Website 404 Error and Redirect Audit: Find and Fix Broken Links Before They Cost Traffic
404-errorsredirectsbroken-linkstechnical-seowebsite-maintenanceurl-management

Website 404 Error and Redirect Audit: Find and Fix Broken Links Before They Cost Traffic

MMegastorage Editorial Team
2026-11-21
8 min read

Learn how to find 404 errors, prioritize broken URLs, map 301 redirects, and verify fixes without chains or redirect loops.

A 404 audit helps you find broken URLs, decide which pages need redirects, and verify that repairs work without creating chains or loops. This guide provides a repeatable workflow, a redirect-mapping template, prioritization method, and practical checks for redesigns, migrations, permalink changes, and routine website maintenance.

Overview

A 404 Not Found response means the server cannot find the requested resource at that URL. Some 404s are harmless: a mistyped address or a permanently removed page may not need a replacement. Others indicate a preventable problem, such as an internal link pointing to an old permalink, a migration that omitted content, or a server configuration error.

The objective of a website redirect audit is not to eliminate every 404. It is to separate expected missing pages from broken user journeys and then apply the simplest correct fix. Usually, that means updating an internal link, restoring a page, or adding a redirect to the closest relevant destination.

Run the audit after a redesign, domain or hosting migration, CMS change, permalink update, product removal, or major content consolidation. A smaller crawl can also be part of a monthly technical SEO review. For related checks, see the website uptime monitoring setup guide and the website-down troubleshooting checklist.

404 and redirect reference table

IssueWhat it usually meansRecommended response
404 errorThe requested URL has no available resource.Fix the link, restore the page, redirect to a genuinely equivalent page, or leave the 404 if no useful replacement exists.
Deleted pageContent was intentionally removed.Use a 301 only when a relevant replacement exists. Otherwise, provide a useful 404 or 410 response according to your platform’s capabilities and editorial policy.
Changed URLA page moved because of a slug, category, or permalink change.Map the old URL to the new equivalent with a 301 and update internal links.
301 redirectA permanent URL change.Use for a lasting migration or consolidation. Point directly to the final destination.
302 redirectA temporary URL change or test.Use only when the original URL is expected to return or the change is genuinely temporary.
Soft 404The page may return a successful response while displaying “not found” content or no meaningful content.Return an appropriate not-found response or provide a real page with useful, matching content.
Redirect chainOne redirect leads to another before reaching the destination.Update the source rule so it points directly to the final URL.
Redirect loopRules send the browser between two or more URLs indefinitely.Inspect overlapping CMS, server, proxy, and HTTPS rules; remove the conflicting rule and retest.
Server or CMS configuration errorRouting, rewrite, permissions, or deployment settings prevent a valid page from resolving.Check application routes, rewrite files, canonical host settings, and recent deployments before adding redirects.

How to estimate

For a practical audit, estimate the repair workload and potential user impact rather than treating every URL equally. Use this simple prioritization score:

Priority score = affected visits or links × business importance × fix confidence

You can rate each factor from 1 to 3. A URL with regular referral traffic, several internal links, or a critical conversion role should receive more attention than an obscure, unlinked address. If traffic data is unavailable, use internal-link count, external-link evidence, and the page’s role in navigation as substitutes. Record the assumptions so the result can be recalculated later.

Begin with four discovery sources:

  1. Internal crawl: Crawl the site as an authenticated or unauthenticated visitor where appropriate. Export URLs returning 4xx or 5xx responses, redirect targets, source pages, and link depth.
  2. Analytics or access logs: Look for requested URLs that receive visits, especially URLs linked from campaigns, bookmarks, or external sites. Treat log and analytics data as directional because collection settings vary.
  3. Search and webmaster tools: Review reported not-found URLs and inspect representative examples. These reports may include old or externally linked URLs that your current crawl does not discover.
  4. Migration inventories: Compare the old sitemap, CMS export, database URL list, or staging crawl with the new site. This is particularly important when a domain, hosting environment, or permalink structure changed.

After collecting the data, normalize URLs before comparing them. Decide how to handle uppercase letters, trailing slashes, URL parameters, HTTP versus HTTPS, alternate hostnames, encoded characters, and duplicate query-string variants. A redirect audit can appear much larger than it is if the same URL is counted in several forms.

