Website checker cannot reach your page: what to do
Understand failed website checks, record the evidence and decide what to retry or ask your developer to investigate.

You run a check on your website and instead of findings you get "could not reach this page". It's the digital equivalent of the delivery driver leaving a card saying they called, when you were sitting in the front room the whole time. Annoying, but useful. A failed check from a third-party tool doesn't, on its own, tell you whether Google or anyone else was affected. To know that, look at the response Google actually received, in Search Console or your server logs. What the failed check does give you is a lead worth following. Here's how to work out which kind of failure it is and what to do about it.
Start with what the error actually said
"Couldn't reach" covers several different problems, and the fix depends on which one you've got. Look for:
- a status code such as 403, 404, 429 or 503;
- a timeout, where the page took too long to answer;
- a network or DNS error, where the checker couldn't find or connect to the server at all.
Write down the exact message, the page address and the time. You'll need all three if this goes to your web person or your host.
Rule out the simple causes
Before anything technical, check the obvious:
- Does the page open for you? Use a private browser window so you aren't seeing a logged-in version.
- Is the address right? Look for typos, a missing "www", or http instead of https.
- Is the page published? Builders often have draft, password-protected and members-only settings.
- Has it moved? An old address with no redirect will fail even though the page exists somewhere else.
When the server says no
Codes in the 400s mean the server received the request and either refused it or couldn't find the page. One of them, 429, is different, and it's covered in the next section.
- 404 or 410: the server is reporting that page as not found, or gone, for that request. That isn't proof the content has disappeared everywhere. If the page should exist, restore it. If it has genuinely moved, redirect the old address to a relevant replacement, not just the home page.
- 401 or 403: the server is refusing access. This is often a login wall, a security plugin, or a firewall or bot-protection service that blocks automated visitors.
A 401 or 403 doesn't mean your page is gone. It means that visitor wasn't let in. For Google specifically, its documentation says Googlebot treats 400, 401, 403, 404, 410 and 411 responses the same way: it ignores the content, and pages already indexed are removed from Search. So a page your customers can open, but which returns 403 to Googlebot, may drop out of Google. Other crawlers, including AI ones, have their own rules, and Google's documentation doesn't speak for them.
If you use a security or bot-protection service, check its settings or logs for blocked requests. If you want search crawlers to reach the page, make sure they aren't being challenged or blocked.
When the server says "not now"
A 429 means "too many requests". Codes in the 500s mean something went wrong on the server: it's overloaded, misconfigured or temporarily down.
For Google, 429 is the exception among the 400s. Its documentation treats 429 as a sign of an overloaded server, the same as a 5xx error, not as a missing page. Google's crawlers temporarily slow down. Pages already indexed are kept at first, but URLs that keep returning server errors are eventually dropped. Google gives no timeframe for this, so don't assume any outage is safely short. A server that keeps returning errors to Google's crawlers is a real problem.
What to do:
- try again later, at a quieter time;
- check whether your host reports downtime or resource limits;
- if it keeps happening, send your host the times and the pages involved.
Timeouts and intermittent failures
A timeout means the page didn't answer in time. Common causes are a slow server, a heavy page or a problem along the way.
If the same page sometimes loads and sometimes doesn't, treat that as a finding in its own right, not bad luck. Note when it fails. A pattern at particular times often points to load or scheduled jobs on the server.
Don't fill the gap with guesses
If a check couldn't reach a page, it can't verify that page's current content. Earlier, dated findings for the page may still exist, and the access failure is itself evidence worth recording. What it isn't is a pass.
A good report shows which pages it checked and which it couldn't reach. It doesn't leave the failures out. If you've just fixed something and the recheck fails to load, record the repair as not yet verified until a check actually succeeds.
That's how CAFMO handles it. The list of checked pages shows coverage and any pages it couldn't reach, rather than hiding them, and you can recheck a page once it's reachable. See how CAFMO checks work.
Retry, then escalate
A simple order:
- Retry once or twice, a few minutes apart.
- Check the page in a private browser window and on mobile data.
- Rule out the simple causes above.
- If it's a 4xx code other than 429, look at login, publishing and security settings.
- If it's 5xx, 429 or a timeout that keeps happening, send your host or developer the page address, the exact error and the times. Our website handover checklist covers agreeing who owns the hosting and the accounts.
Frequently asked questions
The page opens fine for me. Why can't the checker reach it?
Your browser and an automated checker don't look the same to a server. Firewalls, bot protection and some security plugins treat automated visitors differently, and you may be seeing a logged-in or cached version. Test in a private window, and check your security settings.
Will one failed check hurt my search visibility?
A failed check from a third-party tool doesn't tell you whether Google was affected. Look at the response Google actually received, in Search Console or your server logs. For Google's own fetches, its documentation says repeated server errors can eventually lead to URLs being dropped. It gives no safe duration, so fix recurring errors rather than waiting.
Should I block all bots to stop spam?
Be careful. Blocking every automated visitor can also block the search and AI crawlers you want to reach your pages. Block what you need to, and check what you've allowed. Our guide to checking whether ChatGPT can find your website covers the access side.
Sources and further reading
- Google for Developers, "How HTTP status codes affect Google's crawlers" (last updated 4 Feb 2026): https://developers.google.com/crawling/docs/troubleshooting/http-status-codes. Primary source, read at 09:58 UTC on 16 Nov.