A broken link checker can reveal URLs that no longer work, but a reliable maintenance process must also explain why a link failed, whether a redirect is correct, and which fixes deserve attention first. This workflow shows how to audit internal and external links, run a practical URL redirect test, record recurring changes, and turn link checks into a repeatable monthly or quarterly task.
Overview
Link health affects more than user experience. A broken internal link can interrupt navigation, prevent readers from reaching an important page, or weaken the path between related pieces of content. An incorrect redirect can send visitors to an irrelevant destination, create a redirect chain, or preserve an outdated URL without delivering the page users expect. Broken links in technical documentation, product guides, campaign pages, and support content can be especially costly because those pages are often used when someone is trying to complete a task.
The goal of a link audit is not to make every URL look perfect at all times. The goal is to maintain a clear, intentional route from the linking page to the destination. That requires three related checks:
- Availability: Does the destination respond?
- Routing: If the original URL redirects, does it reach the intended final URL efficiently?
- Relevance: Does the destination still match the context and promise of the link?
A redirect management tool may help teams review redirect rules at scale, while a broken link checker is useful for discovering references that need investigation. These tools support the process; they do not replace a decision about whether a link should be repaired, redirected, replaced, or removed.
What to track
Start with an inventory that is detailed enough to support action. For each flagged URL, record the source page, destination URL, link type, status observed, date checked, and recommended action. A simple spreadsheet can work for a small site; larger sites may send results into an issue tracker or link monitoring software.
1. Internal link status
Track links that point to pages on your own domain. Separate hard failures from softer problems. A 404 or similar not-found response usually requires a replacement, removal, or redirect decision. A server error may be temporary and should be checked again before a page is changed. A page that loads successfully but no longer contains the expected information is not technically broken, but it may still be a content-quality issue.
2. Redirect behavior
For every important old URL, record the complete path to the final destination. A useful URL redirect test should answer:
- Does the old URL return a redirect rather than an error?
- Does it reach the correct page?
- Does the destination resolve to the preferred protocol and hostname?
- Are there unnecessary intermediate redirects?
- Does the final page load successfully?
Check common URL variations where relevant, including HTTP and HTTPS, trailing-slash versions, uppercase or lowercase paths, and legacy subdomains. Do not assume that a redirect is correct simply because it eventually reaches a working page.
3. External link health
External URLs can change without notice. Record the destination domain and the page context so that a link can be replaced with a more suitable source if necessary. Automated checks can produce false positives because of rate limits, access controls, temporary outages, or pages that require a browser session. Treat an external failure as a review item until it has been confirmed.
4. Link context and priority
Not all broken links deserve the same response time. Give higher priority to links on high-traffic pages, conversion paths, navigation, product documentation, legal or support resources, and pages receiving substantial internal or external references. Also flag links used in email templates, QR codes, campaign materials, or printed assets. A destination behind a QR code may continue receiving scans long after the original campaign has ended, so review it with the same care as a normal public URL. For campaign links, keep naming and destination records consistent by following a documented UTM naming convention.
Cadence and checkpoints
A recurring schedule is more dependable than a one-time cleanup. The right cadence depends on site size, publishing frequency, release activity, and how often URLs are changed, but a practical baseline is a focused monthly check with a broader quarterly review.
Monthly check
- Run an internal link check across newly published and recently edited pages.
- Review errors on navigation, landing pages, documentation, and high-traffic content.
- Retest open issues to distinguish a temporary response failure from a persistent problem.
- Check redirects introduced during the previous month.
- Confirm that important campaign, short, and QR-linked destinations still resolve as intended.
Quarterly review
- Crawl the wider site, including older content and less frequently visited sections.
- Review redirect chains, loops, and rules that point multiple old URLs to an unsuitable destination.
- Compare the current URL inventory with the site’s content and product structure.
- Review external links on evergreen guides and documentation.
- Archive resolved issues while retaining the date, cause, and fix for future reference.
Run an additional check after a domain migration, redesign, information-architecture change, documentation move, CMS migration, or major product release. These events change URL relationships and should not wait for the next scheduled audit.
How to interpret changes
Look for patterns rather than treating every flagged URL as an isolated defect. A sudden increase in internal 404s after a template change may point to a publishing or routing problem. A group of failures under one directory may indicate a moved content section. A growing number of redirects can signal that links are being updated late, rather than being changed at the source.
Classify each issue into one of four actions:
- Repair: Correct the destination or source link when the intended page still exists.
- Redirect: Send an obsolete URL to the closest relevant replacement when users or external references still need the old address.
- Replace: Choose a new destination when the original resource has been retired or no longer supports the page’s claim.
- Remove: Delete the link when it no longer adds value and no replacement is appropriate.
Be cautious with broad redirects. Sending many unrelated old URLs to a homepage can hide the original problem without preserving useful context. A redirect should represent a reasonable destination, not merely a working response.
When a checker reports a failure, reproduce it with a browser and, where useful, a command-line or header inspection tool. Compare the response code, redirect location, final URL, and page content. For shortened links, first inspect the destination safely with an URL expansion and unshortening tool before deciding whether the link is broken or simply routed through a short-link service.
When to revisit
Use the audit record as a living maintenance queue. Revisit it monthly for active sites and at least quarterly for stable sites. Bring forward any item connected to a launch, migration, campaign, documentation release, or URL restructuring. If link failures rise sharply, shorten the interval until the underlying cause is understood.
Before each review, prepare a small checkpoint list:
- Run the internal link checker and export new failures.
- Retest unresolved URLs and remove temporary false positives.
- Run a redirect checker on changed, retired, and high-value URLs.
- Prioritize issues by user impact, page importance, and replacement clarity.
- Apply the fix, test the final destination, and record the date and reason.
- Review the report for patterns that suggest a template, CMS, or release-process problem.
Keep the process lightweight enough to repeat. A consistent record of source pages, destination URLs, response behavior, ownership, and next action is more useful than an occasional large cleanup with no follow-up. With a recurring broken link checker and redirect review, link maintenance becomes a normal part of SEO and website operations rather than an emergency response.