Should You Redirect All 404s to the Homepage? The Soft 404 Risk

Your site has piled up hundreds of 404 errors, and someone suggested a one-line fix: send them all to the homepage. It sounds tidy. However, Google explicitly advises against it. In this guide we explain why the habit to redirect 404 to homepage is risky, how to build a proper redirect map, and when a 410 status makes more sense.

This is not just our opinion. Google Search Central's guide to site moves with URL changes makes exactly this point. In other words, the homepage is not a backup for a missing page. It sends the wrong signal to both search engines and people.

There is one exception. If you genuinely merged several old pages into one new page, you can redirect those old URLs to the consolidated page. In short, the destination must be the true successor of the old page.

For that reason, the rest of this guide offers a method that makes the decision for each old URL easier. The method has four stages: inventory, mapping, testing and monitoring.

What should you do first if you already did it?

If the blanket redirect is already live, do not rip everything out in a hurry. Measure first, then fix in stages. The sequence below is the basic flow our team follows in client work.

1. Write down the full scope of the rule: which URLs, which status code, which destination.
2. Open the page indexing report in Search Console and check the soft 404 count.
3. Export the old URLs that still earn traffic or backlinks to a separate list.
4. Map that list to one-to-one destinations and leave the rest as 404 or 410.
5. Test the new rules on staging, then deploy them.

Follow this order and you avoid irreversible mistakes. Also, documenting each step makes it much easier to see what changed if something breaks later.

The most common panic mistake is swapping one catch-all rule for another. For example, sending everything to the blog index instead of the homepage only moves the same problem to a different URL. So stick to the principle of measuring first and mapping second.

What is a soft 404, and how does it relate to mass redirects?

A soft 404 happens when the server returns a success code like 200, yet the page looks like an error or an empty page. Google's HTTP status code documentation states that Search Console shows a soft 404 when the content suggests an error, an empty page or an error message.

With mass redirects the mechanism is slightly different. A user looks for an old product page and lands on the homepage. If Google does not treat that destination as a match for the old content, it may handle the redirect like a soft 404. As a result, the redirect does not pass the signals you hoped for.

We cover the technical side in our guide to finding and fixing soft 404 errors. Here the focus is the blanket redirect decision itself.

In addition, a soft 404 classification works as a warning sign. Google is telling you that it does not see a real counterpart for that URL. Seeing this in the report is a good moment to review your redirect strategy.

Why do so many sites redirect 404 to homepage?

There are a few understandable reasons. First, plugins and themes offer the option with one click. Second, red numbers in a report make owners nervous. Third, an old myth still circulates: 404 errors get your site penalized.

• A ready-made "send all 404s to the homepage" switch in a plugin.
• Time pressure and an incomplete URL list during a migration.
• Fear of losing link equity.
• Treating the 404 count like a health score.

That said, none of these reasons produces results. Google accepts that a page that no longer exists returns 404. So instead of trying to push the 404 count to zero, focus on sending the right URLs to the right places.

Another source of this perception is third-party audit tools. Some tools turn the 404 count into a health score. However, those scores are not Google's assessment, so read them in context.

Does a 404 error hurt your site?

Not by itself. According to Google's HTTP documentation, a 4xx code tells the processing system that the content does not exist, and crawl frequency for those URLs gradually decreases. In practice, a deleted page that returns 404 is expected behavior.

The problem starts with pages that should not return 404. For example, if the URL of a product still on sale breaks, that is a real loss. Therefore, do not lump every 404 into the same bucket.

To separate the cases, use our guide to finding unindexed pages in Search Console. That way you can tell real losses apart from routine cleanup URLs.

Also, large sites will always have some 404s. Products disappear, campaigns end and content gets merged. What matters is that these errors stay under control and valuable URLs do not get lost.

How do you build a correct redirect map?

A correct map answers one question for every old URL: what was the user most likely looking for when they opened it? If the answer is a clear page, send it there with a 301. If no clear match exists, let the page return 404 or 410.

1. Identify the topic and purpose of the old URL.
2. Search the new site for a page that serves the same purpose.
3. If no exact match exists, consider the closest subcategory.
4. If the subcategory is unrelated too, skip the redirect.
5. Record the decision in the mapping table.

Doing this by hand gets tiring. For that reason, our redirect mapping tool can turn old and new URL lists into a first draft based on similarity. You still make the final call.

One more tip for mapping: when choosing a destination, look at the page's real purpose rather than its title. For example, an old price list page may be closer to a quote page than to a product page on the new site. That way you answer user intent more accurately.

What should a redirect map table look like?

The table below is a sample map for an example scenario. We used example.com as the domain. Each row stands for one decision.

Notice that the homepage is not the destination in any row. Also, a category redirect appears only where the product truly has a counterpart in that category. As a result, every redirect gives a meaningful answer to what the user wanted.

We recommend keeping this table in a spreadsheet. Then add the decision date and the person who made it to each row. As a result, you can explain each decision easily even when someone on the team changes.

When should you use a 410 status instead?

A 410 status explicitly says the content was removed permanently. It fits campaign pages, expired events and content you deliberately retired.

However, Google's HTTP documentation does not describe a distinct processing difference between 404 and 410. Both belong to the 4xx class and signal that the content does not exist. So think of 410 as a clear label of intent, not as a speed hack.

• Use 410 when your removal decision is final.
• Keep 404 when you are not sure; it leaves you more room.
• For spam or accidental URLs, 404 is enough.
• For discontinued products, check for a close alternative first.

