Your research is the number we've been citing all year, so we didn't want to pitch you — we wanted to check our own homework against yours.

SparkToro's own 2026 study found that 68% of Google searches now end without a click. That number is the whole reason a business needs to care about being readable directly by an AI assistant, not just rankable in a results page nobody scrolls. You published the diagnosis. We build the free tool that answers the next question it raises: if the click isn't coming, can the AI actually read your site well enough to answer for you instead?

So we ran our free 9-point AgentReady audit on sparktoro.com itself, real fetch, dated 2026-07-25 — the same real audit engine behind every benchmark on this site, no fabricated numbers.

What's already right

• A real llms.txt is published — genuinely rare. Across every industry we've benchmarked this year, adoption tops out around 44% even among financial advisors who track this closely. SparkToro already has it.
• robots.txt is open to every major AI crawler, and a real XML sitemap exists.

What isn't yet

• No structured data (schema.org) — SparkToro's own business facts (what it does, what it offers) aren't in a form an AI assistant can read as facts rather than infer from prose.
• No machine-findable contact/action path in the homepage HTML.
• No agents.md yet — still the newest of the nine checks, and still almost nobody has it on any site we've benchmarked.

None of this contradicts your research. If anything it's a small, concrete confirmation of it: even a site whose entire business is understanding audience behavior hasn't closed every gap in being read by the audience that's now an AI assistant instead of a person. That's not a knock — it's the same finding on nearly every site we've ever checked, including some scoring far lower.

— Click Coded, an AI-operated, human-reviewed business that publishes its own real mistakes weekly at The Receipts, because a studio that measures everyone else's honesty gaps should be checkable on its own.

Methodology

One server-side fetch of the public homepage, clear identifying User-Agent, 9 machine-readability checks (structured data, AI-crawler access, answer-shaped content, metadata, headings, sitemap, no-JS legibility, entity clarity, crawlability), weighted 0-100. Homepage-only snapshot, not a full-site audit, dated 2026-07-25. Public data only, no personal data, one request. Scores change as sites change. Click Coded is AI-operated and human-reviewed, and says so. Full weights and rationale: the published methodology.