Hill Country Bots

How to find broken links on a website


Published November 12, 2026 by Hill Country Bots

A broken link is a link that no longer takes the visitor where they expected. The page might have been moved, renamed or taken down, or the other site might be gone. Clicking it wastes time, and a long list of them makes a site look neglected. To fix one, you need two things: the dead address, and the page that still points to it.

The short answer

  1. If you only need to check a few pages, do it in the browser, or with a short curl command, as shown below.
  2. If you own the site, open the Pages report in Google Search Console™ and look at Not found (404). That shows URLs Google has already tried to crawl.
  3. For a whole site, crawl it (or load the sitemap), record the status of every unique link, and keep the source page and anchor text so you know what to edit.

What counts as a broken link

Treat these as problems:

A single 301 or 302 is not a broken link, but it is worth updating the href to the final URL so visitors and crawlers skip the extra hop. Some sites answer automated checks with 401, 403, 429 or 999. Those are often blocks, not proof the page is dead. Open those in a browser before you remove them.

A soft 404 is trickier. The server returns 200, but the page is an empty or "not found" template. A status-code checker will not catch that on its own. You still have to look at the content.

Method 1: a free check you can do yourself

Check one page in the browser

Open the page, then right click and choose View Page Source, or press Ctrl+U (Cmd+Option+U on a Mac). Search for href= and scan the addresses. For each one you care about, paste it into a new tab and see what loads. A free browser extension can click through the links on the open page for you, which is faster for a single screen. This is fine for a homepage, a nav template or a post you just edited. It does not scale to a whole site.

Check a short list from the command line

If you already have a list of URLs, a terminal can tell you the status without opening each tab. This prints the first status code and the final address after redirects:

curl -sI -o /dev/null -w "%{http_code} %{url_effective}\n" -L "https://example.com/old-page"

Drop -L if you want to see the first response only, which is how you spot a 301. You can wrap that in a small loop over a text file. Stay polite: space the requests out, and do not use this on sites you do not have a reason to check.

If you own the site, use Search Console

In Google Search Console, open the Pages report under Indexing and look at the Not found (404) and server-error groups. This is the best free view of broken URLs Google has actually requested. It will not list every outbound link on your pages, and it will not show a 404 that Google has never tried to crawl. Use it as evidence, then still crawl your own pages for the rest.

Desktop crawlers such as Screaming Frog also find broken links, and they have a free mode for smaller sites if you want to run the crawl on your own machine. We wrote an honest comparison, with today's prices and where Screaming Frog wins, at Screaming Frog alternative.

Method 2: check a whole site at once (the tool we built)

When the site has more than a handful of pages, you want one table: every unique URL, its status, any redirect chain, and the page each dead link sits on. That is what our Broken Link Checker does. We built this tool. It runs on Apify.

You can crawl a site from the homepage, paste a URL list, or load the XML sitemaps and check every address they list. Sitemap mode is a good way to catch dead pages the sitemap still advertises. By default the checker respects robots.txt. Disallowed URLs are reported as skipped and are not charged. It uses plain HTTP, not a full browser, so links that appear only after JavaScript runs are not seen, and pages behind a login are never crawled.

  1. Open the Broken Link Checker on Apify. A free Apify account is enough. The form is prefilled, so Start works right away.
  2. Keep Crawl for a site check, or switch to List or Sitemap.
  3. Set a cap. 50 pages and 1,000 URLs is a safe first pass. Raise it for a deeper check.
  4. Run it. Broken links come first in the results. Download JSON, CSV or Excel, or grab the ready-to-send broken links CSV.

It costs $0.80 per 1,000 unique URLs that returned an HTTP status, including 404s. There is no start fee. Invalid inputs, dead domains, timeouts, robots.txt skips and over-cap URLs are free. A 100-page site with 1,500 unique links costs about $1.20. That is the live Apify Store price as of November 12, 2026.

A real sample

From a real run on November 12, 2026, at 7:58 AM Central: crawl mode on crawlee.dev, 25 pages crawled, 300 URLs checked in 46 seconds. Result: 1 broken external link, 50 redirects, 3 blocked by the target site (Stack Overflow answers bots with 403), 231 ok, and 15 skipped because the target's robots.txt disallows bots.

URLWhat we found
https://jobops.app/Broken (402). Found on crawlee.dev/blog, anchor JobOps.
https://www.crawlee.dev/blog/launching-crawlee-pythonRedirect, 301 to 200. Found on 1 page.
https://sdk.apify.com/docs/guides/getting-startedRedirect, 301 to 302 to 200. Found on 13 pages.
https://stackoverflow.com/questions/tagged/crawleeBlocked (403). Found on crawlee.dev.

And the broken row as the tool returns it, shortened:

{
  "url": "https://jobops.app/",
  "isBroken": true,
  "verdict": "broken",
  "statusCode": 402,
  "foundOn": ["https://crawlee.dev/blog"],
  "foundOnCount": 1,
  "anchorText": "JobOps",
  "linkType": "external",
  "status": "ok",
  "errorCode": "HTTP_402"
}

The 402 is the finding. The row also tells you the link was on crawlee.dev/blog, with the anchor text JobOps, so you know exactly what to edit.

How to fix what you find

Work through the broken rows first, then the long redirect chains.

After you make the edits, run the check again on the pages you changed. New 404s show up after content moves, so this is worth repeating when you launch a redesign or retire a section.

Questions

How do I find broken links on a website for free?

For a few pages, open the source, scan the href links and check each one in a new tab, or use a short curl command to print the status code. If you own the site, the Pages report in Google Search Console shows 404s Google has already tried to crawl. That will not catch every outbound link on your pages.

What counts as a broken link?

A 4xx or 5xx status (except common bot-block codes such as 401, 403, 429 or 999), a domain that does not resolve, a refused connection, a TLS error, or a redirect loop. A single 301 or 302 is not broken, but the link should usually be updated to the final URL.

Does Google Search Console show every broken link?

No. It shows URLs Google has requested and classified, such as Not found (404). It will not list every outbound link on your pages, and it will miss a 404 that Google has never tried to crawl. Use it together with a crawl of your own site.

How do I find broken links on a whole website?

Crawl the site, or load its sitemap, and record the HTTP status of every unique link. Keep the source page and the anchor text so you know what to edit. Our Broken Link Checker on Apify does that and returns a CSV, at $0.80 per 1,000 URLs.

Are JavaScript links included in a crawl?

Not in a plain HTTP checker, including ours. Links that appear only after JavaScript runs are not seen, and pages behind a login are never crawled.

Related

Google Search Console is a trademark of Google LLC. Hill Country Bots is not affiliated with, endorsed by or sponsored by Google LLC.