In August 2026, Kanyon Studio ran this same audit on 15 well-known San Diego businesses across dentistry, med spas, fitness, real estate, and restaurants. Use these benchmarks to read your own score.
Above 70 puts you in San Diego's top tier. Anything under 50 means AI assistants are struggling to describe your business at all. Read the full study →
Every issue this audit flags is fixable. Here is exactly how, in the order that moves your score most. Each fix is doable yourself if you can edit your website; all of them are included in a flat-rate AEO sprint.
Why it matters: the title tag is the first thing every AI system and search engine reads. “Home” or your business name alone tells them nothing.
The fix — use this formula:
Keep titles under 60 characters and meta descriptions under 155, and give every page its own. DIY: ~30 minutes.
Why it matters: AI can only recommend what it can quote. If your homepage never says in plain text what you do and where, the assistant guesses — or skips you.
The fix — the one-paragraph test: your homepage's first text block should read like this:
Real words in real HTML — not text baked into images or a hero video. DIY: ~20 minutes.
Why it matters: “how much does X cost in San Diego” is one of the most common questions AI assistants get. If your site has no number, the answer cites a competitor who published one. Only 4 of the 15 San Diego businesses we studied showed any price.
The fix: you don't need a full price list — a floor is enough:
Put it on the relevant service page in plain text. DIY: ~15 minutes.
Why it matters: AI assistants lift question-and-answer text almost verbatim. An FAQ is the single easiest format for them to cite.
The fix: add 4–6 real questions customers actually ask, each answered in 2–4 full sentences that stand alone. Start with these:
Then mark it up with FAQPage schema (see #5). DIY: ~1 hour.
Why it matters: schema is machine-readable labeling — it tells AI systems “this is a business, here's its name, area, prices, and FAQs” with zero ambiguity.
The fix — paste this in your site's head code and fill in the brackets:
This one needs access to your site's head code — the most common fix we handle in sprints.
Why it matters: llms.txt is a plain-text map of your site written for AI crawlers — like robots.txt, but for assistants. Only 3 of the 15 San Diego businesses we studied had one, which makes it the cheapest competitive edge on this list.
The fix — create a text file at yoursite.com/llms.txt:
DIY: ~1 hour if your platform lets you host a root text file.
AI engines recommend businesses they can actually read. The audit fetches your site the way an AI crawler does and grades the four things that decide whether ChatGPT, Perplexity, and Google AI Overviews can cite you.
Whether your business, services, and prices exist as machine-readable schema — the format AI engines trust most when deciding who to recommend.
Whether your pages contain direct, quotable answers to the questions buyers actually ask, or bury them in marketing copy AI skips over.
Whether AI crawlers are allowed in at all, and whether files like llms.txt and your sitemap guide them to the pages that matter.
Whether each page states plainly what you do, where you do it, and what it costs — the details AI engines quote in answers.
01 — SCAN
The audit fetches your homepage exactly the way an AI crawler sees it — no styling, no images, just the structure and signals engines parse.
02 — SCORE
Each signal is graded against the checks AI engines rely on when they choose which businesses to cite, then rolled into one visibility score.
03 — FIX
You get the specific gaps holding your score down, ordered by impact — so you know exactly what to fix first, whether you do it yourself or hand it off.
Yes. It runs instantly in your browser, takes about 60 seconds, and there is no obligation attached to the results.
The score reflects how easily AI engines can read, trust, and quote your site. A low score usually means missing structured data, no direct answers to buyer questions, or crawl signals that keep AI engines out.
Some fixes — clearer titles, plain-language answers on your pages — are very doable yourself. Schema markup, llms.txt, and FAQ structured data usually require access to your site's head code. Kanyon Studio offers flat-rate optimization sprints that handle all of it, typically live in 2–4 weeks.
SEO earns your pages a position in a list of results. AI optimization makes your business quotable inside the answer itself — when someone asks ChatGPT or Perplexity who to hire, the engine recommends businesses whose sites it can parse and verify. The two overlap, but the deliverables are different.
