The Biggest Missed Sale in Online Wine Retail
Picture this. It’s 5pm on a Friday. A customer has just decided what’s for dinner, barbecued salmon, let’s say. They want a bottle of wine to go with it. They open your website.
And they’re met with a wall of filters.
Country. Region. Grape variety. Style. Price range.
They don’t want to filter. They want to ask a simple question: what wine goes with salmon?
They can’t. So they leave. Maybe they grab something from the supermarket on the way home. Maybe they Google it and end up on someone else’s site. Either way, you’ve lost that sale, not because you didn’t have the right bottle, but because your website couldn’t understand a simple question.
This isn’t a hypothetical. It’s happening on wine retail websites every single day. And we have the data to prove it.
What 687 Real Conversations Told Us
At Pinpointed, we analysed 687 real customer conversations with AI Sommeliers deployed on wine and spirits retail websites across three continents between January and March 2026.
The second most common category of question, after operational queries about delivery and orders, was food pairing. These weren’t vague questions, they were specific.
Real questions customers asked:
- “I’m cooking lamb shanks, what red wine under €15 can I get?”
- “Wine for a BBQ — steak, sausages, chicken.”
- “What bourbon pairs well with chicken?”
- “A beer that should be cold but has been allowed to go warm. The toast will have cream cheese on it.”
Read that last one again. That’s a customer who knows exactly what they want, they just can’t express it in the language your search bar speaks.
These customers have already made their purchasing decision. They’ve planned dinner. They’ve committed to cooking. They just need one more thing, they’re not browsing, they’re buying. But most online wine stores have no mechanism to help them.
The Problem with Filters
Filters work brilliantly for a specific type of wine customer: the one who already knows what they want.
They want a Barossa Valley Shiraz under $40. Filter by region. Filter by grape. Filter by price. Done.
But that customer represents a small fraction of your traffic. The vast majority of people shopping for wine online don’t think in grape varieties, appellations, or tasting notes. They think in meals, occasions, and budgets.
“Something for a dinner party.” “A nice white for the weekend.” “What goes with the roast?”
Try finding the answer to any of those questions using a dropdown menu.
The average cart abandonment rate for food and beverage e-commerce sits at around 74%. That’s nearly three out of every four shoppers who add something to their cart but leave without buying. For wine, the problem is arguably worse, because the purchase decision is more complex than buying groceries. Wine requires guidance and online, there’s nobody to give it.
In a physical wine shop, this problem doesn’t exist. A customer walks in, says “I’m making salmon tonight, what should I get?”, and your best staff member handles the rest. They ask a couple of follow-up questions, how are you cooking it? What’s your budget? Do you prefer something lighter or more full-bodied and within sixty seconds, the customer is walking out with a bottle they feel confident about.
Online, that conversation doesn’t happen. The customer is on their own.
Why Dinner-Planners Are Your Most Valuable Customers
There’s something important about the customer who arrives on your site with dinner already planned. They have three qualities that make them the ideal buyer:
They have urgency. Dinner is tonight. This isn’t a browse-and-bookmark session. They need to buy now. If you can help them quickly, the conversion is almost guaranteed.
They have context to share. A customer who says “I’m making roasted lamb shanks for six people” has given you more useful information in one sentence than any filter could extract. The meal tells you the style. The number of guests suggests quantity. You know this is a considered occasion, not a Tuesday night on the sofa. All of that context is gold for making the right recommendation.
They’re open to guidance. A customer who knows they want a specific bottle doesn’t need you. A customer who only knows what’s for dinner needs you. They’re actively looking for expert advice. They trust the recommendation. And our data shows they consistently spend more than their initial budget when they receive confident, personalised guidance, on average, 27% more.
These customers aren’t a niche. They’re a huge portion of online wine traffic. And right now, most online stores are letting them slip through the cracks.
What Happens When You Can Answer the Question
When a customer can type “what wine goes with lamb shanks?” and get a real answer, not a list of search results, but an actual recommendation from the store’s inventory with a price and a reason, something changes.
The conversation opens up.
From our data, here’s what a typical food-pairing conversation looks like:
Customer: “I’m cooking lamb shanks tonight, what red wine under €15 can I get?”
AI Sommelier: Asks how the lamb is being prepared (braised, roasted, with spices?) and whether the customer has a preference for lighter or fuller reds.
