Teardown9 min read

How ChatGPT recommends cold brew: the most wide-open category we’ve measured

No default, niche-friendly, blog-and-Reddit-driven — and a buy box that runs through the grocery aisle. The most winnable category for a challenger.

Cold brew has no king. Where business software has HubSpot, ChatGPT’s cold-brew answers are a genuine free-for-all — ask the same question three times and you can get three different winners. That’s the defining fact of the category, and for a challenger or DTC brand it’s the opportunity: there’s no incumbent baked into the answer to dislodge.

BrandLedAvg shareSentimentScore
Lifeboost234%+1.5
Blue Bottle229%+1.7
Stumptown228%+1.2
Chameleon227%+1.1
SToK124%+0.9
La Colombe118%+1.2
Starbucks026%+0.5
championrecommendedneutralcautioned
Aggregate across all intents. The leaders are small and DTC; Starbucks is named across several intents but never leads and is never championed.

Seven questions, almost no repeat winner

Across the intents we ran — brands, ready-to-drink, low-acid, concentrate, decaf, organic, flavored — the leader was different almost every time. “Best cold brew coffee brands” crowned three different winners across three runs (La Colombe, Stumptown, Blue Bottle). “Ready-to-drink” produced four winners across four runs. “Low-acid” gave Lifeboost the edge but Stone Street, Chameleon and Puroast all took a run. Concentrate went to Chameleon, organic to Wandering Bear, decaf to SToK, flavored to Dunkin. There is simply no anchor.

best cold brew coffee brands” · 3 live runs · leader: 3-way split
Blue Bottle30% · led 1/3
La Colombe28% · led 1/3
Stumptown28% · led 1/3
Starbucks25%
SToK24%
Pick an intent — note how often there’s no single winner, and how often the leader is a small or specialist brand rather than a household name.

Niche brands win; the giants are named but never championed

The brands that lead are mostly small or DTC — Lifeboost, Blue Bottle, Stumptown, Bare Brew, Wandering Bear. Starbucks tells the story in the sentiment: it appears across several intents but never leads and is never spoken of warmly (a neutral score, run after run). In a fragmented, taste-driven category, brand size buys you a mention, not a recommendation.

“In cold brew, being big gets you named. It doesn’t get you recommended.”

Tiny coffee blogs decide it — and one brand’s own blog won

There’s no Runner’s World here. ChatGPT leaned on Reddit and a cluster of small coffee blogs — Drop of Drink, Bean Ground, Roast Critic, Coffee Scan. The striking one: drinkbarebrew.com — Bare Brew’s own blog — was a cited source, and Bare Brew got championed. In a thin category that mainstream media ignores, your own content can become the source ChatGPT trusts. That’s an opening you don’t get in crowded verticals — and it makes generative engine optimization (GEO) unusually winnable for an FMCG brand, because your AI search visibility can be built with content you control.

ChatGPT· most-cited sources across all cold-brew intentslive
reddit.com
60
dropofdrink.com
40
foodnetwork.com
30
beanground.com
30
roastcritic.com
30
coffeescan.com
30
Source counts from live Chaice analyses — small coffee blogs and Reddit, not mainstream media.

ChatGPT rewrites the question — and goes to Reddit

It adds the year, “reviews,” the format, and — tellingly — writes “reddit” right into its own search (e.g. “best cold brew concentrate reviews reddit 2025”). Your Reddit presence isn’t a nice-to-have here; it’s effectively a ranking factor the model reaches for by default.

There is a buy box — and it runs through the grocery aisle

Unlike running shoes, cold brew lit up a shopping panel built on grocery and mass retail — and this time the giants are present: Target, Walmart and Whole Foods led the sellers. Whether the big retailers appear in ChatGPT’s buy box is category-specific: shoes route through specialty chains and exclude them; cold brew routes through grocery where they dominate. Your buy-box visibility is a grocery-feed problem.

ChatGPT· who appeared as the seller, by number of offerslive
Target
15
Walmart
12
Whole Foods Market
10
Sprouts
8
Hy-Vee
6
Kroger
6
The grocery and mass retailers ChatGPT surfaced — the exact giants absent from the shoe shelf.

And the prices swing hard: the same Starbucks cold-brew concentrate ran from $8.97 to $15.38 across stores in one answer. The leading offer is whichever feed ranks highest on price and availability — winnable, but not yours by default.

Starbucks Cold Brew Concentrate$8.97–$15.38
Pop & Bottle Classic$10.19–$15.49
Wandering Bear Straight Black$12.40–$44.99
Chameleon Organic Concentrate$9.99–$11.99
Bizzy Organic Espresso$6.99–$9.99
The same cold brew at different grocery stores, in one answer.

The most volatile category we’ve measured

Repeating the headline intents shows near-total leader churn — three runs of “best cold brew coffee brands” produced three different winners; “ready-to-drink” produced four. A single check here is meaningless. You have to sample heavily and read the distribution, not the snapshot.

What to do if you sell an FMCG product

Treat it as a land-grab

There’s no default to dislodge and the leader rotates every run, so broad, sustained presence beats chasing a single #1. Show up everywhere and you’ll lead some share of the time — which, over many runs, is the win.

Build the content ChatGPT reads — including your own

Tiny coffee blogs and Reddit decide this category, and one brand’s own blog became a cited source. Invest in a genuinely useful brand blog and an authentic Reddit presence; here, unusually, that works.

Win a quality, not a size

The giants get named but never championed. Own a specific, ownable quality — low-acid, organic, strongest — because that’s how the niche brands keep taking the lead.

Get your grocery feeds clean

The buy box is Target, Walmart, Whole Foods and Sprouts. Make sure your best-priced grocery listings are complete and accurate.

See what AI says about your brand

Run your own prompt through the live Chaice console and get the same breakdown — attributes, brands and sources — in seconds.

Try the console
#GEO#generative engine optimization#AI visibility#AI search#FMCG#teardown#cold-brew
AJ
Aaron Joyce

Founder of Chaice. Reverse-engineering how AI engines decide what to recommend — and turning it into things you can ship.

Keep reading