Wine Menu Engineering: BTG Elasticity, Attach Rate and Dead Stock

What is wine menu engineering?

Wine menu engineering is the practice of sorting every wine on your list by how well it sells and how much it earns, then using placement, price and staff recommendations to move guests toward the wines that work for both them and the business. It turns a wine list from a static catalog into something you manage with sales data.

The method most operators use comes from the menu engineering model Michael Kasavana and Donald Smith published in 1982. It was built for food menus, but it works for wine because every bottle and glass has the two numbers it needs: how many you sold and how much each one contributed.

How do you sort a wine list with a menu engineering matrix?

Plot each wine on two measures, popularity and contribution margin, and it lands in one of four groups that each call for a different action. Run the analysis within one category at a time, such as wines by the glass or red bottles, so you compare like with like.

1. Pull sales by item for one full period, such as a month, from your point of sale.
2. Calculate contribution margin for each wine: menu price minus cost. Use the margin in dollars, not the markup percentage, because dollars are what pay the bills.
3. Set the popularity line. In the original method, a wine counts as popular if its share of sales in the category is at least 70 percent of an equal share. With 20 wines by the glass, an equal share is 5 percent, so the line sits at 3.5 percent.
4. Set the margin line at the average contribution margin across the category, weighted by sales.
5. Sort each wine into its group and act on it.

• Stars (popular, high margin): keep them visible and in stock. These are the wines your pre-shift list should never show as out.
• Plowhorses (popular, low margin): guests like them, so change carefully. Review your cost first, then test a small price change.
• Puzzles (unpopular, high margin): the margin is there but the sales are not. Move them to a better spot on the list, add a pairing note, or have staff recommend them.
• Dogs (unpopular, low margin): replace them unless they serve a clear purpose, such as anchoring a region or a pairing.

What is BTG price elasticity?

BTG price elasticity measures how much your by the glass sales change when you change a glass price. It is the percentage change in glasses sold divided by the percentage change in price. If a 10 percent price increase cuts glasses sold by more than 10 percent, that wine is elastic and the increase probably cost you money. If sales barely move, there is room to price up.

You find a wine's elasticity by testing, not by guessing:

1. Change one price at a time and leave the rest of the list alone.
2. Compare equal periods, such as the four weeks before and after, and watch for holidays or events that would skew either one.
3. Track glasses sold and contribution margin together. Fewer glasses at a higher margin can still be a win.
4. Watch the neighbors. If guests switch to the next wine down the list, the change moved sales rather than losing them.

This guide covers how guests respond to glass prices. For setting the bottle markups behind them, see our guide on how to price a restaurant wine list.

What is attach rate, and how do you raise it?

Attach rate is the share of checks, or covers, that include wine. Divide the number of checks with at least one wine sale by total checks for the same period. It tells you how often wine is part of the meal, which matters as much as what each wine earns.

• Put pairings on the food menu. A suggested glass next to a dish removes a decision for the guest.
• Train staff on a short list. Servers recommend what they know and have tasted. Give them three or four wines to talk about each week, and favor your Stars and Puzzles.
• Offer an easy entry point. A well chosen glass at a comfortable price gets hesitant guests to say yes.
• Track it by shift and by server. Attach rate varies across a team, and the gaps show you where training will pay off.

How do you find and clear dead stock?

Dead stock is wine that is not selling and is tying up cash and space. Measure it with inventory turnover: cost of wine sold for a period divided by the average value of wine inventory at cost over that period. A low turnover number for a wine, a category or the whole list means bottles are sitting.

Set your own threshold for what counts as dead, such as no sales in a set number of weeks, and review that list on a regular schedule. Then move it:

• Feature it by the glass or as a staff pick for a limited run.
• Build a pairing dinner or tasting around a few slow bottles.
• Ask your supplier about returns or exchanges where your agreement allows them.
• Stop reordering it. The cheapest dead stock fix is not buying more.

Not every slow bottle is dead stock. Wines you are deliberately aging for a reserve list are an investment, so track them separately from your working inventory.

What data do you need to engineer a wine menu?

You need item level sales from your point of sale, an accurate cost for every wine, and inventory counts you trust. Menu engineering is only as good as the cost and count data underneath it, so start there.

• Run consistent counts with our bar wine inventory count guide.
• Log breakage and spills so they do not distort your numbers, using our bar wine waste and breakage tracking guide.
• Keep the printed list in step with what is in the cellar with our weekly guide on how to update a restaurant wine list.
• Catch missing bottles between counts with variance alerts and pre-shift lists.

InVintory for hospitality tracks every bottle by producer, vintage and location, so the inventory side of this analysis starts from records you can rely on.

FAQ

How often should you re-engineer a wine list?

Re-run the analysis whenever you change the list or the season changes, and always use at least one full period of sales. Decisions made on a single busy weekend tend to reverse themselves.

Should every wine on the list be a Star?

No. A list needs range, and some wines earn their place by anchoring a region, a price point or a pairing. The goal is to know why each wine is there, not to make every wine a top seller.

Should you always cut a Dog?

Not automatically. Check first whether it is miscosted, badly placed or simply never recommended. If none of those changes the result, replace it with something that fits the gap.

Does menu engineering work for bottle lists as well as wines by the glass?

Yes. Run bottles and glasses as separate categories, since they sell at different volumes and margins, and compare each wine only against others in the same category.

Ready to put real inventory data behind your wine list? Explore InVintory for hospitality.