- Google processes JavaScript websites in three phases. Crawling comes first, then rendering, then indexing. Google indexes the rendered HTML.
- Before rendering, there is a queue. According to Google, a page may stay there for only a few seconds, but it can take longer than that.
- Static generation and server-side rendering are best for fast indexing, because the content is already in the server's first response. Google calls dynamic rendering a workaround.
- Most AI crawlers don't execute JavaScript. That is why the main content, links, title, canonical and robots directives belong in the raw HTML.
JavaScript SEO means building a website so that search engines can also find, render and index the content that only JavaScript creates. This guide describes how the process works at Google, compares the rendering strategies and collects the best practices. The statements about Google come from the Google Search Central documentation. We checked them on November 16, 2026.
Two terms run through the text. The raw HTML is the server's response, meaning the source code before any script runs. The rendered HTML is the state after a browser has executed the JavaScript.
How does Google crawl, render and index JavaScript?
According to Google's JavaScript SEO basics, Google processes JavaScript web apps in three main phases. Crawling comes first, then rendering, then indexing. Googlebot queues pages for both crawling and rendering, so there is one queue for each.
Crawling
Googlebot fetches a URL from the crawl queue and first reads robots.txt. If the URL is disallowed there, Googlebot doesn't fetch it at all. From the HTML response, it reads the URLs in the href attributes of links and adds them to the crawl queue. According to Google's In-depth guide to how Google Search works, Google also discovers new pages through submitted sitemaps.
For classical websites and server-side rendered pages, this response already contains all the content. Some JavaScript sites use the app shell model, where the raw HTML doesn't contain the actual content. Google then needs to execute JavaScript to see it.
Rendering and the render queue
Googlebot puts every page with a 200 HTTP status code into the render queue, which Google calls the rendering queue. This also applies to pages without JavaScript. Google names one exception. If a robots meta tag or the HTTP header contains noindex, Google may skip rendering. Pages with a different status code, such as error pages with 404, may not be rendered either.
Google doesn't put a number on how long a page waits. The documentation says the page may stay on this queue for a few seconds, but it can take longer than that. According to Google, it is not immediately obvious whether a page is waiting for crawling or for rendering.
Once Google's resources allow, a headless Chromium renders the page and executes the JavaScript. According to Google, it uses an evergreen version of Chromium, meaning an up-to-date one. Googlebot passes the fetched HTML on to the Web Rendering Service, or WRS for short. The WRS downloads the scripts and stylesheets the HTML references and constructs the page as a user's browser would. That is how the post Crawling December on the Google Search Central Blog describes it.
The renderer behaves differently from a user's browser in three ways. It doesn't click or scroll, as Google's guidance on lazy loading states. According to the guide Fix Search-related JavaScript problems, it clears cookies, local storage and session storage across page loads. The WRS caches scripts and stylesheets for up to 30 days, regardless of your caching headers. What follows from this is covered under Best practices.
Indexing
After rendering, Googlebot parses the rendered HTML for links again and queues the URLs it finds for crawling. Google uses the rendered HTML to index the page. That is why Google discovers a link that only JavaScript inserts one step later than a link in the raw HTML.
Rendering alone doesn't put a page in the index. In its In-depth guide to how Google Search works, Google itself writes that indexing isn't guaranteed and that not every page Google processes will be indexed.
What should be in the raw HTML?
Google reads a lot from the rendered HTML too. But whatever is already in the raw HTML is available without rendering, for Google on the first fetch and for every bot that doesn't execute JavaScript. The table shows the difference for six parts of a page. The information comes from Google's JavaScript SEO basics, and for structured data from the guide Generate structured data with JavaScript.
| Part of the page | In the raw HTML | Only created by JavaScript |
|---|---|---|
| Main content | Every crawler reads it on the first fetch. | Google sees it after rendering. Bots without JavaScript don't see it at all. |
| Internal links | Google adds them to the crawl queue while crawling. | Google discovers them after rendering if they are in the DOM as a elements with an href. |
| Title and meta description | Bots without rendering read them too. | Allowed, according to Google. |
| Canonical | Recommended by Google. | Allowed if the value is the same as in the raw HTML and the page has only one canonical. |
| noindex | Google may skip rendering. A script that removes noindex again then doesn't work as expected. | Possible, according to Google, for example on error pages of a single-page app. |
| Structured data | Bots without rendering read it too. | Google processes JSON-LD that is in the DOM when the page renders. According to Google, this can make Shopping crawls less frequent and less reliable for Product markup. |
The status code always comes from the server, before any script runs. Google may not render pages with a code other than 200 at all.
How do Bing and other search engines handle JavaScript?
