How to Check for Broken Links on Your Website: 5 Methods Compared

Broken links cost you SEO position, frustrate visitors, and quietly bleed conversions out of every flow that depends on them — checkouts, signups, documentation. The hard part isn't acknowledging that. The hard part is picking a method that actually finds them without wasting your week.

This is the practical, opinionated walkthrough of the five distinct approaches. Each one is right for a specific kind of site and a specific kind of effort budget.

Why this matters in 30 seconds

• SEO. Search engines treat outbound rot as a quality signal and burn crawl budget on dead destinations.
• Conversion. A broken link in the checkout funnel is a measurable revenue hole.
• Trust. Visitors don't separate "your site" from "the third-party script your site loads." A red browser warning on your URL becomes your problem regardless of root cause.
• Compliance. Regulated industries get audited on outbound-link hygiene. "We didn't notice" doesn't pass.

The actual question is which method matches the site you have.

Method 1: Manual checking

The 1997 approach: open the page, click every link, write down what doesn't work.

Pros. Free. Zero learning curve. You see exactly what users see, with the context. Useful for verifying a single critical page (checkout, signup, the most-visited landing page) after a release.

Cons. Doesn't scale past about 20 pages. Easy to miss a link in the footer or an inline reference. Can't catch links that broke since you last looked.

Best for. Spot-checks of specific pages. Confirming a fix worked. Manual QA on a small portfolio site. Almost never the primary method.

Time budget. Two to five minutes per page if you're being thorough. Multiply.

Method 2: Browser extensions

Install something like Check My Links (Chrome), Link Checker (Firefox), or Broken Link Checker (Chrome). Click the extension on the page you're viewing; it tests every link on that page and color-codes the results — green for working, yellow for redirects, red for broken.

Pros. Fast on a single page (5–30 seconds for a normal article). Free. Works on pages behind login because the browser is already authenticated. No setup beyond installing.

Cons. One page at a time. No site-wide scan, no automation, no monitoring over time, no exportable report worth keeping.

Best for. Editors checking their work before publishing. Spot-checking after a content update. Verifying authenticated pages that crawlers can't see.

Time budget. Roughly a minute per page. Tractable for 50 pages, painful at 100, untenable past that.

Method 3: Free web-based crawlers

Paste a URL into one of the hosted tools and let it crawl your site:

• BrokenLinkCheck.com — free up to 3,000 pages
• DeadLinkChecker.com — free single-site checks, paid tier adds scheduling and multi-site
• W3C Link Checker — free, standards-focused, no page cap (and no UI polish)
• Dr. Link Check — free tier, adds SSL and basic reputation checks

The crawler runs from their servers. You give it a URL, wait somewhere between 5 minutes and an hour, get a list of broken links with the source page and the error code.

Pros. No install. Site-wide. Free for small-to-medium sites. Exportable CSV reports. Handles the most common ask ("just find me the 404s") well.

Cons. Public pages only — anything past /login is invisible. JavaScript-rendered links often get missed unless the tool has a headless-browser mode. Free tier page caps land somewhere between 1,000 and 3,000. Your server may rate-limit or block the crawler if it's aggressive. No monitoring between scans.

Best for. A one-time audit of a small-to-medium public site. The "I want a CSV of what's broken right now" use case.

Time budget. 5–60 minutes for the crawl plus 10–30 minutes to triage the report. Call it an hour or two for a single audit pass.

Method 4: Desktop crawlers

Real software you install on your machine:

• Screaming Frog SEO Spider — free up to 500 URLs, paid for unlimited; the de-facto pro standard
• Xenu's Link Sleuth — free, Windows-only, last meaningfully updated some time ago but still works
• Integrity — free, Mac-only
• Sitebulb — paid, broader auditing scope than just link checking

The crawler runs locally so it isn't subject to a hosted tool's page cap or rate limits. Configuration surface is much larger: control crawl depth, user-agent, request rate, what kinds of links to follow, headless-Chromium rendering for SPA sites, custom extraction rules.

Pros. Handles large sites (Screaming Frog routinely scales past 100k URLs with enough RAM). Detailed technical output beyond just broken links — redirect chains, missing meta, slow responses. Headless-browser mode catches JS-rendered links the free tools miss. Local execution means no third-party data sharing.

