You don’t need a crawl budget, a login, or a consultant to find out whether AI search engines can read your pages. You need 30 seconds and a browser extension. This walkthrough shows you exactly how to run a free AEO audit on any page with the SellOnLLM Chrome extension, how to read the result, and what to fix first—with specific tips for D2C, SaaS, and hospitality pages.
Why a 30-second audit beats waiting
Most teams postpone AI-search work because it sounds like a project. It isn’t—the diagnosis is instant. A single page audit tells you whether ChatGPT, Claude, Perplexity, and Google AI Overviews can actually parse and trust the page, and gives you a prioritized list of fixes. Do it on your homepage now, and you’ll know your starting point before you finish your coffee.
The 4-step walkthrough
Step 1 — Install the free extension (10 seconds)
Add the SellOnLLM extension from the Chrome Web Store. No account, no credit card. Pin it to your toolbar so it’s one click away.
Step 2 — Open the page and click the icon (5 seconds)
Navigate to any page—your product page, a landing page, or even a competitor’s—and click the SellOnLLM icon. The audit runs on the page you’re viewing, so there’s nothing to paste or configure.
Step 3 — Read your AI Readiness Score (10 seconds)
You’ll see an overall AI Readiness Score (0–100) plus two sub-scores: Technical and Content. The Technical Score tells you whether machines can access and parse the page; the Content Score tells you whether the content is clear, answerable, and trustworthy enough to reuse.
Step 4 — Work the prioritized fix list (5 seconds to read, longer to fix)
The audit ranks issues by impact. Start at the top—a blocked crawler or missing structured data matters far more than a cosmetic tag. Fix, re-audit, repeat.
What the audit checks (and why each matters)
The audit runs 21+ checks. The high-leverage ones to look for first:
- Structured data — valid Schema.org JSON-LD so models can extract facts. Google’s structured data docs are the baseline.
- Content in HTML, not just JavaScript — if your copy only appears after JS runs, most AI crawlers miss it.
- Heading hierarchy — a clear H1 and logical H2/H3s that map the page.
- llms.txt — a spec-compliant llms.txt file. Don’t have one? Use the free llms.txt generator.
- Answerable content — a direct summary near the top and an FAQ section models can lift.
For the full list and scoring logic, see the AI Readiness Score breakdown.
Vertical-specific tips
D2C & ecommerce product pages
Audit a bestseller first. The most common failures: price, materials, and shipping details trapped in images, and no Product or FAQ schema. Move specs into text and add JSON-LD. Full playbook on the D2C & ecommerce page.
SaaS & B2B pages
Audit your pricing and “alternatives” pages. Watch for vague H1s, no comparison structure, and missing author/citation signals that hurt the Content Score. See the SaaS & B2B guide.
Hospitality & hotels
Audit your rooms and location pages. The classic issue is JavaScript-only rendering that leaves crawlers an empty shell, plus missing LocalBusiness schema. See hospitality.
From free audit to real gains
The free audit is the on-ramp. Once you know what’s broken, you have three paths:
- Fix it yourself using the prioritized list—most Technical Score wins are quick.
- Generate AI-ready content in the extension to rewrite weak meta and copy.
- Track citations with an AI Visibility check to see if the fixes get you recommended by ChatGPT, Claude, and Perplexity.
For crawling many URLs at once instead of one page, pair the extension with our free multi-page LLM audit and E-E-A-T scanner.
Audit a competitor to set your bar
Because the audit runs on any URL, run it on the competitor who keeps getting cited when you don’t. Their score and structure show you the standard to beat—often it’s simply cleaner schema and more answerable content, not a bigger budget.
Do it now
Install the extension, open your most important page, and click once. Thirty seconds from now you’ll know exactly where you stand with AI search—and exactly what to fix first.