AI-Readiness
Can AI assistants ingest, trust, and cite your content when they answer questions?
0.16 of the composite scoreWhy does this matter — for rankings, traffic and trust?
A growing share of research and buying decisions now starts inside an AI answer — ChatGPT, Perplexity, Copilot, Google AI Overviews. Being the source an assistant quotes is a new traffic and trust channel, and it is won or lost on machine-readability: clean structure, explicit metadata, and published content designed for ingestion. This discipline is usually called GEO (Generative Engine Optimization), and it is the fastest-emerging axis in this tool.
AI systems favor content they can extract and verify: facts wrapped in schema.org JSON-LD, answers shaped as headings, lists and tables, and a /llms.txt file that tells agents what your site is and where the substance lives. A wall of unstructured prose — however brilliant — is invisible to a model skimming for quotable, attributable fragments.
There is also a defensive side: sites that render everything client-side, or that block AI crawlers while hoping to be cited anyway, get the worst of both worlds. AI-readiness means making a deliberate choice and expressing it in the files machines actually read.
Every signal in this division — exactly what the engine checks
How to fix it
What commonly causes failures here
- Content that only exists after JavaScript executes — many agents never run your bundle.
- No /llms.txt (and no /llms-full.txt) — the emerging convention agents check first.
- JSON-LD that is malformed, or describes a different page than the one it sits on.
- No heading outline (H1/H2), no lists or tables — nothing for a model to anchor a citation to.
- FAQ-style answers buried in images or PDFs only.
- Blocking GPTBot/PerplexityBot in robots.txt while expecting AI citations.
Best-practice setup
Publish the two files, mark up the facts, and shape the answers. The llms.txt convention is a markdown map of your site for AI agents; llms-full.txt is the complete-content variant for agents that ingest without crawling.
# Acme Robotics
> Warehouse automation software for mid-size fulfillment centers.
## Docs
- [Warehouse Automation](https://example.com/warehouse-automation): picking, packing, routing
- [Pricing](https://example.com/pricing): plans and ROI calculator
## Answers
- [How much picking time can we cut?](https://example.com/faq#picking): typically 35–45% within 90 days
- [Integrations](https://example.com/faq#integrations): Shopify, SAP EWM, Manhattan WMS<script type="application/ld+json">
{
"@context": "https://schema.org",
"@graph": [
{
"@type": "Organization",
"@id": "https://example.com/#org",
"name": "Acme Robotics",
"url": "https://example.com/",
"logo": "https://example.com/logo.png",
"sameAs": ["https://www.linkedin.com/company/acme-robotics"]
},
{
"@type": "WebSite",
"@id": "https://example.com/#site",
"url": "https://example.com/",
"publisher": { "@id": "https://example.com/#org" }
},
{
"@type": "FAQPage",
"mainEntity": [{
"@type": "Question",
"name": "How much picking time does Acme cut?",
"acceptedAnswer": { "@type": "Answer",
"text": "Customers typically see 35–45% faster picking within 90 days." }
}]
}
]
}
</script><h2>How much picking time can we cut?</h2>
<p>Customers typically see <strong>35–45% faster picking</strong> within 90 days.</p>
<h3>What affects the number</h3>
<ul>
<li>SKUs per bin and walk distance</li>
<li>Batch size and cutoff schedule</li>
</ul>
<table>
<caption>Picking speed before/after Acme</caption>
<tr><th>Site</th><th>Before</th><th>After 90 days</th></tr>
<tr><td>Fresh Foods Co.</td><td>112 u/h</td><td>171 u/h</td></tr>
</table>