Type in your website address. The engine reads your site the same way Google and AI assistants do, scores each kind of visibility separately — then writes the fixes for you, ready to copy and paste. New to SEO and GEO? Start here ↓
No jargon required. Both are about the same thing — whether people find your business when they go looking for what you do. They just cover two different ways people look.
Someone needs a plumber, so they type "emergency plumber melbourne" into Google and get a page of links. SEO is everything you do so your website appears high in that list — mostly by making your site easy for Google's software to find and read.
Increasingly, people skip the list and ask an AI assistant instead — ChatGPT, or the AI summary at the top of Google. The AI replies with one answer that names only a few businesses. GEO is everything you do so the AI knows, trusts and mentions yours.
Each score is built from weighted checks. Every point you lose maps to a specific finding — with a fix attached. And where a check comes from research, the finding cites its primary source: Google's and Bing's own documentation, or the peer-reviewed GEO benchmark — where attributed quotes lifted AI-answer visibility by 41% and citing sources lifted lower-ranked sites by 115%.
Can search engines find, read and understand your pages?
Will AI assistants — ChatGPT, Perplexity, Google's AI Overviews — mention and link to you?
SEO and GEO are different funnels. A blended number hides which one is broken — every lost point maps to a visible finding.
JSON-LD, FAQPage schema, robots.txt, llms.txt, answer-block templates and a 30-day brief — generated from your own pages.
For every page, a verdict: can Google's AI features and Bing's Copilot actually use it? Judged by the engines' own published rules — with the reason whenever the answer is no.
Research-backed findings cite their primary source — Google's and Bing's documentation or peer-reviewed research — right in the report. Including checks nobody else runs, like the prompt-injection scan.
Pro asks ChatGPT, Perplexity and Claude your real customer questions — through their official APIs, never scraped — and tracks whether you're named and cited, week over week.
The Auditor answers questions about your sites by voice or text, grounded in your own reports. Set score floors and get alerted — with the fixes attached — the moment a score drops or the engines change their rules.
The free audit is the full audit — crawl, both scores, every finding and the generated fixes. Paid plans add your real data and tracking over time.
The full research-grounded audit for one site, free.
Connected data, AI-drafted fixes, and continuous tracking for 5 sites.
Run client audits at scale under your own brand.
Most visibility tools ask you to take their word for it. This engine was built the opposite way: every score must explain itself, and every research-backed claim must cite the document it came from.
"To be eligible to be shown as a supporting link in AI Overviews or AI Mode, a page must be indexed and eligible to be shown in Google Search with a snippet … There are no additional technical requirements."— Google Search Central, AI features documentation. The kind of primary source every research-backed finding in your report links to.
Research-derived findings link to Google's and Bing's primary documentation or peer-reviewed work. Checks that failed adversarial verification were never built.
Connected data is display-only — it never changes the score. Two audits of the same site are always comparable, whatever integrations are switched on.
Every pillar weight is printed in the report, so you can see exactly why a number is what it is — and disagree with it.
A diagnostic, not a ranking guarantee. llms.txt is experimental and weighted near zero. The report says so too — right where it matters.
No. Type your website address and the engine reads up to 8 pages of your public site, the same way Google does. No code, no plugins, no login.
Yes. Every finding has a "what does this mean?" plain-English explanation, and every fix is generated for you — copy and paste it, or forward the report to whoever runs your website.
No — and be wary of any tool that promises it. The engine finds and fixes the technical levers you actually control: eligibility, extractability, entity clarity and evidence.
The whole diagnostic: crawl, both scores, all findings, the AI eligibility matrix and the generated assets. Paid plans add your real Google data, AI drafting, scheduled re-audits with history, AI answer monitoring across ChatGPT/Perplexity/Claude, score-floor alerts and The Auditor assistant.
Right after you deploy fixes, then weekly or monthly. Pro re-audits on a schedule and charts both scores over time, so you can see the recovery.
Very. Ranking well on Google while being invisible to AI assistants — or the reverse — is common, and it's exactly why the report gives you two scores instead of one.
The self-serve diagnostic for classic search and AI answers.