What is an AI SEO audit?
An AI SEO audit is a checkup for one of your web pages.
It looks at two things at once.
First, the normal SEO basics that Google has always cared about. Your title, your links, how fast the page loads.
Second, something newer. Can AI tools like ChatGPT, Perplexity, Claude, and Google's AI Overviews find your page, read it, and quote it?
That second part is the big shift.
For years you typed a question into Google and clicked a blue link. Now a lot of people just ask an AI and read the answer it writes back.
That answer is built from pages the AI was allowed to crawl and could actually read. If your page is blocked or hard to read, the AI leaves you out.
An AI SEO audit tells you, in plain words, whether you made the cut.
You will sometimes see this called GEO, short for generative engine optimization. It is the same idea as SEO, just pointed at AI answers instead of only the ten blue links.
Why an AI SEO audit matters now
AI search is not some far-off idea. It is already here.
Google now shows AI Overviews right at the top of many results, above the normal links.
Tools like ChatGPT and Perplexity answer questions and list the sources they used. Being one of those sources is the new front page.
Here is the catch.
AI engines do not work like the old Google crawler. They use their own bots. They often skip JavaScript. And they lean hard on clear, well-marked content they can trust.
So a page that ranks fine on Google today can still be invisible to AI tools.
The only way to know is to test it. That is what this audit does, one URL at a time, in a few seconds.
Everything this AI SEO audit checks
Here is the full map of what we measure. Each group is explained below.
- GPTBot & OAI-SearchBot (ChatGPT)
- PerplexityBot
- ClaudeBot & Claude-User
- Google-Extended (Gemini)
- robots.txt allow / disallow rules
- JavaScript vs raw HTML render diff
- Words visible without JavaScript
- llms.txt & llms-full.txt
- Main content in the HTML source
- HTTP status & redirects
- noindex / robots meta
- Canonical tag
- XML sitemap
- hreflang for language versions
- JSON-LD schema types
- Article / BlogPosting markup
- Author, publish & update dates
- Breadcrumb & Organization data
- Title tag length & quality
- Meta description
- One clear H1
- Heading order
- Word count & depth
- Largest Contentful Paint (LCP)
- Interaction to Next Paint (INP)
- Cumulative Layout Shift (CLS)
- Server response time (TTFB)
- Internal links in the content
- External & nofollow links
- Images missing alt text
- Descriptive file context
- HTTPS & valid TLS
- HSTS, CSP, X-Frame-Options
- Open Graph tags
- Twitter / X card
What the audit measures, and why each part matters
1. AI crawler access: can AI bots even reach your page?
This is the first check, and the most important one.
If the AI cannot get in, nothing else matters.
Every AI company sends out its own bot to read the web. Each bot has its own name.
| AI crawler | Run by | What it powers | robots.txt name |
|---|---|---|---|
| GPTBot | OpenAI | ChatGPT browsing & training | GPTBot |
| OAI-SearchBot | OpenAI | ChatGPT Search results | OAI-SearchBot |
| PerplexityBot | Perplexity | Perplexity answers | PerplexityBot |
| ClaudeBot | Anthropic | Claude training data | ClaudeBot |
| Google-Extended | Gemini & AI Overviews | Google-Extended | |
| Bingbot | Microsoft | Bing & Copilot | bingbot |
A small file on your site called robots.txt decides which of these bots are allowed in.
Here is the trap. Many privacy plugins, security tools, and hosts block AI bots by default. Often they never tell you.
If PerplexityBot is blocked, Perplexity can never cite your page.
Our audit lists each major AI bot and shows a simple Allowed or Blocked badge. The fix is usually one line in your robots.txt.
User-agent: GPTBot
Disallow: /User-agent: GPTBot
Allow: /
User-agent: PerplexityBot
Allow: /You can learn the exact format in the official robots.txt guide, with each bot's rules in the OpenAI, Perplexity, and Anthropic docs.
2. Content AI can read: the JavaScript render check
Letting a bot in is only half the job. It also has to read your words.
Here is the tricky part.
A web browser runs JavaScript to build the page you see. But many AI crawlers grab the raw HTML and skip that step.
If your main text only shows up after JavaScript loads, those bots may see an almost empty page.
Our render check fetches your page two ways. First the raw HTML, then the full version. Then it compares the word counts.
A small gap is fine. A big gap is a warning sign that AI tools, and even Google's own crawler, might miss most of your page.
Content that often hides behind JavaScript:
- Text tucked inside tabs or accordions that only open on click
- Comments, reviews, and ratings pulled in after the page loads
- Products or posts hidden behind a "load more" button
- Whole pages built by a single-page app framework
The safe rule is simple. Keep your main text in the HTML, not locked behind scripts.