Cons. Real learning curve. Free tier is meaningfully capped on the most popular tool (Screaming Frog). Resource-intensive on the host machine during a big crawl. Still a manual operation — you have to remember to run it. Paid licenses are non-trivial ($259/year for Screaming Frog at this writing, plus or minus).

Best for. SEO consultants and in-house teams doing quarterly deep dives. Sites large enough that the free hosted tools cap out. Anyone who wants the broader audit signal (not just broken-link detection).

Time budget. 10–15 minutes to set up the first time, 10–120 minutes per crawl, 15–30 minutes to triage. Call it 1–3 hours per audit pass.

Method 5: Continuous monitoring services

The category that replaces "scan-and-report" with "watch-and-alert." You install something on the site (a script tag, a CMS plugin, or a Tag Manager entry), and the service watches the outbound link inventory in the background, firing alerts when state changes.

The real players in 2026:

• LinkSentry. Crowdsourced discovery — outbound destinations are reported from real visitor sessions instead of being crawled. Continuous checking against threat-intel, parking, TLS, content-drift, and re-registration signals. Coverage matches what users actually load (behind auth, dynamic, personalized). Starter from $19/mo.
• Ahrefs Site Audit. Part of the Ahrefs SEO suite. Crawler-based, weekly cadence. Right answer if you already pay for Ahrefs.
• SEMrush Site Audit. Same shape, inside SEMrush.
• Sitechecker. Standalone, simpler interface, daily checks.

Pros. Continuous. Catches issues in minutes instead of weeks. Detects past 4xx — re-registered domains, parking, malware/phishing/threat-intel, TLS, content drift. Email/Slack/webhook alerts. Historical data for trends. Scales to any site size. The category that closes the actual detection-time gap.

Cons. Real subscription cost. Initial setup (script tag, configuration of policies and alert preferences). The "I just want to run one scan" use case isn't its strength — for that, a free crawler is better.

Best for. Anything where outbound rot matters over time: e-commerce, SaaS dashboards, content platforms, regulated-industry public sites, agencies managing multiple client properties.

Time budget. 15–30 minutes one-time setup. ~zero ongoing time investment. 5–10 minutes a week reviewing the dashboard or replying to alerts.

How to actually pick

Match the method to the site rather than the marketing:

Practices that work with any method

• Match cadence to traffic and revenue exposure. Continuous for revenue-bearing flows, weekly for content sites, monthly is the floor for anything important.
• Triage by impact, not by alphabetical order. Checkout > homepage > top landing pages > recent posts > archives. Fix in that order.
• Fix at the source. Internal: update the URL. External: replace, remove, or — if it's borderline — rel="nofollow". Don't paper over with redirects that rot in their own right.
• Avoid redirect chains. A → B → C → D is hostile to crawlers and users. Collapse to A → D.
• Re-audit after structural changes. Migrations, CMS upgrades, theme changes, plugin swaps, content imports — all of these break links by accident.
• Inventory third-party dependencies. Analytics tags, fonts, polyfills, chat widgets, OAuth iframes. When one of them is compromised at the source, you want a list to check.

Common ways this goes wrong

• One-and-done. A scan run in January and never repeated tells you about January. Pick a cadence.
• Internal-only checking. Outbound links are where the worst failures live. Don't skip them.
• No prioritization. Fixing 100 obscure archive links while a broken signup CTA goes unaddressed is a rookie mistake. Severity-rank first.
• Treating redirects as fine. A 301 chain works until one of the hops dies. Audit and collapse.
• Forgetting to verify the fix. Re-test after change. Trust nothing.

The short version

If your site is small and static, a free web tool is the right answer. If it's a real product with auth and dynamic content, monitoring is the right answer. The hand-rolled manual methods have their place in spot-checking, but as a primary tool they're rarely the right call past 2010.

Pick the one that matches your shape. Run it on a cadence. Triage by impact. Re-audit after changes. That's the whole playbook.

Try LinkSentry's continuous monitoring →

---

Related Posts:

• The Ultimate Guide to Broken Link Checkers
• Broken Link Checker vs Link Monitoring: What's the Difference?
• Best Broken Link Checkers: An In-Depth Comparison