How to Find Broken Links on Your Website Before Visitors Do

In short

A broken link is any link on your site that fails to reach a working page: a
404, a dead external site, a missing image, or a redirect that loops instead
of landing. Visitors leave the moment one fails, and each one is a small,
avoidable dent in trust. Checking manually catches the obvious ones; a
scheduled check catches the ones that break weeks after you last looked.
SiteHealth is a plain-English website health service that checks a site's
links, among 23 other things, on a repeating schedule.

Why do links break on their own?

Links break because the internet under them keeps moving, not because you
did anything wrong. A vendor page you linked to a while back gets reorganized. A
supplier shuts down their site. An image you uploaded gets deleted during a
redesign and every post that used it now shows a broken icon. Your own
internal links break just as easily: you rename a page, merge two blog
posts, or restructure your navigation, and every old link pointing at the
original URL now dead-ends.

None of this shows up as an error on your side. Your site still loads fine.
The failure only appears the moment someone clicks the specific link that
broke, which is exactly why so many go unnoticed for months.

What kinds of broken links should you actually worry about?

Not all broken links cost you the same amount, so it helps to sort them by
where they sit and who they hurt.

• Internal navigation links — your own menu, footer, or in-page links
• pointing to a page that moved or was deleted. Highest priority: every
• visitor who uses your site hits these.
• Internal content links — links inside blog posts or product pages
• pointing at another page on your own site. High priority, but easy to
• find and fix once you know where they are.
• Broken images — an image tag pointing at a file that no longer
• exists. Shows as a blank box or a broken-icon placeholder, and it reads
• as neglect even when the rest of the page is fine.
• External outbound links — links to other people's sites that changed
• their own structure. You cannot fix the target, only remove or replace
• the link.
• Social icons that point nowhere — a Twitter or Facebook icon still
• linking to an account you closed or a URL you typo'd when you built the
• site, often left unchecked for years because nobody clicks a footer icon
• during normal browsing.
• Redirect loops and long chains — a link that bounces through several
• redirects before landing, or one that redirects back to the same page it
• started on. No visible error, but a slower and often failed page load.

How do you find broken links before a customer does?

The manual version works for a small site, and it is worth doing at least
once even if you plan to automate the rest.

1. Open your site on a phone, using mobile data rather than office wifi, so
2. you are testing the same conditions a real visitor uses.
3. Click every item in your main navigation menu, one at a time.
4. Scroll to the footer and click every link there, including social icons.
5. Open your three or four most-visited blog posts or product pages and
6. click every link inside the body text.
7. Check that your images load. A broken image usually shows as a small
8. icon with alt text instead of the picture.
9. For a fuller sweep than clicking by hand allows, run a link checker like
10. the W3C Link Checker against your
11. homepage. It follows every link on the page automatically and reports
12. which ones failed and why, including redirect chains.
13. Note every broken link with its page and target, then fix or remove each
14. one rather than leaving a running list you never act on.

That covers the site as it exists today. The harder problem is tomorrow:
every new blog post, every deleted product page, every vendor that
reorganizes their site creates a new broken link you have not found yet,
and clicking through the whole site by hand every week is not something
anyone actually keeps doing past the second month.

This is the point where a scheduled check earns its keep. SiteHealth
crawls a site's links on a repeating schedule and flags a new break the
same day it appears, instead of waiting for a visitor to report it or for
you to remember to look. You can see exactly which checks it runs, link
checking among them, on the features page.

Does Google actually penalize a site for broken links?

Not directly, and not for one or two of them. Google's own documentation on
HTTP and network errors
describes how its crawler handles dead pages and server errors during a
crawl, treating scattered errors as normal rather than as a penalty signal.
The real cost is behavioral: a visitor who clicks a dead link on your site
leaves immediately, and a site with several broken links reads as poorly
maintained to anyone who happens to hit one, human or crawler.

Where broken links matter more is in aggregate, on a site with dozens of
them. That pattern tells a crawler the site is not well kept, which affects
how thoroughly and how often it gets crawled. For a small business site,
the direct loss of a frustrated visitor almost always outweighs any ranking
effect.

What should you check first if you have never done this before?

Start with anything a customer is likely to click on purpose: your main
navigation, your "contact us" and "get a quote" links, and any link inside
your top three pages by traffic. Those get the most clicks, so a break
there costs you the most. Save your footer, your old blog archive, and your
social icons for a second pass, since traffic to those is lower but they
still add up over a full site.

If you run a client site for someone else, the same priority order applies,
with one addition: check the client's own links to their booking tool,
payment page, or scheduling widget first, since those are the links most
directly tied to them getting paid. The
monthly client checklist
covers what else belongs in a recurring agency review.

What if you already use Google Search Console for this?

Search Console is a reasonable objection here, and it does report some
broken links Google's crawler encountered. But it only shows what Google
happened to crawl, on Google's own schedule, which can lag days or weeks
behind a change on your site. It also does not check external outbound
links, does not check images consistently, and gives no plain-English
explanation of which link on which page needs fixing first. It is a useful
second opinion, not a substitute for checking your own site directly on a
schedule you control.

Broken links rarely travel alone. A site that has not been checked in a
while for dead links usually has not been checked for the other quiet
failures either, the kind covered in
what uptime monitoring alone misses.

Fix what you can find right now

Pick your three most-visited pages, open them on your phone, and click
every link on each one. Anything that fails, fix or remove today. Then run
your homepage through the free checker at SiteHealth — it takes about
twenty seconds, needs no account, and shows you what else on the page might
be quietly broken alongside your links.