We also check for an llms.txt file. This is a newer idea, like robots.txt but made to point AI models to your best, cleanest content. You can read the proposal at llmstxt.org. It is not required, but it is an easy signal, so we flag it if it is missing.
3. Crawlability and indexing
This group checks the plumbing that lets any search engine find, trust, and store your page. Here is what we look at and why it matters.
- HTTP status. A healthy page returns 200 OK. We flag errors and dead ends.
- Redirects. A few are fine. Long redirect chains waste a crawler's time and slow it down.
- noindex tag. One stray noindex can quietly wipe a page off the map. We check for it.
- Canonical tag. Names the one true address for a page so duplicate versions do not compete.
- XML sitemap. The simple map that helps crawlers find every page you care about.
- hreflang. Points each visitor to the right language version of a page.
Most of these are quiet failures. Nothing looks broken on the page, but a crawler walks away. Google covers the basics in its crawling and indexing docs.
4. Structured data: helping AI understand and cite you
Structured data is hidden code, usually in a format called JSON-LD, that labels what your page is about in a language machines understand.
It says, in plain terms, "this is an article, written by this person, published on this date." People never see it. Search engines and AI engines love it.
{
"@context": "https://schema.org",
"@type": "Article",
"headline": "Best Running Shoes for 2025",
"author": { "@type": "Person", "name": "Jane Doe" },
"datePublished": "2025-03-01",
"dateModified": "2025-06-12"
}For a blog post or guide, the most valuable type is Article schema with a clear author, a publish date, and an updated date.
This is exactly the kind of signal an AI uses to decide if your page is trustworthy enough to quote.
Our audit lists every schema type it finds and warns you if Article markup is missing. You can explore all the types at schema.org and read Google's rules in its structured data guide.
Real author bios with real credentials help too. They feed into what Google calls helpful, people-first content, which both Google and AI tools reward.
5. On-page SEO: titles, descriptions, and headings
These are the classic on-page parts. They still matter a lot.
The title tag is the clickable headline in search results, and often the first line an AI reads. The meta description is the short summary below it.
Here are the quick targets we check against:
| Element | Aim for | Why it matters |
|---|---|---|
| Title tag | 50 to 60 characters | Stops it from getting cut off in results |
| Meta description | 120 to 160 characters | A full summary earns more clicks |
| H1 heading | Exactly one | Sets one clear main topic per page |
| URL slug | Short, keyword first | Easy to read, share, and trust |
We also check that you have exactly one H1 heading and that your other headings flow in a logical order.
Clean headings help an AI break your page into clear chunks it can pull a direct answer from. We also flag pages that are too thin to fully answer the question.
6. Links and internal structure
Links are how both people and bots move through your site and judge what is important. We count three kinds.
| Link type | What it is | Why it matters |
|---|---|---|
| Internal links | Point to your own pages | Spread ranking strength and help crawlers find more |
| External links | Point to other sites | Add trust when they cite solid sources |
| Nofollow links | Tagged rel="nofollow" | Tell engines not to pass credit through that link |
Knowing the mix helps you build a stronger, better-connected page.
7. Images and media
Search engines and AI tools cannot see a picture the way you do.
They rely on alt text, a short written description attached to each image.
Good alt text helps you show up in image search, makes your page work for people who use screen readers, and gives AI more context.
Our audit counts how many images are missing alt text so you can fix the gaps fast. The W3C accessibility guide is a good, plain-language reference for writing it.
8. Speed and Core Web Vitals
A slow page loses readers and rankings.
Google measures speed and stability with a set of scores called Core Web Vitals. There are three to know.
| Metric | What it measures | Good | Poor |
|---|---|---|---|
| LCP | How fast the biggest thing on screen loads | Under 2.5s | Over 4s |
| INP | How fast the page reacts to a tap or click | Under 200ms | Over 500ms |
| CLS | How much the layout jumps while it loads | Under 0.1 | Over 0.25 |
We also look at your server response time, often called TTFB. That is how long the server takes to start sending the page.
Speed helps every visitor, human or bot. A fast, stable page is also easier for AI crawlers to process at scale. Core Web Vitals come with the full report, and you can test any URL on Google's own PageSpeed Insights.
9. Security and trust signals
AI engines, like people, prefer sources they can trust.
The baseline is HTTPS, the lock icon that means your connection is encrypted. A page without it looks risky and outdated.
We also check security headers like HSTS, Content-Security-Policy, and X-Frame-Options. These protect your visitors and signal that the site is run with care. MDN's reference explains what each one does in plain terms.
