Why can a page that looks like 404 still be HTTP 200?

Question

Weirdo · Accepted Answer

No 404 was sent. “Not found” is copy in the HTML. HTTP `200 OK` means the request succeeded: a custom error page or a single-page app’s client-side route can still travel as 200. Google calls that a soft 404.

What the page says is not the status code. MDN defines `200 OK` as a successful request. For GET, the server retrieved a resource and put it in the body. That resource can be a custom “not found” template and still be a success.

`404 Not Found` is the code that means the server cannot find the requested resource. MDN treats a custom 404 page as something you show visitors — the status remains 404. Google Search Central is blunter: custom 404 pages exist only for users; they are useless to a search engine, so the server must return HTTP `404` or the page can be indexed. Serve the same HTML with `200`, and the protocol records a hit.

Single-page apps often route in the browser. Google’s JavaScript SEO guide says that with client-side rendering and routing, meaningful HTTP status codes can be impossible or impractical: an unknown URL may still receive the app shell as `200`, then JavaScript paints “not found.” Their workarounds: redirect in JavaScript to a URL whose server responds `404` (for example `/not-found`), or add a `noindex` robots meta tag on the error view.

Google calls a page that tells the user it does not exist while returning 200 a soft 404; empty pages and pages with no main content count too. Once the algorithms decide from the content that it is an error page, Search Console’s Page Indexing report shows a soft 404, and the URL is excluded from Search. If the content is gone, send `404` or `410`. If it moved, send `301`.

Why can a page that looks like 404 still be HTTP 200?

Success is the request, not “this page exists”

Client-side routing, and the soft 404