Bing renders JavaScript too. In October 2019, Microsoft announced that Bingbot renders pages with the most recent stable version of Microsoft Edge, which, like Chrome, is based on Chromium. A year earlier, the Bing team wrote that Bingbot is generally able to render JavaScript, but that it is difficult to process JavaScript at scale on every page of every website. At the time, Bing recommended dynamic rendering for websites relying heavily on JavaScript. Both statements come from the Bing Webmaster Blog, in the posts The new evergreen Bingbot and JavaScript, Dynamic Rendering, and Cloaking.
Google is cautious about all other search engines. Its page on dynamic rendering says that other search engines may choose to ignore JavaScript and won't see JavaScript-generated content. Whatever is in the raw HTML, on the other hand, reaches every search engine.
Rendering strategies compared
The rendering strategy determines where a page's HTML is created. That can be on the server, at build time or only in the browser. This decides what a search engine sees on the first fetch. The terms follow the overview Rendering on the web on web.dev, Google's site for developers.
| Strategy | What the search engine sees on the first fetch | Effect on indexing | What it suits |
|---|---|---|---|
| Client-side rendering (CSR) | An HTML skeleton with scripts. The content is only created in the browser. | Google sees the content after rendering. Bots without JavaScript don't see it at all. | Areas behind a login and applications that shouldn't be indexed. |
| Server-side rendering (SSR) | The complete HTML, generated on the server for every request. | Content, links and meta tags are in the raw HTML and don't need rendering. | Pages whose content changes often or depends on the user, such as categories and products in a shop. |
| Static generation (SSG) | The complete HTML, generated at build time and served as a finished file. | Same as server-side rendering, plus a consistently fast server response. | Content that rarely changes, such as guides, documentation and landing pages. |
| SSR or SSG with hydration | The complete HTML. Scripts then make the page interactive in the browser. | Same as SSR and SSG, as long as the scripts don't change the content afterward. | Websites on frameworks such as Next.js or Nuxt that combine content and interaction. |
| Static with background regeneration (ISR) | The complete HTML from a static file that the server regenerates in the background. | Same as static generation. | Websites with a very large number of pages whose content changes regularly. |
| Dynamic rendering (prerendering for bots) | Detected bots get a prerendered version without JavaScript, users get the client-side rendered one. | Detected bots get the content without rendering. Google calls the approach a workaround and doesn't recommend it. | A transition for existing client-side apps until SSR, SSG or hydration is in place. |
Hydration combines server-side and client-side rendering. The server delivers finished HTML, and scripts then make the page interactive in the browser. According to web.dev, frameworks such as Next.js and Nuxt offer both static rendering and server-side rendering. At Next.js, regenerating static pages in the background is called Incremental Static Regeneration, or ISR for short.
Google calls dynamic rendering a workaround
With dynamic rendering, the server detects crawlers by their user agent and sends them a prerendered version as static HTML. Users still get the client-side rendered page. In practice, this is often called prerendering for bots. If the server delivers prerendered HTML to all visitors, that is static rendering. Google recommends that. Google is clear about the approach on its page Dynamic rendering as a workaround. According to Google, dynamic rendering was a workaround and not a long-term solution, because it creates additional complexities and resource requirements. Instead, Google recommends server-side rendering, static rendering or hydration.
Google generally doesn't consider dynamic rendering as cloaking, as long as bots and users get similar content. Serving completely different content to users and crawlers can be considered cloaking.
Which rendering strategy is best for fast indexing?
Static generation and server-side rendering are best. With both, the complete content is already in the server's first response. Google can read content, links and meta tags while crawling and doesn't have to wait for rendering. There are three reasons for this.
- The content doesn't wait for rendering. Google only sees client-side rendered content once the page has gone through the render queue. Google doesn't say how long that takes.
- Links are discovered earlier. Links in the raw HTML go into the crawl queue while crawling, links inserted by JavaScript only after rendering.
- The content reaches every bot. Google calls server-side rendering and prerendering a great idea, because not all bots can run JavaScript.
Which of the two fits depends on the content. Static generation suits pages that are the same for all visitors and rarely change. Server-side rendering suits content that changes often or is personalized. With a very large number of URLs, the static build becomes laborious because an HTML file has to be created in advance for every URL. Hybrid approaches such as ISR exist for this.
Client-side rendering makes indexing depend on rendering. Google can process the content, but it needs the extra step, and bots without JavaScript get nothing. No strategy comes with a guarantee. In its In-depth guide to how Google Search works, Google writes that it doesn't guarantee crawling or indexing, even if a page follows the Google Search Essentials.
Best practices for JavaScript SEO
The following 13 points come from Google's documentation. The source is given at the end of each point.
