Rendering
Rendering is the step where a search engine executes a page's JavaScript to see the content that only exists after scripts run. It happens on a deferred schedule after the initial HTML is processed, so script-generated content is discovered later than content present in the source.
A crawler fetches HTML. If that HTML contains the page’s text, indexing can proceed immediately. If the text is built in the browser by JavaScript, the engine has to run a headless browser to see it — which costs far more resources than parsing HTML, and so is queued separately.
Google has described this as a two-wave process: the raw HTML is processed first, and rendering follows when capacity allows. The delay is not fixed and not observable from outside.
Three failure modes are worth knowing:
- The render times out, and the page is indexed as whatever the raw HTML contained.
- A resource the render needs — a script, an API endpoint — is blocked in
robots.txt, so the content never appears. - The content depends on user interaction, such as a click, which no crawler performs.
The reliable test is to fetch your page with JavaScript disabled, or use the Live Test in URL Inspection and read the rendered HTML it returns.