Soft 404: a 200 page that's really an error
A soft 404 returns HTTP 200 while the content says "not found". Google won't index it and wastes crawl budget on it. Find soft 404s free and return a real status code.
What it means
A soft 404 is a page that returns HTTP 200 while its content is empty or an error message. The URL says "everything is fine" but the body says "page not found" or "no results". Google sees the mismatch, decides the page has no real content, and treats it as a 404 anyway — while still spending crawl budget rediscovering it.
Why it matters
Because the 200 status lies, Google cannot use it to prune dead URLs. Its systems inspect the rendered content, judge the page thin or error-like, and refuse to index it — the URL then sits under "Crawled – currently not indexed" or is flagged as a soft 404 in Search Console, and Googlebot keeps recrawling it because a 200 invites return visits. The cost is twofold: crawl waste (out-of-stock products, empty search pages and deleted articles served as 200 drain fetch quota that should go to real pages) and quality signals (a site that serves many thin, error-like 200s looks lower quality in aggregate). Returning the correct status lets Google drop the URL cleanly.
How to fix it
- Identify the URL's real state: genuinely gone, temporarily unavailable, or valid but empty by mistake.
- For genuinely missing pages return a true 404 Not Found (or 410 Gone if permanent) — do not render a friendly "not found" page with a 200 status.
- For content that moved, return a 301 redirect to the replacement URL.
- For legitimately empty states you still want live (e.g. an empty category), add real content or noindex the page so Google does not read the emptiness as an error.
When it's not a problem
Crawlinx requires two conditions together before flagging: thin content AND a not-found marker ("not found", "404", "no results", "page doesn't exist") in the title, H1, or first heading. A rich article that merely mentions "404" in its body is not flagged, and pages already carrying noindex are never flagged — the owner is clearly handling the state deliberately. If your empty-results page is intentional and noindexed, no action is needed.
How common is it?
1 audited site in our corpus currently show this issue. The breakdowns below show which platforms, gatekeepers, verticals and countries are most exposed.
Related guides
Audit your own site — free
77 checks, internal PageRank, render-diff. No signup, results in ~30s.