- Deliver the main content in the server's HTML. According to Google, server-side rendering or prerendering is still a great idea because it makes your website faster for users and crawlers, and not all bots can run JavaScript. Google Search Central, Understand JavaScript SEO basics
- Use a elements with an href attribute for links. Generally, Google can only crawl links of this kind. Google doesn't reliably extract links built with
onclick, aspanor ajavascript:call in thehref. Links that JavaScript inserts are fine as long as they use this markup. Google Search Central, Link best practices for Google - Give every view its own URL. Use the History API in single-page apps. Googlebot can't reliably resolve fragments such as
#/products. Google Search Central, Understand JavaScript SEO basics - Give every page its own title and meta description. Google allows you to set both with JavaScript. If they are in the raw HTML, a bot without rendering reads them too. Google Search Central, Understand JavaScript SEO basics
- Set the canonical in the HTML. If you have to use JavaScript, set the same value as in the raw HTML and only one canonical per page. According to Google, conflicting or multiple canonical tags may lead to unexpected results. Google Search Central, Understand JavaScript SEO basics
- Only put noindex in the raw HTML if the page shouldn't be indexed. When Google encounters
noindexthere, it may skip rendering. A script that removes the tag then may not work as expected. Google Search Central, Understand JavaScript SEO basics - Use meaningful status codes. Single-page apps often report a 200 status code for error pages. Google names two ways to avoid these soft 404 errors. You use a JavaScript redirect to a URL for which the server responds with 404, or you add a
noindexto the error page with JavaScript. Google Search Central, Understand JavaScript SEO basics - Allow scripts and CSS in robots.txt. Google doesn't render JavaScript from blocked files. If the renderer can't fetch a rendering-critical resource, Google may have trouble extracting the page's content. Google Search Central, Understand JavaScript SEO basics · Google Search Central Blog, Crawling December
- Load content without clicks and without scrolling. Google Search doesn't interact with your page. Lazy loading should load content when it is visible in the viewport. Infinite scroll also needs its own URL for every chunk. Google Search Central, Fix lazy-loaded content
- Test structured data that JavaScript creates. Google processes JSON-LD that is in the DOM when the page renders. The Rich Results Test shows whether it gets through. Google Search Central, Generate structured data with JavaScript
- Fingerprint your script file names. Google's renderer caches scripts and stylesheets for up to 30 days and may ignore caching headers. If the file name changes with the content, such as
main.2bb85551.js, Google loads the new version. Google Search Central, Understand JavaScript SEO basics · Google Search Central Blog, Crawling December - Keep HTML and scripts lean. For Google Search, Googlebot crawls the first 2MB of a file, measured on the uncompressed data. The limit applies to each script file separately. For users, Google recommends good Core Web Vitals, meaning LCP within 2.5 seconds, INP under 200 milliseconds and CLS under 0.1. Google Search Central, Googlebot · Google Search Central, Understanding Core Web Vitals
- Check the rendered HTML. The URL Inspection tool in Search Console and the Rich Results Test show how Google crawls and renders a URL. You see the loaded resources, the JavaScript console output and the rendered DOM. Google Search Central, Fix Search-related JavaScript problems
The same applies to web components. When Google renders a page, it flattens the shadow DOM and light DOM content and indexes only what is visible in the rendered HTML.
What do AI crawlers do differently?
Most AI crawlers don't render. They read the raw HTML and don't execute JavaScript. Our overview lists 39 bots and tokens. For 14 of them there is a statement from the operator or a measurement, and 8 of these 14 don't execute JavaScript, including GPTBot and ClaudeBot. The full list with the sources for every bot is in the guide AI crawlers and JavaScript rendering.
The same applies to these bots as to any search engine without rendering. Content, links and structured data that only appear in the browser don't reach them. Anything that should show up in AI answers belongs in the raw HTML.
How do you check what search engines see on your page?
The check is always the same comparison. You put the raw HTML next to the rendered HTML and look at what JavaScript adds, removes or changes.
Checking a single page
- The free JavaScript Rendering Check loads the page once and shows both values for title, meta description, canonical, meta robots, hreflang, structured data, headings, text, images and links.
- The URL Inspection tool in Search Console shows what Google itself fetched. Under “View crawled page”, you see the HTML, the HTTP headers, the JavaScript console output and the loaded resources of the last crawl. “Test live URL” also gives you a screenshot of the rendered page. The buttons are described in the Search Console Help on the URL Inspection tool.
- The Rich Results Test also renders a URL. It also works for websites that you don't own in Search Console.
Checking many pages
Spot checks show individual URLs. But differences are often tied to the template and only occur on certain page types. searchVIU's JavaScript crawling study from 2020 found differences only on subpages for 81 of over 200 domains. For the whole website, you therefore need a crawler that renders every URL in a browser, stores both versions and compares them element by element. How to set up such a crawl is explained in the guide Crawling a website for technical SEO analysis.
