Skip to content

Soft 404 Checker

This soft 404 checker asks a site for pages that cannot exist and reads the answer.

Try it on or paste any public URL. A page inside a directory gives the nested probe somewhere to go.

Note

We request the URLs from our server as a desktop browser, follow each redirect ourselves and read the status codes and the HTML exactly as the server sent them, before any JavaScript runs. The made-up URLs end in -nf and a random name, so they cannot match a real page, and requesting them changes nothing on the site. Nothing is stored.

It requests a random made-up path at the root, the same path inside a real directory of the site and once more with .html on the end, follows every redirect and shows each hop with its status code. A missing page should end at a 404 or 410. If it ends at a 200, either served directly or after a redirect to the homepage, crawlers cannot tell your real pages from fake ones. It also reads the page you entered for the signs Google uses to label a single URL a soft 404: not-found wording in the title or H1, an almost empty body, a redirect to the homepage and text that matches the site's error page. No sign-up and no email.

Use this tool often? Bookmark this page, so it is one click away next time.

Email me this result

Want a copy for your records or a client? Leave your email and we'll send you this result. It is written by a real person at Crawl Cove, not an automated blast, so it may take a little while to arrive. We'll only use your address to send it, unless you also tick the box below. Privacy policy.

That email address doesn't look right. Please check it and try again.

Something went wrong saving your address. Your result above is unaffected; please try again.

Got it. We'll email your result to you. Nothing else lands in your inbox unless you ticked the tips box.

How to check a site for soft 404s

  1. Paste a URL

    Enter any public address, with or without https://. A deeper page is better than the homepage, because its directory is where the nested probe goes and its own content is checked as well.

  2. Read the verdict

    Missing pages answer 404, some missing URLs answer 200, or missing pages answer 200 across the site. The last one is the finding Crawl Cove raises when it crawls the site, in the same words.

  3. Check each probe

    Each made-up URL is listed with every hop, its status code and where it ended up, so a catch-all redirect shows exactly where it sends visitors and crawlers.

  4. Apply the fix

    The fix under the result matches what the probes found: send a real 404 status, remove the catch-all redirect, or handle not-found routes in a JavaScript app. Run the check again once the change is live.

What this soft 404 checker tests

Two things. First the server: does it answer a URL that does not exist with a 404 status, or with a page that says 200? Three made-up URLs are requested, each with a random name ending in -nf so it cannot match a real page: one at the root of the site, one inside a real directory taken from the page you entered or its links, and one at the root ending in .html, because some servers hand file-like paths to a different handler. Every redirect is followed and each hop is listed with its status. Then the page: if the URL you entered answers 200, it is checked for the signs of a soft 404 one page at a time.

Soft 404: what it means and how to fix it

Why a made-up URL is the test

A soft 404 is a page that tells people it is missing while its status code tells crawlers it is fine. Google's documentation calls returning a success code for a missing page a bad practice, because Google may keep crawling such URLs instead of your real pages. The quickest way to see whether a server does this is to ask for a page that cannot exist. One 200 can be a coincidence, a single directory with its own fallback, so Crawl Cove only reports a trap when a made-up URL at the root and another inside a real directory both end at 200. This tool uses that same rule and the same wording, so the tool and the app agree. The .html probe is extra, and on its own it never makes the trap verdict.

How crawl budget works

Catch-all redirects to the homepage

Many small sites redirect every unknown URL to the homepage, often through a plugin setting or a rewrite rule. It feels tidy, but Google treats a redirect from a missing page to an unrelated page, the homepage included, as a soft 404. The checker follows each redirect to the end: a made-up URL that lands on a 200 page counts as a 200, and the result names the page it landed on. A redirect that ends at a 404 page is fine, though the extra hop is wasted. When we sent one made-up URL to each of 1,000 UK domains in October 2026, 489 gave a readable answer: 33 served a soft 404 and 41 more redirected the missing URL to their homepage.

