One diagnosis, three very different symptoms. The market average health score is 35 out of 100 — but what that 35 means depends entirely on the niche.

web3 & crypto — strong products, invisible brands

The most paradoxical segment: strong products, real volume, systemic invisibility.

typical values across audits
metrictypical value
average GEO health score37/100
branded recall~100% — AI knows nearly every brand
category recall~0 — "best crypto card / DEX / wallet" repeats the same 3–4 rival names
Wikidata / Wikipedia0 of audited brands
llms.txtpresent at ~20% — the best score of any niche, still a failure
typical organic traffic10–200 visits/mo, against products doing millions in volume

The product-versus-visibility gap is the widest on the market. We audited products with billions in volume whose visibility was single-digit visits and zero AI citations. Web3 builds product before distribution, and AI punishes it mercilessly.

Every category answer is already held by three or four brands — and they are not always the best products. They are the ones sitting in the listicles, with machine-readable whitepapers.

Name collisions are the chronic disease. Short, clever names collide with words in other languages, other products and meme brands. In one audit, 99% of a brand's traffic turned out to belong to someone else's similarly-named product.

Categories are still cheap — and repricing fast. In several audits the category's visibility leader collected under 1,000 visits a month. One audited category grew 16.8× in 24 months; entry into AI answers by 2027 is priced at 2–4× today's cost.

iGaming — AI doesn't ignore you, it warns people about you

The harshest segment. Here models don't merely stay silent about brands — they caution users against them.

typical values across audits
metrictypical value
average GEO health score24/100 — the market's worst
category recall0 of 10 live prompts
negative AI tone ("high-risk", "possible scam")2 of 3 audited brands
schema / JSON-LD0/100 across all audited

In a category where trust is the product, an unverifiable brand does not get a neutral answer. It gets a warning.

B2B services — world-class portfolios the machine can't read

Agencies, studios and consultancies with genuine track records and awards — none of it in a form a model can parse. The work exists as PDFs, case decks and prose; the machine needs entities, markup and named people.

what it means wherever you sit

If your category inside AI answers is already taken by three names, or still free, the entry price today is the same and it is minimal: entity anchor, schema, readability, and two or three placements in the sources the model cites. In a year the same entry becomes a full-scale SEO war.