Key takeaways
- An AI-ready website is a normal, well-built website: crawlable, fast, clear and specific. Google says there are no extra technical requirements for its AI features.
- The most common failure is access: robots.txt rules, Cloudflare bot settings or JavaScript-only content that keep crawlers from reading the page.
- AI assistants quote facts. Service pages, your prices, address, hours and FAQs need to be in plain text, not in images or PDFs.
- Allow search and answer crawlers (Googlebot, Bingbot, OAI-SearchBot, PerplexityBot, Claude-SearchBot); deciding on training crawlers is a separate choice.
- Measure AI referrals in GA4's AI Assistant channel and AI Overviews and AI Mode in Search Console.
An AI-ready website is one that AI assistants can crawl, read and quote: server-rendered pages, fast loading, one clear page per service, and your facts (your prices, address, hours, areas) written in plain text. Google says there are “no additional technical requirements” to appear in AI Overviews or AI Mode beyond being indexed and eligible for a snippet. This AI-ready website checklist covers the 15 things that most often decide whether ChatGPT, Gemini, Perplexity and Google’s AI features can use your site.
Why AI search needs a website at all
AI assistants answer questions by searching indexed web pages and citing them. Most social platforms block AI crawlers or put content behind a login. We checked this in detail in Can ChatGPT and Google AI read your Facebook page? Your website is the source you control, so it has to be easy to read. The full argument is in why your website matters more than social media in AI search.
The AI-ready website checklist: 15 things to check
| # | Check | Done when |
|---|---|---|
| 1 | Crawlable, server-rendered HTML | Main text, prices and contact details are in the page source, not loaded only by JavaScript |
| 2 | Speed and Core Web Vitals | Pages load fast on mobile; Search Console’s Core Web Vitals report shows no “poor” URLs |
| 3 | Clear service pages | One page per main service: what it is, who it’s for, process, timing, price or range |
| 4 | Business facts in text | Address, opening hours, service areas and your prices written out on the page |
| 5 | Structured data | Organization or LocalBusiness schema that matches the visible facts; FAQ markup only for FAQs visible on the page |
| 6 | Thai and English with hreflang | Native versions of key pages, linked to each other with hreflang tags |
| 7 | Named authors and an About page | Real people, roles, company history and location |
| 8 | Updated dates | Time-sensitive pages show a visible “updated” date that changes only when content changes |
| 9 | Google Business Profile consistency | Name, address, phone, hours and website identical on your site and your profile |
| 10 | robots.txt and Cloudflare AI settings | Search and answer crawlers allowed; training crawlers set by choice |
| 11 | XML sitemap | Sitemap lists all indexable pages and is submitted in Search Console and Bing Webmaster Tools |
| 12 | No key content only in images or PDFs | Menus, price lists and brochures also exist as HTML text |
| 13 | Internal links | Every service page is linked from the menu or related pages with descriptive anchor text |
| 14 | LINE OA link | “Chat on LINE” button on every page, plus the LINE ID in text on the Contact page |
| 15 | AI referrals measured in GA4 | AI Assistant channel visible in Traffic acquisition; baseline recorded |
The sections below explain the ones that need more than a line.
Access: can crawlers read your pages? (1, 10, 11)
Many AI crawlers don’t run JavaScript the way a browser does. Google can process JavaScript “as long as it isn’t blocked”, but says JavaScript-heavy sites are harder to optimise. The safe route is to have every important word in the HTML the server sends.
Then check who is allowed in. Your robots.txt should allow:
| Allow (search and answers) | Your choice (training) |
|---|---|
| Googlebot, Bingbot | GPTBot (OpenAI) |
| OAI-SearchBot, ChatGPT-User (OpenAI) | ClaudeBot (Anthropic) |
| PerplexityBot, Perplexity-User | Google-Extended (Gemini training and grounding) |
| Claude-SearchBot, Claude-User (Anthropic) | CCBot (Common Crawl) |
OpenAI says sites that block OAI-SearchBot won’t appear in ChatGPT search answers. Blocking a training crawler like GPTBot is a separate decision and doesn’t remove you from ChatGPT search.
If your site runs on Cloudflare, also check its AI crawler settings. Cloudflare’s AI Crawl Control lets you allow or block AI crawlers, and a block there overrides an “allow” in robots.txt. Make sure search crawlers aren’t blocked by accident.
Content: give AI facts it can quote (3, 4, 12)
AI assistants quote specific facts, not slogans. Write each service page so its first two sentences answer “what is it and who is it for?”, then cover process, timing and your prices or ranges.
Don’t leave key information only in a picture of a menu, a price list in a PDF or text inside a video. Add it as HTML text too. Google calls the content that stands out “non-commodity”: your own experience, local detail and data, not what every other site says.
Trust and identity (5, 6, 7, 8, 9)
Structured data isn’t required for Google’s AI features, Google says, but it helps machines understand who you are. Keep it consistent with the visible page. Add Organization or LocalBusiness schema, with sameAs links to your official profiles.
If you serve Thai and international customers, publish native Thai and English pages and connect them with hreflang. People ask AI assistants in their own language.
Named authors, an About page and visible updated dates help people and AI judge whether to trust you. Your Google Business Profile should say exactly what your site says. Google lists an up-to-date Business Profile among its recommendations for AI features.
Connect and measure (13, 14, 15)
Internal links help crawlers find every page and understand how they relate. A visible LINE OA button turns an AI-referred visitor into a conversation; see our LINE OA marketing service.
For measurement, GA4’s default channel group includes an AI Assistant channel for visits from sources like ChatGPT, Gemini and Copilot. It excludes Google’s AI Overviews and AI Mode, which count as Organic Search. For those, use Search Console. Our AI SEO guide covers the reports in detail, and our Google AI Mode guide explains how AI Mode picks sources.
Who builds websites this way
We build websites to this checklist: fast, server-rendered pages, Thai and English, clear service pages and the technical setup done — a standard business site is usually live in about a week. Afterwards you order updates and see your reporting in FoxWorks, our customer portal. Web apps and LINE integrations are built by our development team, FoxSoft Solutions.
Already have a site and want it cited more often? Our AI search & SEO service starts with this checklist.
FAQ
Do I need special code or an llms.txt file to be AI-ready?
No. Google says you don't need AI text files, special markup or Markdown to appear in its AI features, and that Search doesn't use llms.txt. A well-built, crawlable site is what counts.
Should I block GPTBot and other training crawlers?
It's a business decision. Blocking GPTBot doesn't remove you from ChatGPT search, which uses OAI-SearchBot. Most small businesses that want to be known by AI leave training crawlers allowed.
Is FAQ schema still worth adding?
Only if the questions and answers are visible on the page. Google says structured data must match the visible text. The visible Q&A is what AI assistants quote.
How often should I review this checklist?
Every three to six months, and after any redesign, hosting move or change to Cloudflare settings. Crawler rules and platform settings change often.
Sources
- Google Search Central — AI features and your website
- Google Search Central — Optimizing your website for generative AI features on Google Search
- OpenAI — Overview of OpenAI crawlers
- Perplexity — Perplexity crawlers
- Anthropic Help Center — Anthropic crawlers (ClaudeBot, Claude-User, Claude-SearchBot)
- Cloudflare Docs — AI Crawl Control
- Google Analytics Help — Default channel group