What is rendering? Definition, types and SEO impact

What is rendering?

Rendering is the step that executes a page’s code and builds the DOM, the structure from which text, links and images are read. A plain HTML page needs little rendering because the content is already in the source. A JavaScript application often sends an almost empty HTML shell, and the content exists only after scripts run. Search engines face the same choice as browsers: they either read the raw HTML or spend resources to render the page fully.

How does Google render pages?

Google renders pages in a separate stage between crawling and indexing, using an evergreen version of headless Chromium. Google Search Central names three phases (crawling, rendering, indexing), which run in four steps:

1. Crawl the URL with Googlebot, check robots.txt and read links from the raw HTML.
2. Queue every page with a 200 status for rendering, whether it uses JavaScript or not.
3. Render the page in headless Chromium when resources allow, then read new links and content from the rendered HTML.
4. Index the rendered version of the page.

The render queue is the reason JavaScript content can appear in Google later than HTML content. In the Vercel and MERJ data from nextjs.org, 25% of pages were rendered within 4 seconds, 75% within 26 seconds and 90% within about 3 hours. The same study found that 100% of indexable HTML pages were eventually fully rendered, which also shows that rendering spends part of the site’s crawl budget and server resources.

Types of rendering

Web developers choose between four main rendering strategies, described by Addy Osmani and Jason Miller in the web.dev guide “Rendering on the Web”.

Server-side rendering (SSR)

Server-side rendering means the server builds full HTML for every request. Bots receive complete content at once, which makes SSR the safest option for SEO on dynamic sites.

Static rendering (SSG)

Static rendering means HTML is generated once at build time and served from a CDN. It suits blogs, documentation and landing pages that change rarely.

Client-side rendering (CSR)

Client-side rendering means the browser downloads JavaScript that fetches data and builds the page. Without server help, bots that do not run JavaScript see an empty page.

Hydration and dynamic rendering

Hydration means server-rendered HTML becomes interactive after client scripts attach to it, the default model in frameworks such as Next.js and Nuxt. Dynamic rendering, which serves prerendered HTML only to bots, is called “a workaround and not a recommended solution” in Google’s documentation.

Why rendering matters for SEO and AI search

Rendering matters because content that exists only after JavaScript runs is invisible to many crawlers. Google renders, but Vercel’s December 2024 analysis found that none of the major AI crawlers render JavaScript: GPTBot, OAI-SearchBot, ClaudeBot, PerplexityBot and Bytespider read only the initial HTML. Gemini, which uses Googlebot’s infrastructure, and Applebot are the exceptions. A client-side rendered product page can therefore rank in Google and still be missing from ChatGPT or Perplexity answers, which hurts Generative Engine Optimization.

In SaaS and e-commerce audits I see three recurring problems: product descriptions loaded by scripts, internal links built from JavaScript click handlers instead of <a href> tags, and canonical or hreflang tags injected only after rendering. I cover the full checklist in the guide to on-page SEO.

SSR vs CSR for SEO