the 1,000 site soft 404 study · trace any redirect hop by hop

Signs one page is a soft 404

Google decides that a page is a soft 404 from what it sees on the page, after it renders it, so no outside tool can give you Google's verdict. What this checker gives you is the signals that usually lead to it, and it labels them as a heuristic: a title or H1 that says not found, page not found, 404 or does not exist; a short page with the same wording in its text; fewer than 50 words of body text; a deeper URL that redirects to the homepage; and text that matches the page the server sends for a made-up URL. It reads the HTML as the server sends it, before any JavaScript runs. On a site that builds its content with JavaScript the body can look empty here while Google, which renders scripts, sees more. A long article that mentions "page not found" is not flagged.

see a page as a crawler reads its HTML

How to fix a soft 404

Match the fix to what the probes found. If missing URLs are served a 200 page, keep the friendly not-found page but send it with status 404, or 410 for a page removed for good; in a CMS that is usually a theme template or plugin, in a framework the last catch-all route, and on a static host a rewrite that sends every path to the index page. If missing URLs redirect to the homepage, remove the catch-all redirect and keep redirects for pages that have a genuine replacement; on Apache, an ErrorDocument 404 line that names a full URL makes the server redirect instead of serving the error page. In a JavaScript app with routing in the browser, have the server return 404 for unknown paths, or follow Google's advice: add a noindex robots meta tag to the not-found view with JavaScript, or redirect with JavaScript to a URL that answers 404. For a real page Google flagged, add the content it is missing.

Not found (404) in Search Console · Page with redirect in Search Console · check the status and headers of any URL

Frequently asked questions

What is a soft 404?
A page that tells visitors it does not exist, or has nothing on it, while the server answers with status 200, the code for a working page. Google spots these from the content, leaves them out of the index and reports them as Soft 404 in the Page indexing report.
Is this soft 404 checker free?
Yes. It is free, needs no sign-up and no email, and the only limit is a rate limit of ten checks a minute, because each check sends several requests to the site.
What is the difference between a 404 and a soft 404?
A 404 is a status code: the server says the page is not there, whatever the page looks like. A soft 404 is a page that looks missing or empty but is sent with a 200 status, so a crawler has to guess from the content. A custom not-found page sent with status 404 is not a soft 404.
Should I redirect missing pages to my homepage?
No. Google treats a redirect from a missing page to an unrelated page, the homepage included, as a soft 404, so it fixes nothing and hides the problem from you. Redirect a removed page only to a page that replaces it, and let everything else answer 404 or 410.
Why does the checker request URLs that do not exist?
Because that is the only way to see how the server treats a missing page. The names are random and end in -nf, so they cannot match a real page. Three are sent, one at the root, one inside a real directory and one ending in .html, and none of them changes anything on the site.
Does it see the page the way Google does?
It sees what Google sees first: the status code and the HTML the server sends, before JavaScript runs. Google renders the page later and makes its soft 404 call on the rendered content, so on a JavaScript site Google may see more text than this check does. For your own pages, the URL Inspection live test in Search Console shows the rendered page.
Why does Search Console say Soft 404 when my page loads fine?
Because Google judged the content, not the status. The usual causes are a page with almost no text, an empty search or category page, content that only appears after a script that failed for Googlebot, or a redirect to an unrelated page. Enter that URL here to see which signs it shows.
Can I check every page of my site for soft 404s?
Not here: this tool checks the server and one URL at a time. Crawl Cove, the desktop crawler, runs the same made-up URL probe on every crawl, and it flags thin pages and links to missing pages among the pages it crawls. It has a permanent Free plan.
Can I share the result?
Yes. After a check the page address includes the URL you checked, so copying it gives a link that runs the same check. Results are cached for ten minutes.

From one page to your whole site

This tests the server's answer for missing pages and one URL. Crawl Cove runs the same probe on every crawl and checks the pages it crawls for thin content and links to missing pages, so the soft 404 trap and the pages that cause it show up in one report. See Soft 404 in Search Console: fix it or leave it? or run a full site audit.

Download free