Customer: “Braised, with rosemary and garlic. Something full-bodied.”
AI Sommelier: Recommends two or three specific bottles from the store’s live inventory, a Côtes du Rhône at €13, a Malbec at €14, and a slightly more premium Châteauneuf-du-Pape at €19 while explaining why each works with the dish.
The customer clicked the €19 bottle. Their budget was €15.
That’s not an upsell. That’s what happens when someone gets confident guidance. They don’t feel pushed. They feel helped. And when people feel helped, they spend more willingly.
Across our dataset, customers who received AI-guided recommendations spent an average of €56, compared to their stated average budget of €44. That’s the difference between a website that lists products and a website that sells them.
How In-Store Thinking Translates Online
If you run a wine shop, you already know how to sell wine to someone who only knows what’s for dinner. Your staff do it every day. The challenge is replicating that expertise online at scale, at any hour, in any language.
Here’s what your best staff member does instinctively that your website currently can’t:
They listen first. They don’t start by listing every red wine in the shop. They ask what’s being cooked, how it’s being prepared, how many people are coming. The meal is the starting point, not the wine.
They read the customer. Is this person a wine enthusiast who wants to geek out, or someone who just wants a solid recommendation without the lecture? Your best staff adjust their approach. Online, that means your recommendation engine needs to understand “something nice” just as well as it understands “a medium-bodied Sangiovese with good acidity.”
They work within the customer’s budget, then stretch it gently. Nobody walks into a shop saying “I’d like to spend exactly €14.50.” They say “around fifteen euros” or “not too expensive.” A good recommendation gives them a solid option at their price point and a slightly better one just above it. Not pushy. Just present.
They recommend from what’s actually on the shelf. This sounds obvious, but it’s where most online recommendation tools fail. If you suggest a bottle that’s out of stock, you’ve broken the customer’s trust. Every recommendation needs to come from your live inventory.
They speak the customer’s language. In our dataset, customers asked questions in English, German, French, Portuguese, Spanish, Dutch, and Italian, often without prompting. One customer even switched languages mid-question. If your online experience can’t respond in the customer’s language, you’re not just losing a sale. You’re telling them they’re not welcome.
The Practical Takeaway for Wine Retailers
If food pairing is one of the most common reasons customers engage with your store and our data strongly suggests it is, then your website needs a way to handle that conversation.
Here’s what that means in practice:
Your product pages aren’t enough. A product page tells customers what you have. It doesn’t help them figure out what they need. The customer who’s cooking salmon tonight doesn’t want to read tasting notes on 40 different white wines. They want someone to say “this one.”
Your search bar isn’t enough. Most site search is keyword-based. “Wine for salmon” returns every product with “wine” or “salmon” in the description which is probably nothing useful. Conversational search is fundamentally different from keyword search. It requires understanding intent, not just matching words.
Filters definitely aren’t enough. If your customer knew enough to use your filters effectively, they wouldn’t need your help. The customers who need the most guidance are the ones your filters serve the worst.
An AI Sommelier trained on your inventory changes the equation. It listens to what the customer is cooking. It asks the right follow-up questions. It recommends bottles you actually have in stock, at prices that match the customer’s budget. And it does this 24 hours a day, in multiple languages, without taking a day off.
The technology exists right now to give every online wine shopper the same experience they’d get from your best in-store staff member. The retailers who adopt it first aren’t just improving their conversion rates — they’re capturing the customers that everyone else is losing.
The Customer Is Ready. Is Your Website?
Here’s the uncomfortable truth: your customers are already trying to have this conversation with your website. They’re typing “wine for BBQ” into your search bar. They’re scanning your category pages looking for pairing advice. They’re leaving when they can’t find it.
The question isn’t whether customers want guided recommendations. The 687 conversations we analysed make that clear. The question is whether your store is set up to provide them.
The customer who only knows what’s for dinner is not a problem. They’re an opportunity. They have money to spend, a decision to make, and a willingness to trust your expertise.
All you have to do is give them a way to ask the question.
Pinpointed AI Sommelier helps wine, beer, and spirits retailers turn food-pairing questions into sales 24/7, in 16 languages, using only products from your live inventory. Book a free demo to see it in action on your store.