Broken Link Checker: Find Dead Links & 404 Errors on Websites

What is Broken Link Checker?

Broken Link Checker crawls a website and checks every link it finds, internal and external, then lists the ones that fail:

• broken links: 404 and 410 errors, and links to domains that no longer exist
• server errors (5xx), timeouts and unreachable servers
• redirect chains, hop by hop, when you ask for them
• for each bad link, the page where it was found and its anchor text, so you know exactly what to fix

It is built for SEO specialists, webmasters, agencies and developers who audit sites, prepare migrations or check a site every week.

Why use it?

• Fix links in minutes: every row says which page holds the bad link and what the link says.
• No false alarms from bot blockers: sites that refuse automated checks, such as LinkedIn (answers 999) or pages answering 403 or 429, are listed apart as blocked, not as broken.
• Accurate checks: links that refuse a light HEAD request are checked again with a full GET before they count as broken.
• Redirects explained: each hop and status (301, 302...) after a migration.
• Polite crawling: respects robots.txt and spreads requests.
• Pay per link checked, with a cap you set per site.

Because it runs on Apify, you also get scheduling (check your site every week), API access, integrations (Slack, email, Make, Zapier, n8n, Google Sheets) and exports to JSON, CSV or Excel.

What data do you get?

Field	What it tells you
url	The link that fails
outcome	broken, server-error, timeout, unreachable, blocked or redirect
status, error	The HTTP status, or the network error (not-found-host, connection-refused, ssl-error...)
linkType	internal (your site) or external
linkText	The link's anchor text
foundOn	Up to 10 pages that contain the link, with the link text on each
finalUrl, redirects	Where redirects lead, and each hop

The Summary in the Output tab gives the totals per site: pages crawled, URLs checked, and how many were OK, broken, failing, blocked or redirected.

How to use it

1. Enter one or more websites, such as https://example.com.
2. Set the maximum pages to crawl and the maximum URLs to check per site.
3. Choose whether to check external links, images and redirects.
4. Click Start. The Output tab lists the broken links, worst first.
5. To watch a site, save the run as a task and schedule it weekly, with a Slack or email alert.

How much does it cost?

You pay per URL checked, pages and links alike, each once per run: $0.0008 at the Bronze tier, less at Silver and Gold.

Usage	Cost (Bronze)
A small site: 40 pages, about 300 URLs	$0.24
A medium site: about 2,000 URLs	$1.60
500 URLs every week for a month	$1.60

Set Maximum URLs to check per site to cap what a run can cost.

Input example

{
    "startUrls": ["http://info.cern.ch/"],
    "maxPagesPerSite": 40,
    "maxLinksPerSite": 2000,
    "checkExternalLinks": true,
    "includeRedirects": true
}

Output example

Real results for info.cern.ch, CERN's restored copy of the first website, captured on 2026-11-05: 40 pages crawled and 287 URLs checked in 31 seconds.

Outcome	Links
OK	191
Broken	66 (38 domains that no longer exist, 28 pages answering 404)
Unreachable	29
Blocked automated checks	1
Redirected	10

A broken internal link:

{
    "website": "http://info.cern.ch/",
    "url": "http://info.cern.ch/Help/Webmaster",
    "outcome": "broken",
    "status": 404,
    "error": null,
    "linkType": "internal",
    "kind": "link",
    "linkText": "Webmaster",
    "foundOn": [{ "page": "http://info.cern.ch/hypertext/DataSources/WWW/Servers.html", "text": "Webmaster" }],
    "finalUrl": null,
    "redirects": []
}

A redirect chain:

{
    "url": "http://fatty.law.cornell.edu./usr2/wwwtext/lii.table.html",
    "outcome": "redirect",
    "status": 200,
    "linkType": "external",
    "linkText": "Cornell Law school",
    "finalUrl": "https://www.law.cornell.edu/",
    "redirects": [
        { "url": "http://fatty.law.cornell.edu./usr2/wwwtext/lii.table.html", "status": 301 },
        { "url": "http://www.law.cornell.edu/", "status": 301 }
    ]
}

Use cases

• SEO audits: broken links waste crawl budget and frustrate visitors.
• Site migrations: list every redirect chain and every link that broke.
• Weekly monitoring: schedule a check and get an alert when a link breaks.
• Content maintenance: find dead external sources in blog posts and documentation.
• Agency reporting: one list of fixes per client site.

FAQ

What counts as broken?

A 404 or 410 answer, any other 4xx answer except 401, 403 and 429, and a link to a domain that no longer exists. Server errors (5xx), timeouts and refused connections are reported with their own outcome, so you can tell a dead page from a server having a bad moment.

Why are some links "blocked"?

Some sites answer every automated request with 401, 403, 429 or 999 (LinkedIn), even for pages that work in a browser. They are listed apart, and you can hide them with Report links that block automated checks.

Does it run JavaScript?

No. It reads the HTML that the server sends, which holds the links on most sites. Links added only by JavaScript in the browser are not seen.

Does it follow robots.txt?

Yes, by default: pages that the site's robots.txt keeps crawlers out of are skipped.

Limits

• Up to 20 websites per run, 10,000 pages crawled and 50,000 URLs checked per site.
• Pages larger than 3 MB are checked but their links are not read.

Other Actors you may find useful

Actor	What it does
Website Contact Scraper & Email Extractor	Emails, phones and social profiles from a list of websites
Keyword Search Volume Checker	Google search volume, CPC and difficulty for your keywords
AI Share of Voice	How often AI assistants recommend your brand

Run it from the API

from apify_client import ApifyClient

client = ApifyClient("YOUR_TOKEN")
run = client.actor("spokentext/broken-link-checker").call(run_input={
    "startUrls": ["https://example.com"],
    "maxPagesPerSite": 200,
})
for link in client.dataset(run["defaultDatasetId"]).iterate_items():
    if link["outcome"] == "broken":
        print(link["url"], link["status"], "on", link["foundOn"][0]["page"])

Feedback

Something failed or gave a wrong result? Open an issue in the Issues tab of this page, with the error message from your results: every report is read and fixed. If the Actor saved you time, a rating helps others find it.