Checking all pages with searchVIU
searchVIU shows you what JavaScript changes on every URL and renders every single URL in Chrome, even on websites with millions of URLs. The “JS Rendering” report names the changed signal, such as title, meta description, canonical or robots directives, with the value in the raw HTML and the value after rendering. If JavaScript removes content or internal links, the report shows that too. The full crawl runs every week, and in between, Page Watch checks your most important pages continuously, throughout the day. More on this on the JavaScript SEO page.
What do searchVIU's tests show about JavaScript SEO?
We have been testing how Google handles JavaScript since 2017. The four tests date from 2017 to 2020. The method, figures and current state are in each guide.
- Experiment on hreflang in the rendered HTML from 2017. Google used the rendered HTML for hreflang and ignored the tags in the source code.
- JavaScript SEO experiments with Google Tag Manager from 2017. Google picked up five changes that only appeared during rendering. On the homepage it took 8 days, on the other pages around two months.
- Canonical via JavaScript from 2018. Google canonicalized all four test pages, after 10, 23, 24 and 34 days.
- JavaScript crawling study from 2020. For 96% of over 200 domains and 56% of URLs, the raw and the rendered HTML differed.
The times in these tests cover the entire process until the change showed up in search. Three of the four tests ran before Google switched Googlebot to a continuously updated version of Chromium in May 2019.
FAQ
How do search engines crawl and index JavaScript websites?
Google works in three phases. Googlebot fetches the URL and reads the links from the raw HTML. Then the page goes into the render queue, and a current version of Chromium executes the JavaScript. Google indexes the rendered HTML.
According to Microsoft, Bing renders with Microsoft Edge. According to Google, other search engines may ignore JavaScript.
How does JavaScript rendering affect a website's SEO?
Rendering decides when and for whom your content is visible. If the content is in the raw HTML, every crawler reads it on the first fetch. If it only appears in the browser, Google sees it after rendering, and bots without JavaScript don't see it at all.
The same applies to the title, canonical, robots directives, internal links and structured data.
How do different JavaScript rendering methods affect SEO?
Server-side rendering and static generation deliver the content in the raw HTML, with or without hydration. Client-side rendering makes indexing depend on rendering. Dynamic rendering serves detected bots a prerendered version, and Google considers it a workaround.
Which rendering strategy is best for fast indexing?
Static generation and server-side rendering are best. With both, the complete content is in the server's first response, and Google doesn't have to wait for rendering to see content and links.
Static generation suits content that rarely changes. Server-side rendering suits content that changes often or is personalized.
What are the best practices for JavaScript SEO?
Deliver the main content in the server's HTML, use a elements with href for links and give every view its own URL. Title, canonical and robots directives belong in the raw HTML, error pages need a real status code, and content should load without clicks.
Then check the rendered HTML with the URL Inspection tool in Search Console. All 13 points with their sources are in the Best practices section.
How long does it take Google to render a JavaScript page?
Google doesn't give a fixed time. According to the documentation, a page may stay in the render queue for only a few seconds, but it can take longer than that.
In searchVIU's tests from 2017 and 2018, it took between 8 days and three months for changes from the rendered HTML to show up in search. These tests are older than today's Googlebot and cover the entire process.
Sources
Retrieved on November 16, 2026. If something is out of date or missing, let us know.
- Google Search Central, Understand JavaScript SEO basics: https://developers.google.com/search/docs/crawling-indexing/javascript/javascript-seo-basics · As of Mar 4, 2026
- Google Search Central, Dynamic rendering as a workaround: https://developers.google.com/search/docs/crawling-indexing/javascript/dynamic-rendering · As of Dec 10, 2025
- Google Search Central, Fix lazy-loaded content: https://developers.google.com/search/docs/crawling-indexing/javascript/lazy-loading · As of Dec 10, 2025
- Google Search Central, Fix Search-related JavaScript problems: https://developers.google.com/search/docs/crawling-indexing/javascript/fix-search-javascript · As of Dec 18, 2025
- Google Search Central, Link best practices for Google: https://developers.google.com/search/docs/crawling-indexing/links-crawlable · As of Dec 10, 2025
- Google Search Central, Generate structured data with JavaScript: https://developers.google.com/search/docs/appearance/structured-data/generate-structured-data-with-javascript · As of Dec 10, 2025
- Google Search Central, In-depth guide to how Google Search works: https://developers.google.com/search/docs/fundamentals/how-search-works · As of Dec 18, 2025
- Google Search Central Blog, Crawling December: The how and why of Googlebot crawling: https://developers.google.com/search/blog/2024/12/crawling-december-resources · Dec 3, 2024
- web.dev, Rendering on the web: https://web.dev/articles/rendering-on-the-web · As of Jan 5, 2026
- Microsoft Bing, The new evergreen Bingbot: https://blogs.bing.com/webmaster/2019/10/The-new-evergreen-Bingbot-simplifying-SEO-by-leveraging-Microsoft-Edge/ · Oct 9, 2019