How to Check a Website for Broken Links — And What to Do With What You Find

You update a URL slug. You forget to update the three internal links pointing to it. Six weeks later, Google Search Console flags a spike in crawl errors, your bounce rate on that section is up 12%, and you're tracing the problem back to a change you made in April.

That's the normal way people discover broken links — after the damage is done. But a single crawl with a link checker takes under five minutes and surfaces every broken link on your entire site, before Google does.

This guide covers how to check a website for broken links properly: what the different error types actually mean, which tools find them, and what to fix first once you have the results.

What Counts as a Broken Link?

The term "broken link" gets used loosely, but from a technical standpoint there are four distinct failure modes, each requiring a different response.

4xx errors (client errors) are the classic broken link. A 404 means the resource simply doesn't exist — the page was deleted, the URL changed, or the link was typed incorrectly. A 410 means it's intentionally gone. These are content problems: fix by restoring the content, creating a 301 redirect to an equivalent page, or removing the link.

5xx errors (server errors) mean the server received the request but couldn't complete it. A 500 is a generic server error; a 503 means temporarily unavailable. These are usually transient — check again in 24 hours before taking action. Persistent 5xx on internal pages means something is broken in your application stack.

ERR (network-level failure) is the category most tools miss. When a link returns ERR, there was no HTTP response at all — the TCP connection failed before a single byte was exchanged. Causes include DNS resolution failure (the domain no longer exists), connection refused (server is down or port is closed), SSL handshake timeout, or a firewall dropping the connection silently. These are infrastructure problems, not content problems. Updating the link target won't fix them.

Blocked (403/429) is the category most tools get wrong. A 403 means the server got the request and refused it — typically because of bot detection, a CDN WAF rule, or IP-based rate limiting. A 429 is explicit rate limiting. Neither means the link is broken for real users. Social platforms (LinkedIn, X/Twitter, Instagram) return 403 to all automated crawlers while serving the page perfectly to human visitors. Treating 403 as broken floods your report with phantom issues that waste hours to investigate.

Why Broken Links Hurt SEO

The impact isn't abstract — there are three concrete mechanisms through which broken links cost you rankings.

Crawl budget waste. Googlebot follows your internal links to discover and index your content. Every time it follows a link to a 404, it uses a crawl slot on a dead end instead of discovering a new page. For large sites with limited crawl budget, internal broken links directly reduce indexation coverage.

PageRank leakage. Internal links pass authority between pages — that's the fundamental mechanism of how link equity flows through a site. A link pointing to a 404 passes that authority nowhere. A link pointing into a multi-hop redirect chain leaks some authority with each hop. Neither is catastrophic for a single link, but the cumulative effect of dozens of broken or chained internal links is measurable.

User experience signals. A visitor who hits a 404 mid-flow bounces. On mobile, where back-navigation is friction, that bounce is often permanent. High bounce rates from 404 pages feed back into Google's quality signals over time. One broken link in a conversion funnel can be disproportionately costly compared to its apparent severity.

Four Ways to Find Broken Links

The method you use determines how quickly you find problems and how comprehensively you find them.

Google Search Console is the right tool for ongoing monitoring — it surfaces broken links that real Googlebot actually encountered. But its data lags by days or weeks, and it only covers your own site. A link checker finds problems immediately across both internal and external links. For anything beyond a handful of pages, a crawl-based checker is the only practical approach.

Step-by-Step: Checking for Broken Links With ByteWaveNetwork

The process takes under five minutes for most sites. Here's exactly what to do.

1. Start the crawl. Go to ByteWaveNetwork's Link Checker, paste your site's URL, and click Start. The BFS crawler follows every internal link from your homepage outward, checking each URL it finds.

2. Watch the live stream. Results appear in real time — you don't wait for the full crawl to finish. Broken links show up immediately as they're discovered, so you can start prioritising before the crawl completes.

3. Use the filter tabs. The results table has separate tabs for Broken, Blocked, ERR, and Redirect. Start with the Broken tab — these are your actual broken links. Check ERR next. You can safely deprioritise the Blocked tab for external social links.

4. Check the "Found on" column. This is the key column most people overlook. It tells you which page contains the broken link — not just which URL is broken. That's where you need to make the edit.

5. Export to CSV. For sites with many broken links, or when sharing results with a developer, use the CSV export. The export respects your active filter tab — export just the Broken tab to share a clean fix list without the noise of Blocked and Skipped links.

What to Fix First — A Priority Order

Not all broken links are equal. Work through them in this order to get the most SEO impact per hour of work.

1. Internal broken links (4xx/5xx) — highest priority. These are pages on your own site that return errors. Google is actively crawling them, wasting budget and finding dead ends. Fix by creating 301 redirects to the correct URL, restoring the content, or removing the link if the content is genuinely gone.

2. ERR links — investigate immediately. An ERR on an internal link means something in your infrastructure is failing — DNS misconfiguration, a service that's down, or an SSL certificate issue. An ERR on an external link usually means the target domain no longer exists; remove or replace the link.

3. Redirect chains (3+ hops) — update source links to point directly to the final URL. Each hop adds latency and leaks a small amount of PageRank. For a deep dive on how redirect chains affect your link equity, see our pre-migration link audit walkthrough.

4. External broken links (404 on other domains) — fix when you have time. External 404s don't waste your crawl budget, but they do create a poor user experience. Replace with an equivalent resource or remove the link.

5. Blocked internal links (403) — investigate. Your own pages returning 403 to crawlers is unusual and worth understanding. Check whether a CDN rule or authentication gate is accidentally blocking crawlers from pages you intend to be public.

How Often Should You Audit for Broken Links?

The right cadence depends on how often your site changes. Here's a practical schedule:

• After any URL restructure or slug change — run a targeted crawl immediately. Internal links break silently when URLs change.
• Before a site migration — run a full crawl at least 72 hours before go-live. This gives you time to fix what you find before DNS cutover. Run a second crawl after launch to verify fixes are live.
• After a CMS update or template change — navigation links and footer links can break silently when template structure changes.
• Routine maintenance — monthly for actively published sites, quarterly for mostly static sites. External links rot over time as third-party sites restructure or go offline.

The pattern that catches most teams off guard is the cumulative external link rot. You publish content with external citations, those domains restructure their URLs 18 months later, and you accumulate 404s across old posts without ever noticing. A quarterly crawl keeps this under control before it becomes a credibility problem.

Affiliate disclosure: Some links on this page may be affiliate links. We only mention tools we've personally used and have an honest opinion about. Affiliate revenue helps keep ByteWaveNetwork's tools free and maintained. We are not paid by any of the tools compared in this article for favorable coverage.