For a hundred years, Cheerios owned the answer to "what cereal should I buy." Not because it was better. Because it was there. Shelf, TV, search — the three arenas that decided the sale.

There is a fourth arena now. And Cheerios is losing it.

5W AI Communications this month published the first grocery-focused AI visibility study — the Grocery edition of The Private Label AI Advantage. We measured the Citation Share of 25 grocery brands inside ChatGPT, Claude, Perplexity, Gemini, and Google AI Overviews. Read the full index here.

The headline:

  • Store brands: 44. National brands: 23. Nearly two to one.
  • Trader Joe's (78) and Kirkland (74) at the top of the whole list.
  • Good & Gather, launched in 2019, already out-citing Cheerios.
  • Tyson, Philadelphia, Frosted Flakes, Hellmann's, Campbell's — bottom third.

That is not a rounding error. That is the biggest names in American groceries getting outranked, inside the platforms where more than a third of consumers now start product research, by the private labels they used to define themselves against.

This Is Not About Product

Nobody is arguing Kirkland ketchup is better than Heinz. That is not the question the AI is being asked. The question is: what should I buy. And the AI is answering with the store brand.

Because the corpus around the store brand is bigger, better-structured, and more editorial. Trader Joe's has The Kitchn writing about it every week. It has ten years of Reddit threads. It has Bon Appétit calling out specific products by name. Cheerios has a media plan.

The AI engines read the corpus. They do not watch the ads.

This Is What the AI Communications Era Looks Like

National CPG spent sixty years optimizing for memory — for the second in front of the shelf when a shopper reaches without thinking. Retrieval is a different sport. It rewards the brands that show up in editorial, in comparison content, in named publications, in community threads. It rewards specificity, third-party validation, and structured entity presence. It punishes brand claims that only appear on brand.com.

Store brands built exactly that footprint. By accident, in most cases. Now the AI reads it and rewards it.

The Category Was Already Moving

Private label hit $282.8 billion in the U.S. in 2025 — a record. 21.3% of the dollar share. Growing nearly three times as fast as national brands. AI is not creating that shift. AI is accelerating it.

And the acceleration compounds. Every AI answer that recommends Kirkland over Heinz becomes a training signal. Every Reddit thread comparing Trader Joe's to a CPG competitor becomes retrievable context. Every editorial round-up that names the store brand as the winner deepens the citation graph.

National CPG is not just behind. It is falling behind faster every quarter it does not act.

What CPG Has to Do

Stop measuring unaided awareness as if it is the only awareness that matters. Start measuring Citation Share. Every marketer in America can name what percentage of consumers know their brand. Almost none can name what percentage of AI answers cite it. That is the number the next decade will be won or lost on.

Build the corpus. Editorial coverage in named publications. Comparison content. Category-level authority — not brand-level puffery. Show up in the community threads. Let yourself be compared. Refusing to be compared is refusing to be cited.

And invest in Generative Engine Optimization the way you invested in SEO twenty years ago. Same seriousness. Bigger stakes. The engines are still forming their retrieval habits. The brands that build authority into them now will own the answer for a decade.

The Bigger Point

The chatbox is the new checkout. More people ask AI what to buy than ask Google. The whole system that CPG built to win the aisle was engineered for a shopper standing in front of it. That shopper is now standing in front of a screen — and the screen is telling her the store brand is the smart pick.

Trader Joe's did not out-spend Cheerios. It out-showed-up. That is the entire lesson of the AI Communications era, in one grocery study.

The next four editions — Amazon and e-commerce, pet, pharmacy and OTC, supplements — publish weekly. I expect the pattern to hold. And I expect the CPG conversation to change.

The earned-only AI visibility play is over. So is the paid-only one. The brands that win are the ones that treat the whole corpus around them — trade press, editorial, comparison content, community — as the media plan.

Everything else is nostalgia.

Full study: 5wpr.com/ai-visibility-index/private-label-ai-advantage/grocery

PR Newswire release: First 5W Study of Grocery-Brand AI Visibility: Trader Joe's Beats Cheerios, Kirkland Beats Heinz