Coffee Industry Trends: Hot vs Packaged

Most coffee industry trends reporting measures one of two things: what people say they drink, from consumer surveys, or what scans through a register with a barcode on it, from syndicated retail measurement. Between those two sits the largest coffee business in a convenience store, and neither instrument can see it, because a cup poured at a dispenser has no UPC.

That gap is the whole subject of this page. Hot dispensed against packaged, cup size and daypart, what the cup pulls into the basket with it, and what happens to the mix when you move one rung of the price ladder. It is a read that only first-party transactions produce, and it is the one that decides how a coffee program is run.

Key takeaways

• In a convenience store, hot dispensed coffee is the overwhelming majority of coffee units and coffee gross profit, and it is invisible to every syndicated coffee dataset.
• Price per fluid ounce falls as cup size rises, so cup mix moves gross profit more than a headline price change does.
• The biggest daypart is not the most valuable one. Attachment and basket ring peak in the middle of the day, on a fraction of the cups.
• Move one rung of the ladder and the mix moves to the rung you did not touch. Measure the blended ring, not the price you changed.
• Consumer surveys measure stated behavior and retail scan measures barcoded packages. Neither is a substitute for the dispenser.

What the published coffee numbers actually measure

The best public source on US coffee consumption is the National Coffee Association's National Coffee Data Trends report, and it is a survey. The Spring 2026 edition was fielded 5 to 20 January 2026 among a nationally representative sample of 1,850 US adults aged 18 and over, and it reported that 66% had drunk coffee in the past day and 73% in the past week, at an average of 2.8 cups a day, which the NCA put at roughly 195 million weekly adult coffee drinkers and around 507 million cups a day.

Two figures from the same report matter most to a retailer. Past-week specialty coffee consumption reached 58%, up from 57% in the Fall 2025 edition, against 62% for traditional coffee, which was flat. And espresso-based drinks reached 45% past-week, up from 40% in 2022, with lattes at 21% and espresso at 20%. On preparation, 85% of past-day drinkers had coffee prepared at home and 36% had coffee away from home, split across the workplace at 14%, in transit at 13% and eating places at 10%.

Hold onto what that is: self-reported behavior from 1,850 people. It is a good directional instrument and a bad basis for a cup-size decision at 124 stores. The retailer-side context comes from NACS, which reported for 2025 that foodservice was 28.5% of convenience in-store sales and 38.9% of in-store gross profit, with prepared food making up 73.9% of foodservice sales and the balance coming from commissary and hot, cold and frozen dispensed beverages. Packaged beverages were a separate 18.7% of in-store sales.

Neither source can tell you what your dispensed coffee did last week. That is not a criticism of either. A survey measures people and a syndicated panel measures barcoded packages, and the hot cup is neither.

The two coffees in one store

Here is the split from transactions. A 124-store convenience operator, 52 weeks, everything coffee in the building.

Hot dispensed is 88.5% of the coffee units, 75.2% of the coffee dollars and 87.7% of the coffee gross profit. The barcoded part, packaged plus RTD, is 11.5% of the units and 12.3% of the gross profit, and it is the only part a syndicated coffee report can see. That ratio is the reason a c-store coffee program run off category reports tends to be run backwards: the reports cover about an eighth of the profit in fine detail and seven eighths of it not at all.

The margin column is the other half of it. A 68.5% rate on the cup against 27.4% on the bag is not a small difference in kind, it is a different business. Packaged coffee is a distribution and price-competition category. Dispensed coffee is a foodservice operation with an equipment, waste and labor cost structure, and it should be reported next to the rest of foodservice rather than next to the packaged set.

Cup size and the price ladder

Inside the 812 cups is a price ladder, and it slopes the way every good ladder does: the bigger the cup, the cheaper the ounce.

That mix blends to $2.16 a cup and $1,754 a week per store. Now the interesting part, which is what happened when the operator moved one rung. Over the prior 52 weeks the 16 oz sold at $1.99 and the 12 oz at $1.79. The 16 oz went to $2.09 and the 12 oz was left alone.