Reusable audit checklist

  • Take a backup or export of current redirect rules before editing them.
  • Crawl important templates, navigation, XML sitemaps, and internal search results.
  • Export every broken URL with its source page and response status.
  • Separate true 404s from server errors, blocked requests, soft 404s, and redirect errors.
  • Check whether each broken URL has traffic, backlinks, internal links, or conversion importance.
  • Choose one destination only when it closely matches the old page’s intent.
  • Update internal links so users do not pass through unnecessary redirects.
  • Implement rules in the correct layer: CMS, web server, reverse proxy, or hosting control panel.
  • Test direct requests, destination responses, HTTPS behavior, and alternate hostname behavior.
  • Recrawl the site and retain the before-and-after reports.

Inputs and assumptions

Use a spreadsheet or database with one row per old URL. The following redirect-mapping template is deliberately small enough for a manual review and structured enough to automate later:

Old URLObserved statusSource or evidenceTarget URLActionOwnerTest result
/old-guide404Internal link and crawl/updated-guide301 and update linksWeb adminPending
/removed-item404Legacy sitemapNoneKeep not-found responseContent ownerPending

For each proposed redirect, document why the target is relevant. Avoid sending unrelated pages to the homepage simply to remove a 404 report. That can create a poor user experience and may produce a soft 404 if the destination does not satisfy the original request.

Also record the implementation layer. In a CMS, redirect plugins or routing settings may control the rule. On managed or shared web hosting, a hosting control panel may provide a redirect interface. On a VPS or self-managed server, rules may live in the web server configuration or application. Do not add the same redirect independently in several layers without testing; duplicated rules are a common source of unexpected behavior.

Status-code verification

Test the old URL from a command-line client, browser developer tools, a crawler, or another HTTP inspection tool. Confirm that the old URL returns the intended redirect status, the Location value is correct, and the final destination returns a successful response. A 301 should not point to another 301 unless there is a temporary technical reason; revise the map to point directly to the final URL.

Check both secure and insecure variants where relevant, along with the preferred hostname. Review trailing-slash and case behavior, because a rule that works for one URL form may fail for another. Do not assume that a browser cache proves the rule works: use a fresh request and test more than one client or environment when the change is important.

Worked examples

An old article URL returns 404, and the site’s navigation still links to it. The new article has the same subject and substantially serves the same visitor intent. The correct sequence is to add a direct 301 from the old path to the new path, replace the internal links, update the sitemap if necessary, and recrawl both URLs. The estimate is one redirect rule plus the number of templates or content records containing the old link.

Example 2: A removed product with no replacement

A discontinued product URL receives occasional requests but there is no equivalent product. Redirecting it to a broad category may confuse visitors. First, check whether a useful replacement, successor, or explanatory page exists. If not, retain a clear not-found response with navigation to relevant alternatives. Record the decision in the map so a later audit does not repeatedly reopen the same case.

Example 3: A redirect chain after migration

An old HTTP URL redirects to HTTPS, then to a new hostname, then to a revised path. Although the final page loads, the map contains three transitions. Replace the old rule with a direct redirect to the final canonical URL where your server and platform allow it. Then test the original URL, each intermediate URL, and the final destination. If any request loops, inspect HTTPS enforcement, hostname normalization, CMS canonical settings, and proxy rules together.

Example 4: A soft 404 after a CMS change

A missing URL displays a branded “page not found” message but returns a successful response. Review the CMS route and server configuration so missing content produces the platform’s intended not-found status. Then check that legitimate pages still return successful responses and that the custom error template remains useful to visitors.

When to recalculate

Revisit the audit whenever the inputs change: after a domain transfer, hosting migration, redesign, CMS upgrade, permalink change, content consolidation, product catalog update, or server rewrite change. Run a focused crawl after deployment and a broader review once the new URLs have had time to receive normal traffic. Recalculate priorities when analytics, access logs, backlink information, internal-link structure, or business-critical page assignments change.

Keep a dated redirect map and compare each new crawl with the previous one. A successful repair should remain successful after cache, CDN, SSL, DNS, and application changes. For related infrastructure checks, review the CDN and Cache-Control guide, the DNS records guide, and the cloud storage backup maintenance plan.

Final action plan: export your current URLs, crawl the site, classify each response, score the affected URLs, create one direct target for each valid redirect, implement rules in one controlled layer, and verify every change with a fresh request. Keep unresolved cases in the map with an owner and next review date. That turns 404 troubleshooting from a one-time cleanup into a repeatable migration and website maintenance process.

Related Topics

#404-errors#redirects#broken-links#technical-seo#website-maintenance#url-management
M

Megastorage Editorial Team

Cloud Infrastructure and Website Operations Editors

Senior editor and content strategist. Writing about technology, design, and the future of digital media. Follow along for deep dives into the industry's moving parts.