48.3% of consumption happens in one setting. It's not where you'd guess.

Bar or restaurant? 0.6%. We went through 2,000 written responses to find out where consumers actually drink — and why.

Part of the Functional Beverage Study series — findings from Cohort 1, a real-world, longitudinal study of 2,580 consumers across 20 products.

This one isn't just quantitative. We went back through nearly 2,000 written responses — product reviews and alcohol-replacement reasoning — to understand the occasion behind the numbers. The qualitative data told the same story as the quantitative data. Louder.

The punchline

Across the full cohort (N=24,021 reported consumption days), nearly half of all functional beverage consumption — 48.3% — happened at the end of the day to relax. Another 22.8% happened at home with a meal. Bars and restaurants accounted for 0.6%. Social events: 4.4%. The dominant use case for functional beverages is not social. It's personal wind-down.

Where they drink

Participants reported their setting every time they consumed the product:

• End of the day to relax: 48.3%

• At home, with a meal/food: 22.8%

• Alone: 12.6%

• Outdoor activities: 5.1%

• Social setting / social event: 4.4%

• Other: 3.2%

• Watching sports: 3.0%

• Bar or restaurant: 0.6%'

83.7% of consumption is some version of at-home, evening, low-key, often solo. When participants selected "Other," they described working from home, watching TV, cleaning the house, playing video games, cooking dinner, before bed. Domestic, quiet, routine.

Median use time: 5:30 PM — with the middle 50% of consumption falling between 2:53 PM and 7:11 PM. 71.7% used the product once per day.

What they told us

We asked 1,970 participants to write a product review and 1,972 to explain why they would or wouldn't replace alcohol with the product.

Across 20 different products, unprompted, they described the same ritual:

• "I came to really look forward to my little end of day ritual of winding down with it."

• "Satisfies my desire for something to relax me after work without the incapacitating effects and without the hangover."

• "The ritual is the same but the consequences are better."

Where alcohol still wins — and they drew the line themselves:

63.6% said there are settings where they'd prefer alcohol. Top two: social events (22.4%) and bars/restaurants (22.0%).

581 participants said they'd "sometimes" replace alcohol. Their explanations drew a remarkably consistent line:

• "Alcohol for social parties, [the functional beverage] for nighttime relaxing."

• "I like my alcohol drinks especially when out with friends. I like the [functional beverage] when at home and relaxing."

• "I didn't feel social when drinking it so I would drink it at home instead of wine, but I would never drink it in public, or if I was going out."

Consumers have already sorted this for themselves: functional beverages own the wind-down, alcohol owns the social occasion. They're not confused about it. The industry is.

What this suggests

1. The industry has a setting mismatch. Scan the marketing from any ten functional beverage brands. Count how many reference bars, parties, social occasions. Now compare that to where 83.7% of consumption actually happens. Consumers articulated the gap clearly — brands haven't caught up.

2. "Alcohol alternative at the bar" has a data ceiling. The substitution is real — separately, daily alcohol probability dropped 12.7 percentage points during the product-use phase. But participants told us the substitution happens at home at 5:30 PM. The settings where consumers prefer alcohol are the exact settings the category keeps trying to win.

3. The wind-down occasion is unclaimed. Tea owns relaxation. Wine owns the evening pour. Functional beverages have 24,021 data points and nearly 2,000 written descriptions showing consumers already use them to decompress — and almost no brand is intentionally claiming that occasion. When consumers use words like "ritual" and "routine" unprompted, that's not satisfaction. It's habit formation.

Method note: Self-reported, real-world observational dataset (Cohort 1; N=2,580; 20 beverages).