The cup count fell from 823 to 812, down 1.3%, and dollars rose from $1,747 to $1,754, up 0.4%. But the mix moved underneath: the 12 oz went from 19.4% of cups to 21.9% and the 16 oz went from 44.6% to 42.0%. Two and a half points of share moved to the size whose price had not changed, and roughly the same amount left the size that had.

That is the reason a coffee price decision has to be measured on the blended ring rather than on the item you changed. The 16 oz price rose 5%, the blended ring went from $2.12 to $2.16, under 2%, and the difference is entirely mix. A program that reads the 16 oz line alone concludes the increase was fully banked. It was not, and the leakage shows up permanently in the cup-size mix rather than temporarily in the volume.

Daypart, because a cup at 6am is a different product

Coffee is the most daypart-sensitive line in a convenience store, and the mistake is to assume the biggest daypart is the most valuable. Same operator, same 812 cups.

The early-morning daypart is 44.1% of the cups and attaches a foodservice item to only 38% of them, the lowest rate of the three dayparts before mid-afternoon. The 11am to 2pm daypart is 11.8% of the cups, attaches to more than half, and rings $9.11 against $6.94 for the morning. The morning cup is a commuter transaction that mostly stands alone; the midday cup travels with lunch.

That is an operating decision, not an insight. Promotional pairing, brewing schedule, staffing of the food case, and where the cup lids sit relative to the food case are all daypart decisions, and a single blended attachment number hides every one of them. See attach rate for the metric's failure modes and daypart for the way the day usually splits.

One caution. Attachment measured this way is co-occurrence in a basket, not causation. The midday cup and the midday sandwich share an occasion; that does not prove either sells the other. It does justify testing the pairing, which is the level of certainty a merchandising test needs.

Where packaged and RTD coffee actually compete

The packaged set in a convenience store is not competing with the dispenser, and treating it as though it were produces bad decisions in both directions. It competes with the grocery shop, on a stock-up occasion, at a price the store will usually lose. Its job is to be present rather than to be cheap, and a set of a couple of dozen items turning 34 units a week is doing that job.

RTD coffee is a third thing again: a cold-vault immediate-consumption item competing with energy drinks for the same door space, on the same occasion, at a similar price point. It should be read against the cooler rather than against the coffee program. Beverage Marketing Corporation reported that US RTD coffee declined in both volume and dollars in 2024, in a year when total liquid refreshment beverage volume grew 1.0% to nearly 36.4 billion gallons, so the segment has been losing a fight it is actually in. Beverage industry trends has the full cooler read.

Coffee industry trends worth acting on, and how to falsify each

What this read cannot tell you

Transactions record a cup size, a price, a time and the other items in the basket. They do not record a person, a preference or a reason. So the read above supports cup mix, daypart mix, attachment, price-ladder movement and store-to-store comparison, and it supports no claim about who is buying coffee or why.

Where a genuine consumer question sits, use a consumer instrument and label it as one. The NCA figures quoted above are a survey of 1,850 people and are quoted here as exactly that, not as a measurement of purchases. Mixing the two, quoting a survey percentage next to a velocity figure as though both were transactions, is the most common way a coffee deck loses a room.

How Scout fits

Scout reads inside-store POS directly, including foodservice, so dispensed cups, packaged coffee and RTD sit in one model with their sizes, dayparts, prices and baskets attached. That is what makes the hot-against-packaged split above a standing view rather than a special project, and it is the read that no syndicated coffee source can produce for you at any price, because the cup never had a barcode.

The boundaries worth stating. Scout can hold the item file, maintain cost and retail and push a price file to the POS, and it recommends orders; it does not raise or transmit purchase orders, hold an order guide, or carry an EDI connection to your suppliers. For fuel operators it reads the forecourt as well as the inside store, but connecting a cup of coffee to a fill-up needs the forecourt transaction joined to the inside-store basket, which is a harder problem than either feed alone; nothing here makes that claim. And Scout's numbers are transactions rather than people, so it is not a substitute for consumer research.

Related: quick-service restaurant and daypart.