How does this broken link checker actually work?
Phase 1 fetches each crawled page server-side and parses the HTML with DOMDocument to extract every <a href>, internal and external, along with its anchor text. Phase 2 sends a HEAD request to each unique URL with an 8-second timeout. Any HTTP response 4xx or 5xx, network failure, or zero-response is flagged as a broken link. The tool runs entirely in your browser plus a stateless server-side fetch, no account required.
Do broken links hurt SEO rankings on Google?
Yes, in three measurable ways. First, broken internal links waste crawl budget and leave orphaned pages, so Googlebot revisits less of your useful content. Second, every 404 in your sitemap or internal navigation is a soft signal that the site is unmaintained, which is one of the Quality Rater Guideline checkpoints. Third, broken outbound links to authoritative sources weaken the topical authority signal Google uses for E-E-A-T evaluation. Sites with under 1% broken-link ratio rank better than sites with 5%+, holding all other factors equal. Run a broken link checker monthly to keep the ratio low.
How often should I check my website for broken links?
Monthly for sites that publish weekly. Quarterly for static sites that change rarely. Also run an on-demand scan immediately after these four events: a site migration (new domain or new hosting), a bulk content import, a major theme or page-builder change, and any third-party plugin uninstall that touched the content. Save each month CSV and diff against the previous month. New 404s appearing in the diff almost always trace back to a specific change you can revert or fix at the source.
What HTTP status codes does the broken link checker flag as broken?
Any response that is not a 200 (OK), 201 (created), 204 (no content), or 3xx (redirect that ultimately resolves to 200). Specifically flagged: 400 Bad Request, 401 Unauthorized, 403 Forbidden, 404 Not Found, 410 Gone, 451 Unavailable for Legal Reasons, 500 Internal Server Error, 502 Bad Gateway, 503 Service Unavailable, 504 Gateway Timeout. Network-level failures (DNS lookup failed, TCP reset, TLS handshake failed) are also reported as broken with the specific error message attached.
What is the page limit for a whole-site scan?
Whole-site mode is capped at 100 pages per scan to keep results fast and avoid surprising the target server with a sudden crawl burst. Most blogs and small business sites fit comfortably under that limit. For very large sites, run the tool on each section URL separately (for example, /blog/ first, then /products/, then /docs/), or use a desktop crawler like Screaming Frog for a single end-to-end audit. The CSV export merges across multiple scans cleanly.
Why are some external links flagged as broken even though they work in the browser?
Some sites block automated user agents, require cookies, or only respond to GET requests with a full browser fingerprint. Cloudflare-protected sites in particular often return 403 or 503 to plain HEAD requests even when a regular browser loads them fine. Treat external 403 and 503 results as suspect rather than definitive: open the URL in a real browser to verify. The internal-link results are reliable because your own server responds consistently regardless of user agent.
Is this better than the Broken Link Checker WordPress plugin?
They solve different problems. The WP plugin runs continuously inside your admin via WP-Cron, which monitors over time but adds load to every cron tick and can slow large sites. The Pixellize tool runs on demand from outside your site, with no impact on your hosting and no plugin to maintain. For most agencies and freelancers running audits across many client sites, an external tool is more practical. For solo bloggers who want passive monitoring of their own site, the WordPress plugin is a better fit. Use both for the strongest coverage.
What is the difference between a broken link and a 404 error?
A 404 is one type of broken link, the most common one. The full set of broken links includes 4xx client errors (404 Not Found, 403 Forbidden, 410 Gone), 5xx server errors (500 Internal Server Error, 502 Bad Gateway, 503 Service Unavailable), and network-level failures (DNS resolution failed, connection timed out, TLS handshake failed). A complete broken link checker reports all of these, not just 404s. The Pixellize tool flags every category with the specific status code or error message attached, so you can prioritize fixes correctly.
How do I fix the broken links the tool finds?
Three patterns cover 95 percent of cases. First, an internal link to a moved or deleted page: update the link or add a 301 redirect from the old URL to the new one. Second, an internal link with a typo: fix the typo in the source post content. Third, an external link to a site that went down: either replace the link with an alternative authoritative source or remove the link and rephrase the sentence. Process the CSV in priority order, top-traffic pages first. Search Console organic landings make a useful priority list.
Is my data stored when I use the broken link checker?
No. The crawl runs entirely server-side per request, no database write, no logging of URLs you submit, no account required. The CSV and JSON exports are generated client-side from the result data and saved straight to your downloads folder. Your scan history is not persisted anywhere on our servers, so each scan starts clean. The tool is safe to use on internal staging URLs, password-protected sites you have access to, and other sensitive contexts.
