing is notoriously opaque. Markups typically range from 150% to 400% over retail. Some bottles are priced on name recognition alone. Others are underpriced gems from regions consumers haven't discovered yet. Without expertise — or a tool that has that expertise built in — you're flying blind.
The 5-Dimension Value Model was built to cut through that opacity and give every diner access to the same judgment a world-class sommelier would apply.
Critic Ratings & Producer Reputation
The intrinsic quality of the wine itself, independent of price.
Every wine Somm-AI evaluates is assessed for its underlying quality based on producer reputation, critic ratings, regional pedigree, and independent tasting data. This is the baseline: is this actually a good wine, regardless of what it costs?
A high quality score means the wine consistently performs at or above its category. A low quality score means you're paying restaurant markup on a wine that wouldn't impress you even at retail.
Price and quality are not the same thing on a restaurant wine list. A bottle can be expensive and mediocre, or inexpensive and outstanding. Quality Score separates the two.
Restaurant Markup vs. Retail Market Value
How the restaurant's asking price compares to the wine's fair market value.
This is where most of the value identification happens. Somm-AI compares what the restaurant is charging for a bottle against what that same wine sells for at retail — across multiple markets and data sources — to calculate whether you're getting a fair deal, a bargain, or a significant markup.
A wine priced at 1.5× retail is a much better deal than a wine priced at 4× retail, even if both bottles appear on the same list at similar prices. Price Position makes that visible.
Restaurant markups vary wildly by bottle, not just by restaurant. Identifying low-markup bottles on a specific list — your list, on this evening — is something no generic wine app can do. It requires analyzing the actual menu you're looking at.
Finding Undervalued Wine Regions on Any Restaurant List
Whether the wine's region of origin is currently over- or undervalued in the market.
Wine regions go in and out of fashion. Napa Cabernet commands a premium partly because it's famous — not always because it outperforms a Paso Robles or Columbia Valley Cab at a similar price point. Burgundy's reputation prices even entry-level bottles at a premium, while neighboring appellations in the Côte Chalonnaise offer similar terroir for a fraction of the cost.
Somm-AI tracks regional pricing dynamics and scores each bottle based on whether its region is currently delivering outsized value for the dollar — or charging for its reputation.
The best-value wines at most restaurants come from regions that are genuinely excellent but haven't been "discovered" by mainstream diners yet. Regional Value surfaces those hidden opportunities systematically.
Is This Wine Actually Drinking Well Right Now?
Whether the wine is being served at the optimal point in its aging curve.
A great wine served too young or too old is a wasted opportunity. Certain vintages are exceptional; others were challenging. A 2015 Barolo served today is in a very different place than a 2020 of the same wine — and neither is better or worse in absolute terms, but one may be dramatically more enjoyable right now.
Somm-AI evaluates each wine's vintage against known aging guidelines and vintage assessments to score whether the bottle is drinking well at this moment, past its prime, or still too tight to show its best.
Restaurants hold wine for years. Some bottles are perfectly cellared and ideally timed. Others have been on the list too long. Vintage Timing tells you which is which before you order.
Supply, Demand & Wine Value Signals Beyond the Label
Broader supply, demand, and trend signals that affect a wine's true value at this moment in time.
Some wines are scarce and genuinely hard to find — making a restaurant's allocation a real opportunity. Others are widely available and simply priced for perceived exclusivity. Some producers are on an upward trajectory (underpriced today relative to where they'll be in two years); others are coasting on past reputation.
Market Dynamics captures these signals: production volume, critic attention trajectory, collector demand, and import availability. It's the dimension that reflects timing — not just quality or price, but whether this particular bottle is a smart buy at this specific moment.
Wine is a market. Prices reflect supply, demand, and perception as much as quality. Understanding market dynamics is the difference between buying a bottle because it's actually good value and buying it because it's famous.
Each of the five dimensions generates an independent sub-score. Somm-AI weights and combines them into a single Value Score for each bottle on the list — then ranks every wine by that score within your budget and preferred style.
The result: your top 10 value picks, returned in seconds, based on the same multi-factor analysis a trained sommelier would apply — but faster, more consistent, and without any personal preference bias.
No guessing. No intimidation. No overpaying.
Somm-AI is built for diners who want more from their restaurant experience — not just a good wine, but the right wine at the right price.
Find My Best Value Wines →