10. Social and sharing tags
When your link is shared on social media or in a chat, special tags decide how the preview looks.
Open Graph tags control the title, image, and description on Facebook and LinkedIn. The Twitter card does the same on X.
A missing image or title makes your link look broken and kills clicks. These tags also give AI another clean summary of your page. You can read the spec at ogp.me.
How to read your two scores
The report gives you two main scores, one for each job your page has to do.
- AI search readiness answers one question: can AI engines crawl, read, and cite this page? It blends the crawler-access, render, and structured-data checks into a single grade.
- Technical SEO covers the classic health of the page: indexing, on-page quality, links, images, and speed. This is the score Google has always cared about.
You want both to be high.
A page can be perfect for Google and still blocked from AI tools. Or wide open to AI yet weak on the basics. Looking at both side by side shows you exactly where the gap is.
How to fix the most common problems first
Not every fix is worth the same effort.
Here is a simple order that gives you the biggest wins for the least work. The full report ranks your own issues this way automatically.
- Unblock AI bots in your robots.txt so ChatGPT, Perplexity, Claude, and Gemini can reach you at all.
- Make sure your main content shows up without JavaScript, so crawlers actually read your words.
- Add Article schema with a real author, a publish date, and an updated date.
- Fix any accidental noindex tag and confirm the page is in your sitemap.
- Tighten your title and meta description, and write a direct answer near the top of the page.
- Add alt text to images and improve your slowest Core Web Vital.
How often should you run an AI SEO audit?
Run an audit whenever you publish a new page, make big edits, change your theme or platform, or touch your robots.txt.
For a site that does not change much, a monthly check is a healthy habit.
AI crawlers and AI search features change all the time. A setting that was fine last month can quietly block you this month. Regular checks are how you stay visible.
Frequently asked questions
- What is an AI SEO audit?
- An AI SEO audit is a health check that looks at how well a web page works for both classic search engines like Google and the new AI answer engines like ChatGPT, Perplexity, Claude and Gemini. It checks if AI bots are allowed to crawl the page, if they can read the content without JavaScript, and if the page gives clear signals (like schema and author info) that help AI trust and cite it. A normal SEO audit mostly checks Google. An AI SEO audit adds the AI layer on top.
- How is an AI SEO audit different from a normal SEO audit?
- A normal SEO audit checks things like titles, links, speed and indexing for Google. An AI SEO audit checks all of that too, but it also tests whether AI crawlers are blocked in your robots.txt, whether your content shows up without JavaScript, whether you have an llms.txt file, and whether your structured data makes it easy for an AI to quote you. The goal is to be found in AI answers, not just blue links.
- Can ChatGPT, Perplexity and Gemini see my website?
- Only if you let their crawlers in. Each AI company uses its own bot: OpenAI uses GPTBot and OAI-SearchBot, Perplexity uses PerplexityBot, Anthropic uses ClaudeBot, and Google uses Google-Extended for Gemini. If your robots.txt blocks one of these, that engine cannot crawl your page and is far less likely to mention you. Our audit lists every AI bot and shows you which ones are allowed and which are blocked.
- Why does JavaScript hurt my AI SEO?
- Many AI crawlers grab the raw HTML and do not run JavaScript the way a browser does. If your text only appears after JavaScript loads, those bots may see a near-empty page. Our render check compares the raw HTML to the fully rendered page and tells you how much of your content needs JavaScript to show up. The less you depend on JavaScript for your main text, the safer you are.
- What is llms.txt and do I need it?
- llms.txt is a simple text file you place at the root of your site that points AI models to your most important, clean content. It is a newer idea, like robots.txt but for large language models. It is not required, but it is a low-effort signal that can help AI tools find and understand your best pages. Our audit checks for both llms.txt and the longer llms-full.txt.
- How do I show up in Google AI Overviews and AI answers?
- Make sure AI bots are allowed, keep your main content in plain HTML, add clear Article schema with an author and dates, write a strong title and a direct answer near the top of the page, and earn trust with HTTPS and real expertise. Our audit scores each of these and ranks the fixes by impact, so you know what to do first.
- Is this AI SEO audit really free?
- Yes. You can run a set number of audits per day with no login. The free report covers AI crawler access, the render check, indexability, structured data and on-page SEO. A free account unlocks unlimited audits plus Core Web Vitals, the full JavaScript render diff, an AI summary and a 30/60/90 day action plan.
- How often should I run an AI SEO audit?
- Run one any time you publish or heavily edit a page, change your theme or platform, or touch your robots.txt. For a stable site, a monthly check is a good habit. AI crawler rules and AI search features change often, so checking regularly keeps you from quietly going invisible to AI engines.