Alcohol Sales Trends: What the Register Shows

Alcohol sales trends are easy to read wrong, because the department reports in dollars and dollars move for three reasons at once: price, pack size, and units. The version that survives a category review is narrower. Units per store per week by segment, the average price of the selling unit, and how each of those moved against the same weeks a year earlier.

What follows is that read, worked through a 38-store grocery and convenience set over 52 weeks against the prior 52. Published industry numbers get their own section with the source and the period attached, because they are measured on a different universe than any one operator's register, and mixing the two is how a category deck ends up indefensible.

Key takeaways

• Read the department in units per store per week first. A dollar decline with flat units is a mix story, not a demand story, and the two need different responses.
• Pack migration is where most of the movement hides. A shift out of 12-packs into single-serve cans lowers dollars while the unit count barely moves.
• Beverage alcohol has two peaks, not one. The summer peak is beer and ready-to-drink. The December peak is spirits and wine, at a much higher average ring.
• Non-alcoholic beer, wine and spirits is the fastest-moving line in most sets and behaves nothing like the rest of the department, with its own seasonality, price ladder and placement argument. It gets its own read rather than a paragraph here.
• Point of sale is transactions, not people. It cannot tell you which generation drinks less, and a chart that claims otherwise is reading a panel, not a register.

Reading alcohol sales trends: units before dollars

Start with a level read before any trend read, because half the arguments about alcohol sales trends are actually arguments about a denominator. This is the 38-store set averaged over 52 weeks, one row per segment, stated per store per week.

Beer and flavored malt is 48.6% of department dollars here and 47.3% of units, which is close enough that either denominator tells roughly the same story. Spirits is not: 12.8% of units and 19.6% of dollars. A department managed on dollars over-weights spirits, a department managed on units under-weights it, and a cooler reset planned on the wrong one puts the space in the wrong place.

Ready-to-drink runs the other way, at 16.7% of units and 13.1% of dollars. It is the segment most likely to look flat in a dollar report while it takes real space and real facings, so read it in units and in cooler doors.

Pack migration is the trend under the trend

Now the year-over-year read, on beer, which is the segment with the most pack variety and therefore the most room for a mix effect to masquerade as demand. Both years below are costed at the current shelf price, so what moves in the dollar column is mix rather than inflation.

Beer units fell 1.1%, from 183 to 181. Beer dollars fell 3.0%, from $2,660 to $2,581. The gap between those two numbers is the average selling unit, which lost 27 cents, from $14.53 to $14.26, because the single-serve can and the 15 and 18-pack grew while the 6-pack and the 12-pack gave ground.

That is two shoppers moving in opposite directions inside one line of a report. One is buying a colder, smaller, more immediate pack more often. The other is trading up to the large pack on a stock-up trip. Averaging them produces a department trend that describes neither, and a buyer who cuts the 6-pack because 'beer is soft' has cut the only pack that was actually losing.

The practical move is to run the same table at each year's own prices as well, which separates mix from inflation, and to plan the shelf against pack rather than against brand. That is the price pack architecture question, and beer is where a retailer sees it earliest because the pack ladder is longest.

Two peaks, and they belong to different segments

Seasonality in this department is strong enough that a quarter-over-quarter comparison is close to meaningless. Index dollars per store per week against the 52-week average, weighted by the number of weeks in each window so a long month does not read as growth.

The peak windows run 51% above the January and February trough. The two 115s are the interesting part: same index, different department. The summer peak is carried by beer and ready-to-drink, which blend to a $13.37 average ring in this set. The December peak is carried by wine and spirits, which blend to $15.18. At $6,103 a week that is roughly 457 units in July and 402 in December, about 12% fewer units for the same money, which is a different number of cases through the back door and a different amount of labour.

So split the index by segment before planning against it. An operator who staffs and orders December against a summer-shaped forecast over-orders beer and runs out of the gift-pack spirits that carry the week.

Non-alcoholic, and the published department numbers

Non-alcoholic beer, wine and spirits sits inside this department rather than beside it, at 2.0% of department dollars in the store set above and the fastest-growing line in it. It does not behave like the rest of the department: its peak is January, it prices above the alcoholic equivalent at every pack, and where it sits on the fixture moves it more than assortment does. That read has its own page, worked through the same stores: non-alcoholic beverage trends.

For context measured on a different universe: NIQ reported non-alcohol beer, wine and spirits at $925 million in US off-premise sales and 22% growth year over year in its 2025 report on the category, and projected it past $1 billion by the end of that year. Its 2025 Beverage Alcohol Year in Review, measured over the 52 weeks ending January 3, 2026 against the same weeks a year earlier, confirms the segment passed $1 billion, and reports that beer, wine and spirits all posted dollar declines driven by sustained volume softness.

The same NIQ review puts ready-to-drink at over 12% of total alcohol dollars and calls it the most reliable growth engine in the department, which is close to the 13.1% the store set above shows. Two very different universes landing within a point of each other is a reason to trust the direction. It is not a reason to quote either decimal as if it described your stores.

What point-of-sale cannot tell you about drinkers

Every few months a chart circulates claiming a generation drinks less, sourced to retail sales data. Retail sales data cannot support that claim. A scan is an item, a price, a time and a store. It carries no age, no household, no income and no intent, and aggregating a million of them does not create any of those fields.

What a register does support: which packs moved, which price tiers moved, which weeks moved, which stores moved, and what else was in the basket at the same time. That last one is genuinely useful in this department, because the alcohol basket is one of the largest in the store and the attachment read tells you what the trip was actually for.

If the question is who, you need loyalty-linked demographics or a survey panel, and the honest version of the answer names which one you used. See household panel versus consumer panel data for what each instrument measures, and syndicated versus panel data for why the two are not interchangeable.

Running this read in your own data

• Fix the unit before anything else. A 12-pack is one selling unit and twelve equivalized units, and a department that mixes the two produces a trend that is an artifact of pack mix. See equivalized volume.
• Hold price constant across both periods when the question is mix. Cost both years at the current shelf price, then run it again at each year's own price, and the difference between the two answers is inflation.
• Split by segment before splitting by anything else. Beer, wine, spirits, ready-to-drink and non-alcoholic have different pack ladders, different rings and different weeks, and a department-level growth rate is the average of five unrelated stories.
• Index seasonality on dollars per store per week rather than on period totals, so a five-week period does not read as a 25% lift.
• Decompose every fast-growing line into items per store and units per item. Growth in the first is a shelf decision you already made. Growth in the second is demand.
• Check promoted and non-promoted weeks separately. In a department this promoted, an item whose non-promoted weeks run well under half its promoted weeks is renting volume rather than building a base.

Where Scout fits

Scout ingests store-level point-of-sale directly and holds units, pack, price, promotion and item count in one model, so the reads above are cuts rather than projects: velocity by segment, a constant-price mix split, a seasonality index by store group, and items-per-store next to units-per-item for any line that is moving.

Two boundaries worth stating plainly. Scout reads transactions, not people, so it will not produce a demographic or generational claim about drinkers, and any page that offers you one from POS alone is overselling. And for an operator who also sells fuel, Scout reads the forecourt as well as the inside store, but they are separate instruments: every figure above is a merchandise and foodservice number, not a fuel one.

Related: SPINS for beverage and the alcohol / non-alcohol split, what BWS means and how the category splits, and sales velocity.