How to Handle 404s, Redirects, and Broken Links (Without Losing SEO Value)

A reader clicks a promising headline, waits for the page to load, and lands on a dead end. No answer, no context, just a blank sort of “nope”. They don’t complain, they just leave.

That’s the quiet damage of 404s, bad redirects, and broken links. They waste visits, waste crawl time, and chip away at trust one small frustration at a time. Worse, they can drain link value if strong pages get removed or redirected carelessly.

The fix doesn’t need drama or a big rebuild. You need a clear plan: find what’s broken, choose the right response (update, redirect, keep a true error), then put a light routine in place so it doesn’t keep happening.

Know what you are looking at: 404 vs 410 vs redirect

It’s tempting to treat every error like a leak and slap tape on it. But the goal isn’t “make the error go away”. The goal is to match the user’s intent.

If someone wanted an old guide, send them to the new guide. If the product is gone, don’t pretend it exists. If the URL is a typo, let it be a clean error so Google and humans can move on.

Here’s the simplest mental model:

If you want a deeper refresher on how these codes behave, Moz’s primer on HTTP status codes is a solid reference.

404 Not Found, when the page is missing but might return

A 404 simply means the server can’t find that URL. It doesn’t automatically mean your site is broken. A few 404s are normal because the web is messy.

Common causes include:

Typos and copy mistakes: One wrong character in a newsletter link can create a brand-new 404.
Old URLs: Posts get updated, slugs get changed, folders get renamed.
Removed pages: A product is retired, a page is pruned, a legal page is replaced.
Site migrations: A new CMS or a new structure goes live, but redirect mapping was incomplete.

The real issue is patterns. If Search Console shows a growing list of “Not found”, or your analytics shows key landing pages dropping, that’s when 404s stop being background noise and start costing you.

From the reader’s side, a 404 feels like arriving at a shop with the lights off. Most people won’t wait around for an explanation.

301, 302, and 410, which one to use and why

Redirects are helpful when they point people to the page they meant to reach. They are harmful when they are used to hide problems.

A simple guide:

Use a 301 when the page moved and won’t come back. This passes most of the value and signals a lasting change.
Use a 302 when it’s a temporary switch. Think short-lived campaign pages or a brief test.
Use a 410 when the page is gone and has no true replacement. It tells crawlers to stop expecting it.

Google’s public guidance has stayed consistent: don’t redirect all missing URLs to the homepage. That creates confusion and wastes crawling, because the crawler expects the missing content and gets dumped somewhere unrelated.

If you want a practical overview of redirect types and SEO impact, Moz’s guide to 301 vs 302 redirects lays it out in plain terms.

Find 404s and broken links fast (without guessing)

Broken links feel like the kind of problem you can spot by browsing, but that approach misses most of them. You need a clean list, then a way to sort it by impact.

Think of it like tidying a house. You don’t start by polishing a spoon. You start with the rooms people actually use.

Use Google Search Console and a site crawl to spot the real offenders

Start where Google tells you it hit a wall.

In Google Search Console, look for reports that surface “Not found” URLs. These are pages Google tried to crawl but couldn’t. That’s your first queue of real-world issues, not guesses.

Then crawl your site with a crawler tool. A crawl catches:

• Internal links pointing to 404s (your own pages linking to missing pages)
• Redirect chains (multiple hops)
• Canonical mistakes that point to missing URLs
• Broken image links and resources that slow rendering

Screaming Frog is a common choice and the free version is enough for small sites (it has a crawl limit). The point isn’t the tool, it’s the method: crawl your internal links like a search engine would.

Also separate problems into two buckets:

Internal broken links: You created these, fix them first.
External broken links: Other sites link to you, or you link out to them. It’s still your job to handle them, because your readers don’t care whose “fault” it is.

Prioritise what to fix first: traffic pages, backlink pages, and key journeys

A long error list can make you freeze. Don’t treat every issue as equal.

Use this order:

1. Pages that get visits: If people land there, fix it now. This is pure user experience.
2. Pages with valuable backlinks: If other sites link to a dead page, you’re losing trust and link equity.
3. Pages in key journeys: Home, category pages, sign-up, pricing, checkout, account pages, popular posts.

A good rule: fix the link at the source first. If one menu link is wrong, it can create thousands of broken clicks. Updating that internal link can remove the problem across the site in one move.

Redirect work matters too. If you ignore redirects after a restructure, you can lose the accumulated value of links built over years.

Fix it the right way: redirects that help people and protect SEO

Redirects are signposts. A good one points to the nearest real destination. A bad one points to the building next door and hopes nobody notices.

In 2026, best practice is still boring and still effective:

Keep redirects one-to-one where possible, send people to the closest match, avoid chains, and don’t dump missing pages onto the homepage unless it’s genuinely the best destination (it rarely is).

A simple redirect decision tree (move, merge, remove, or restore)

Use this set of rules when choosing what to do with a missing URL.

If the page moved: 301 redirect to the new URL that serves the same need.
Example: /guides/seo-basics becomes /learn/seo-basics.

