No Signup Unlimited Usage Real Data 100% Free Privacy First Fast Results Export Reports Mobile Friendly

JavaScript Rendering Checker

See what AI crawlers that do not run JavaScript get from a page: raw HTML vs a headless-Chromium render, field by field (title, canonical, robots, headings, words, links, JSON-LD, hreflang), JS dependency %, render-only text highlighted and SSR/prerender fixes.

Fetching live data…

Quick answer

The JavaScript Rendering Checker shows what AI crawlers see on a page compared with what a browser sees. It fetches the raw HTML without running JavaScript, renders the same page in headless Chromium, and compares the two field by field. You learn how much of your content, links and metadata exist only after JavaScript runs, and what to move into the server HTML.

How it works

  1. Enter a page URL.
  2. The tool downloads the raw HTML with a GPTBot-like User-Agent and does not execute any JavaScript, which approximates what most AI crawlers receive.
  3. It then renders the page in headless Chromium with images, fonts and media blocked, waiting for the network to go idle or 15 seconds at most. Every sub-request is safety-checked.
  4. Both versions go through the same parser, and you get a score, a comparison table, a text diff and fix suggestions.

What it checks

  • Word count — visible words in the raw and rendered versions, and the JS dependency: rendered words minus raw words, divided by rendered words, as a percentage.
  • Title and meta description — whether they exist in the raw HTML and whether JavaScript changes them.
  • Canonical and meta robots — flagged as critical if they change in the render or appear only after rendering.
  • Headings — H1 to H3 in each version.
  • Links — internal and external link counts; a critical finding if more than half of the rendered links are missing from the raw HTML.
  • Structured data — JSON-LD types in each version, with a critical finding when JSON-LD exists only in the render.
  • Hreflang — language alternates in the raw and rendered HTML.
  • Thin raw content — a critical finding when the raw HTML has under 100 words but the render has over 300.

Why it matters

Googlebot renders pages, so content added by JavaScript can still reach Google Search, as Google's JavaScript SEO basics explains. GPTBot, ClaudeBot and PerplexityBot do not execute JavaScript. A single-page app that ships an empty shell and builds everything in the browser can look complete to a visitor while an AI crawler receives a title and almost nothing else. The same goes for links added by click handlers and for JSON-LD injected by a tag manager.

The score starts at 100. Each critical finding subtracts 25 points, each warning 10, and every 10% of JS dependency another 5, down to a minimum of 0. This is our own measure of how much a page relies on JavaScript, not a rule from Google or any AI company. A low score suggests that crawlers which do not render may miss parts of the page; it cannot show how a particular service uses what it fetched.

How to fix common problems

  • Read the diff first — the side-by-side text view highlights text that appears only after rendering, so you can see exactly what non-rendering crawlers miss.
  • Render on the server — server-side rendering (SSR), static generation (SSG) or prerendering puts the main content into the HTML response. Most modern frameworks support at least one of these.
  • Use real links — navigation should use <a href> elements with real URLs, not buttons or click handlers that change the page in JavaScript.
  • Put critical tags in the server HTML — the title, canonical, meta robots and JSON-LD should arrive in the first response, not be added or changed later. Confirm the result with the Meta Tag Analyzer and the Structured Data (Schema) Checker.
  • Check access too — content in the raw HTML only helps if the bots can fetch it. Test that with the AI Crawler Access Checker and list your key pages in an llms.txt checked by the llms.txt Validator.

Frequently asked questions

Which crawlers do not run JavaScript?

GPTBot, ClaudeBot and PerplexityBot fetch the HTML without executing JavaScript, and many other AI crawlers work the same way. Googlebot renders pages. This tool's raw fetch approximates the non-rendering crawlers, and its Chromium render approximates a browser or a rendering crawler.

What does the JS dependency percentage mean?

It is the share of rendered words that are missing from the raw HTML: rendered words minus raw words, divided by rendered words, times 100. A value near zero means the text is already in the server response; a high value means most of it appears only after JavaScript runs.

Why are images and fonts blocked during rendering?

They do not change the text, links or tags the tool compares, and skipping them makes rendering faster. Scripts and data requests still run, so content built by JavaScript appears in the render as it would in a browser, within the 15-second limit.

Why might results not update right after I change my page?

Results are cached for 24 hours per URL, and renders run in a small queue so each one gets enough resources. If you have just deployed a fix, the cached report may still show the earlier version until the cache expires.

Is a low score bad for Google rankings?

Not by itself. Google can render JavaScript, and the score is our own criterion, not a Google ranking factor. A low score is mainly a warning about crawlers that do not render, and a reason to confirm that key content, links and tags are in the server HTML.

Is the JavaScript Rendering Checker free?

Yes, it is free to use. Enter a URL and the report shows the raw and rendered versions side by side, with the field-by-field table, the text diff and fix suggestions. For speed and layout metrics, use the Page Speed & Size Test.

Learn & Resources

Guides, tips and tutorials hand-picked for this tool — go from running a check to actually fixing it.

Related tools

Website Technology & CMS Detector

Find out what a website is built with: CMS, e-commerce platform, JavaScript framework, CDN, analytics, tag managers, server and programming language, detected from the HTML, HTTP headers and cookies with version numbers and a confidence level for each match.

Open tool →

Topical Authority Map

Map how completely your site covers a topic — the way Google and AI engines judge topical authority. Crawls your internal links, titles and headings, compares them against the real subtopics people search for, and shows your covered subtopics plus the content gaps to build for a stronger topic cluster.

Open tool →

AI Crawler Access Checker

Check whether GPTBot, OAI-SearchBot, ChatGPT-User, ClaudeBot, PerplexityBot and 12 other AI bots can reach a page: the robots.txt rule (with line number) and a live request with each bot’s real User-Agent, WAF challenge detection and copy-ready robots.txt templates.

Open tool →

Entity Extractor

Extract the named entities (people, organizations, places, concepts) a page contains — the way AI and Google’s Knowledge Graph read it. Lists every entity with mention counts and a type guess, scores entity richness and diversity, and flags where schema markup would define your entities more clearly. Paste text or enter a URL.

Open tool →

AI Prompt-Friendly Content Checker

Check how well your content answers the prompts real people type into AI chatbots. Scores prompt-alignment — direct answers, self-contained context, definitions, steps and comparisons — so ChatGPT and Perplexity can lift a complete answer from your page.

Open tool →

NLP Content Analyzer

Analyze your content the way a natural-language processing engine does: entities, salient terms, sentiment, readability, sentence structure and factual signals. Scores how parseable and unambiguous your text is for search and AI models, with a top-terms chart and linguistic metrics. Paste text or enter a URL.

Open tool →