SEO · FREE

Find the broken links
on your site

A single dead link costs trust — and search engines notice too. Four ways to check your whole site, from zero-install to fully automated in CI.

Updated August 2026 · 5 min read

Check any page for GDPR & cookie issues: Run the Free Scanner →
Ask any EU compliance question — EAA, NIS2 or GDPR — and get a practical answer in seconds: 🤖 Ask the Compliance AI →

Four methods — from fastest to most thorough

  1. Online checker: paste your domain into a free broken link checker. Done in minutes, but depth is limited on large sites.
  2. Sitemap crawl: fetch sitemap.xml and test every URL plus its links. Covers exactly the pages you consider official.
  3. Terminal: wget --spider or a dedicated CLI tool gives full control and output you can pipe into scripts.
  4. CI automation: run the check on every deploy so new broken links never reach visitors in the first place.
# Check one URL's status code
$ curl -s -o /dev/null -w "%{http_code}" https://yoursite.com/old-page

# Crawl an entire site with wget (status only, nothing downloaded)
$ wget --spider -r --no-parent https://yoursite.com 2>&1 | grep -B2 "404"

Once you've found them: how to prioritize

Sort by where the link lives, not just by the error itself. One dead link on your most visited page costs more than twenty on an archive page. Check first:

LocationWhy it matters most
Homepage and main navigationFirst impression — everyone sees them
Pages with the most organic trafficDirect loss of visitors
Pages with many backlinksWastes value others gave you
The conversion pathBreaks the sale midway
Get the full picture of your site →

Read next

Also read: how to find all pages on a website, check redirect chains and technical SEO checks.

Frequently asked questions

Why are broken links a problem?

They send visitors and search engines into 404 pages. That costs trust, wastes link equity from internal and external links, and can hurt rankings over time. Google Search Console reports them directly under coverage.

What's the fastest way to check one site?

An online broken link checker or a sitemap crawl. Paste the URL, wait a few minutes, get a list of every link with its HTTP status. For larger sites a CLI tool is faster and more controllable.

How do I catch them automatically going forward?

Run a link check in your CI pipeline on every release, or schedule a weekly crawl of sitemap.xml. GitHub Actions is free for public repositories and needs no server.

Should every 404 be fixed?

No — some are intentional (removed pages). But they should serve a proper 404 page, and important old URLs deserve a 301 redirect to relevant content. Prioritize links from pages with the most traffic and backlinks.

Try the site analysis now →

Danish version of this guide

Related: Find all pages · Redirect chains · Technical SEO checks