Allbirds. Gymshark. Mejuri. Brooklinen. Death Wish Coffee. Five brands that between them have earned millions of organic and social visitors. We scanned every one through our AEO engine on 2026-06-14. Not one scores above 60. Not one scores above zero on answer-readiness — the dimension that decides whether ChatGPT quotes you.
AI answer engines are eating the top of the purchase funnel. When someone types “best sustainable sneakers under $120” into ChatGPT or “what sheets do interior designers actually use” into Perplexity, they get a direct answer with two or three cited stores — not ten blue links. If your brand isn’t one of those citations, the sale is gone before you ever knew the question was asked.
We built the Hatchloop AEO engine to measure exactly how visible a Shopify store is to these AI systems. It fetches the same public signals an AI crawler sees — your llms.txt, product feed, JSON-LD schema, robots.txt, and homepage content — and scores five sub-dimensions from 0 to 100. We picked five well-known DTC brands with real marketing budgets, ran the engine live, and are reporting the numbers without modification.
| Brand | Score | Grade | Answer-readiness | Biggest gap |
|---|---|---|---|---|
| Allbirds | 47 | D | 0 / 100 | Zero JSON-LD schema; robots.txt blocks AI crawlers; all product descriptions under 30 words |
| Gymshark | 55 | C | 0 / 100 | Zero JSON-LD schema; no FAQ or answer content; 0% image alt text |
| Mejuri | 10 | F | 30 / 100 | Products.json returns 404; no llms.txt; 20 locales, zero AI declarations per market |
| Brooklinen | 31 | F | 0 / 100 | Product feed too large to parse (feed score = 0); zero JSON-LD schema; no FAQ content |
| Death Wish Coffee | 56 | C | 0 / 100 | No FAQ schema or answer content; missing Product/Offer/Rating JSON-LD; 64% “Default Title” variants |
The grade scale: A ≥ 80 · B ≥ 65 · C ≥ 50 · D ≥ 35 · F < 35. No brand in this set scores above C. And critically, four out of five score exactly zero on answer-readiness — the sub-dimension that controls whether an AI engine will quote the brand when a buyer asks a category question.
Gymshark has done the technical AEO infrastructure work: they have llms.txt, agents.md, an agentic sitemap, and a robots.txt that allows AI crawlers. Their product descriptions are solid — 72% pass the 60-word quality bar. The gap is entirely in schema and answer content. There is zero JSON-LD on the homepage, zero FAQ schema, and zero answer-oriented content that an AI engine could quote. One additional gap the engine flagged: 0% image alt text across the product feed.
Mejuri is instructive because it is a genuinely sophisticated, design-forward brand — and it still scores 10. Here is why. Their /products.json endpoint returns HTTP 404: they have locked their public product catalog, almost certainly for competitive reasons. The engine assigns feed completeness = 0. There is no llms.txt, no agents.md, no agentic sitemap. Here is the specific irony: Mejuri serves 20 hreflang locales (en-us, en-ca, fr-ca, en-au, en-gb, de-de, and more) but declares zero per-market context for AI agents — so any AI assistant that surfaces Mejuri may quote the wrong currency or price. The only positive signals the engine found: Article JSON-LD on the homepage and 21 FAQ-style phrases in body copy (which explains the modest 30/100 answer-readiness despite no FAQPage schema).
llms.txt — AI assistants may quote the wrong currency. Adding per-market blocks is the single highest-leverage fix for Mejuri./products.json is a common move among enterprise Shopify stores protecting catalog data. It is a reasonable competitive decision — but it has an invisible cost: AI shopping agents cannot build a product understanding of the store, and citation likelihood drops to near zero for category queries.
Allbirds has done the work to set up llms.txt (86 lines, 11 headings, enriched), agents.md, and an agentic sitemap — a discoverability score of 90. But their robots.txt blocks AI crawlers. The engine found that 0% of sampled products pass the 60-word description quality bar, 0% have alt text, and there is zero JSON-LD schema anywhere on the homepage. The infrastructure file work is genuinely valuable but is being undercut by the crawler block and thin catalog data.
Brooklinen’s main issue is structural: their /products.json feed is too large for the engine to parse cleanly, resulting in a feed completeness score of zero — the same penalty as Mejuri’s 404. Combined with zero JSON-LD schema on the homepage and no FAQ content, the overall score bottoms out at 31. This is a fixable problem: paginating the product feed or using the ?limit= parameter to return a parseable subset would restore feed completeness.
Death Wish is the best performer here, and the gap between them and the others is instructive. They have perfect discoverability (100), strong product descriptions (92% pass the 60-word bar, 100% have SKUs), and two JSON-LD types — Organization and WebSite. These are real wins. But they have no FAQ schema or answer content at all (answer-readiness = 0), and critically: 64% of their products use “Default Title” as the variant name. An AI agent trying to answer “what grind size does Death Wish Coffee come in?” gets nothing useful from “Default Title”. Structured variant naming and FAQ content are the two fixes that would push them from C to A.
Every brand in this set has invested in SEO, content, and distribution. Several have large engineering teams. None of them has cracked AEO. The consistent pattern across all five:
/llms.txt, /agents.md, /sitemap_agentic_discovery.xml, /products.json?limit=25, and /robots.txt. It computes five weighted sub-scores (discoverability 25%, feed completeness 25%, schema coverage 22%, answer-readiness 16%, multi-market hygiene 12%) into a single 0–100 number. No scores were hand-tuned or estimated. This is a public-surface scan only: it grades what AI crawlers see from outside the store, not authenticated or server-side data, not live citation frequency across AI engines. All stores confirmed as Shopify by platform detection signals. Raw engine JSON available on request.
Run the same engine against your own storefront. Free, no signup, results in under 60 seconds. You’ll get a 0–100 score across all five dimensions, plus a ranked list of the fixes that would move your number the most.
Check your AI-visibility score →Free · No account needed · Works on any public Shopify storefront