If two pages merged: 301 redirect the weaker page to the stronger combined page.
Example: /pricing-uk and /pricing-eu merge into /pricing.

If the page is gone with no replacement: keep a true 404 or use 410.
Example: a time-limited event page from 2022 that no longer applies.

If the old URL is valuable (strong backlinks): consider restoring it or creating a close replacement, then redirect to that.
Example: an old “beginner’s guide” that many blogs still link to.

One warning that saves a lot of pain: irrelevant redirects confuse users. If a reader wanted a specific article and you send them to a category page that doesn’t mention it, they bounce. Google sees that behaviour too.

Moz has a useful discussion on whether 404s are always harmful in its post Are 404 Pages Always Bad for SEO?. The takeaway is simple: errors happen, but mishandling them is optional.

Avoid redirect chains and loops, they slow sites and weaken results

A redirect chain is when one redirect points to another redirect, and another.

Example:

/old-post → /older-post → /new-post

Each hop adds delay. It can also muddy tracking, confuse debugging, and waste crawl time.

A redirect loop is worse:

/page-a → /page-b → /page-a

That traps users and crawlers in a circle until the browser gives up.

Fixing chains is usually straightforward:

Point the first URL straight to the final URL.
Then update internal links so your site stops asking for the old URL in the first place.

If you only do one thing after a migration, make it this: remove chains. They are easy to create when you keep “stacking” changes over months.

Update internal links first, then redirect for everything else

Internal links are the cleanest win because you control them.

Redirects are still useful, but they are a safety net. Your own pages should link directly to the right URLs.

Mini checklist for where broken internal links hide:

• Navigation menus and footers
• Breadcrumbs
• In-article links (especially older posts)
• Related posts modules
• Category pages and tag hubs
• XML sitemap URLs

After you fix the links and redirects, re-crawl the site. You want confirmation, not hope. It’s like checking the latch after you close a gate.

Turn dead ends into detours: build a helpful 404 page

Even well-run sites have 404s. People mistype, old links live on, and bots request strange URLs. Your 404 page is your chance to be a calm guide, not a brick wall.

A good 404 page says: “That path is closed, but here are the open doors.”

If you want ideas for what search engines expect, Moz’s best practices for error pages is a quick watch and keeps the focus on clarity.

What a good 404 page includes (search, popular links, and a clear message)

Keep it simple and human. Your reader is already annoyed, so don’t make them read a speech.

Essentials that work:

A plain message: “Page not found” is fine.
A short apology: one line is enough.
A site search box: the fastest escape route.
Links to popular sections: your top categories, latest posts, or key topics.
A clear way back: link to home and one or two major hubs.
Normal header and footer: it reassures users they are still on your site.

Also think mobile-first. The 404 page should load fast, be readable, and not bury the search box below a giant image.

What to avoid on 404 pages (fake content, endless redirects, and blame)

Some “solutions” look tidy in reports but make the real experience worse.

Avoid:

Redirecting all 404s to the homepage: it’s rarely relevant and often flagged as poor practice.
Hiding the status code: returning a 200 “OK” for an error page creates a “soft 404” problem.
Blaming the user: nobody wants to be told off for clicking a link you published last year.
Thin fake pages: don’t generate empty pages just to avoid errors, unless they genuinely help users.

A clean error is better than a confusing lie.

Keep broken links from coming back: a simple maintenance routine

Broken links aren’t a one-time event. They return whenever you publish, prune, migrate, or rename URLs. The goal isn’t perfection, it’s keeping the problem small.

A short routine beats a yearly panic audit.

After every site change: redirect map, re-crawl, resubmit sitemap

Any time URLs change, treat it like moving house. Label the boxes before you unpack.

Post-change routine:

Keep a redirect map: old URL, new URL, status (301, 410, keep 404), and notes.
Test your top pages: home, top categories, top landing pages, and top conversion steps.
Re-crawl the site: confirm no fresh 404s and no redirect chains.
Update and submit your sitemap in Search Console if URLs changed.

Keep important redirects live for at least a year. Longer is often better if the old URLs still get visits or backlinks. Old links can live a long time in blog posts, PDFs, and chat threads.

Set a weekly or monthly link check, and track the source of each 404

Choose a rhythm that matches your publishing pace. For many sites, monthly is enough. For news or fast-moving content, weekly catches problems before they spread.

When you log issues, track four fields:

Also check referrers when it’s worth it. If a partner site links to a mistyped URL that sends real traffic, asking for a fix can pay off. If it’s a random scraper site, a redirect might be the simpler answer.

Keep an eye on campaigns too. Newsletter links, paid ads, and social bios can create a sudden spike of 404s if a URL was typed once and copied everywhere.

Conclusion

Broken links are normal. Ignoring them is optional. Keep a simple three-part system: spot issues (Search Console plus a crawl), choose the right fix (update internal links, use a clean 301, or keep a true 404 or 410), then build a light routine so problems don’t pile up. Run a quick scan today, fix the top 10 issues by impact, and set a reminder to check again next month. Your readers will feel the difference, even if they never tell you.