Most CPG brands assume that being the category leader means showing up first, everywhere, including in AI search. That assumption is wrong, and it is costing brands market share they do not even know they are losing.
On May 27, Barrel hosted a live webinar with partners from Shopify and Good Protein to talk through what it actually takes to get cited by AI search tools. One example from that conversation captures the reality of Generative Engine Optimization (GEO): picture a blogger in Nebraska, with no meaningful following, who writes a single post comparing protein powders. That post could end up cited constantly by AI assistants, sometimes ahead of the brands it reviews. It's a small, specific scenario, but it's playing out across categories right now. If you'd rather watch the full conversation, download the free recording here.
Here's what that story reveals about off-site content, and what to do about it.
If you're the biggest name in your category, AI tools will often recommend you by default. That's the good news. The bad news is that default position is not permanent, and it is not earned by size alone.
"If you are the category leader, AI will recommend you by default. If you are not, you need a reason to be cited instead," said Lucas Ballasy, CEO of Barrel, during the session. "That reason is almost always content quality and specificity."
Challenger brands that publish detailed comparison guides, transparent ingredient sourcing stories, and thorough product education are already outperforming better-known competitors in AI-driven recommendations. The playing field is more level than most established brands realize, and that levelness runs in both directions. A market leader who does nothing here can be displaced by a smaller brand that shows up consistently in the answer.
Lucas described a conversation with a beverage brand that leads its category but does almost nothing in this space. Their question was reasonable: if we lead the market, don't we lead in AI search too? The answer is no. Smaller brands doing the work today are slowly becoming the answer to the questions that customer is asking, and given enough time, that is how market leadership actually shifts.
Matt Hickey, Partner Solutions Engineer at Shopify, made a related point about how this plays out for brands competing on the same platform: every Shopify merchant's product catalog is available to these agents, competitors included, so no one is shut out by default.
"It's a matter of having the structured product data and the relevant sources to make sure that you're outranking your competitors in that space," Matt said. Being on Shopify gets you into the conversation. It doesn't win the conversation for you.
During the webinar, Lucas pointed to something most brands underweight: your off-site footprint. AI models don't just read your Shopify storefront. They read your Amazon listings, your Instacart descriptions, press coverage, podcasts, Reddit threads, and any blog that happens to write about you.
That Nebraska blogger owns a piece of her subject brand's content online that the brand itself never claimed. When someone asks an AI assistant for a recommendation in that category, her post is part of what shapes the answer, whether the brand likes it or not.
Reddit is a particularly clear example. Lucas pointed out that if you search your own brand name on Reddit right now, there's likely a real conversation happening, and it may not be one you'd choose to have. Brands that ignore this are ceding the narrative to whoever shows up. Brands that contribute to the conversation, or work with a PR partner who can, have a say in what gets said.
This connects to a point Jean-Sébastien Crevier, SVP of Growth at Good Protein, raised about reviews specifically. On-site reviews matter to shoppers browsing a product page, he said, but he was skeptical they carry much weight with AI tools.
"There's a line in the sand between on-site reviews and off-site reviews," Jean-Sebastien said, pointing instead to editorial coverage and third-party writeups as the content actually shaping what an AI assistant says about a brand.
Matt agreed, noting that on-site product reviews can be easy to spam and harder for a model to trust, while an article in a publication like a health or lifestyle outlet carries more credibility.
"Having those external sources that really rave about the products and really highlight the brand experience... I think is going to provide a lot more traffic and a lot more credibility for the agents," Matt said. The Nebraska blogger fits squarely into that category: an independent, specific, off-site voice that an AI model has learned to trust.
There's a subtler version of this worth knowing about. Some brands or their competitors build separate content hubs, or write blog roundups, that rank competitors against each other, often with the publisher's own brand placed just above yours. It's a tactic that predates AI search, but it matters more now: if that content is what AI models are reading and citing, it becomes the narrative customers hear, regardless of whether you agree with it.
The fix isn't to ask the AI to stop citing it. As Lucas put it, "you cannot ask the AI to stop. You can only create better content that earns more citations. The brands winning this game are playing offense."
A few starting points pulled directly from the session:
This is one topic among many covered in the full session, including how Good Protein approaches this at the operator level and a broader framework for auditing your current content. Watch the full 45-minute recording here.
Barrel builds and optimizes Shopify storefronts for CPG brands, including the content and off-site strategy that determines whether AI tools recommend you. If you want to talk through what this looks like for your brand, reach out at barrelny.com/contact.
%20(1).jpg)
.jpg)