Stock-specific decisions are covered in our guide to 404, 410 and 301 for out-of-stock product pages.

How does a redirect 404 to homepage rule affect users?

People arrive looking for something specific. For example, they click a search result titled "red running shoes." Then they land on the homepage and have no idea what happened. The page shows no error, but what they wanted is not there.

This confusion usually ends with a bounce. Moreover, the visitor may assume the site is broken. A well-designed 404 page, on the other hand, states the situation honestly and offers search or related categories.

In short, an honest 404 is often a better experience than a confusing homepage redirect. Google's site move guide warns that such redirects can confuse users, and this is exactly that point.

You can even go a step further for user experience. Add a simple block to the 404 page that suggests a few related pages based on words in the requested URL. That way you guide people without any redirect.

How do you avoid blanket redirects during a site migration?

Blanket homepage redirects are born most often during migrations. The team races the clock and the old URL list stays incomplete. Then someone writes an easy catch-all rule for every unmapped URL.

The fix is a full URL inventory before the migration starts. Google's site move guide also stresses mapping old URLs to new ones and recommends permanent server-side redirects.

1. Collect old URLs from sitemaps, server logs and Search Console data.
2. Put URLs with traffic or backlinks on a priority list.
3. Map every priority URL one by one.
4. Do not write a generic fallback rule; let unmapped URLs return 404.

You will find every step in our website migration SEO checklist.

Timing matters too. The mapping table should be ready before the new site goes live. Redirects added after launch can make many URLs look broken during the first crawl period.

Which 404 URLs are worth saving?

Not every 404 URL carries the same value. Some earned traffic in the past, some earned links from other sites. Others are meaningless URLs produced by bots.

• URLs with external backlinks: they carry value, so map them first.
• Pages that used to earn organic traffic: people may still look for them.
• URLs still requested by internal links: fix the internal link first.
• Bot or spam URLs: leave them as 404 and move on.

The last group in particular causes confusion in Search Console. We covered it in our post on unknown spam URLs and 404s. In short, spend your energy on URLs that carry value.

When is redirecting to a category acceptable?

A category redirect looks like a smarter version of the homepage redirect. However, it carries the same risk on a smaller scale. If the user wants a specific product, a broad category may not satisfy them either.

It is acceptable in one situation: the product is gone, but similar products still appear in the same category. For example, if one jacket model was discontinued and the jackets category still offers similar models, the redirect makes sense.

On the other hand, skip the redirect if the product was a different type or the category was removed too. Put simply, the test is always the same: is there an answer close to what the user wanted?

So what if several categories fit? In that case pick the context where the product sold or was searched most often. If the choice still feels hard, that may be a sign the URL has no clear counterpart at all.

Is there a risk of redirect chains and loops?

Yes, especially when new rules pile up on old ones. For example, if URL A goes to B and B goes to C, you have a chain. A blanket homepage rule then quietly attaches to the end of those chains.

Therefore, after mapping, point every redirect straight at its final destination. Also check rule order; if a generic rule runs before the specific ones, the whole map goes to waste.

To find chains, use our redirect checker. We explain the theory in our guide to redirect chains.

Also account for protocol and domain changes. For example, if an old URL first jumps to the secure version and then to the new path, you get two hops. Combine the rules to bring that down to a single hop.

How long should you keep redirects in place?

Only set up a permanent redirect when you are sure you will not revert it. Google's redirects documentation recommends 301 or 308 for permanent server-side redirects.

Our general approach on duration is simple: keep redirects that carry value in place for a long time. Because external links can keep pointing at the old URL for years, removing them early wastes that value.

You will find the detailed answer in our post on how long 301 redirects should stay. The key point here is that keeping a bad blanket rule longer only stretches the problem.

How do you remove an existing blanket rule safely?

Deleting the blanket rule at once makes hundreds of URLs return 404 overnight. That is not technically wrong, but you might lose valuable URLs too. So move in stages.

1. Prepare one-to-one mappings for valuable URLs first.
2. Place these specific rules ahead of the blanket rule.
3. Then remove the blanket rule.
4. Confirm that the remaining URLs return 404 or 410.
5. Watch the Search Console reports for a few weeks.

This way valuable URLs stay protected while the rest move to an honest status code. If you would like us to handle this on your site and track the results, you can request a quote through our SEO consulting service.

You also need a fallback plan. Save the old rule file and note the date of the change. Then, if something unexpected happens, you can return to the previous state quickly.

Which Search Console signals should you watch?

After the fix, read several reports together. A single number can mislead, but the trend tells you whether you are heading the right way.

• Changes in the soft 404 count in the page indexing report.
• The type of URLs listed as Not found (404).
• URLs in the Page with redirect status.
• Impressions for old URLs in the performance report.

Still, these reports take time to update. Google only sees the new state once it recrawls the URLs. Consequently, expecting clear results within a few days is not realistic.

In addition, use the URL Inspection tool on a few sample URLs. It shows how Google last saw each one. Checking them one by one lets you confirm the general trend in the reports.

Example scenario: a blanket redirect on an online store

Example scenario: a clothing store at example.com switches platforms. The old system holds around two thousand product URLs. Under time pressure the team writes one rule that sends every unmapped URL to the homepage.

A few weeks later the soft 404 count in Search Console rises. Moreover, visitors arriving from old product URLs leave the homepage right away. This is exactly the picture Google warns about.

To fix it, the team first separates URLs with backlinks and traffic. Next, it maps them to product or category counterparts. The remaining URLs return 404 or 410. All numbers in this scenario are an example, not real client data.

The lesson is clear: time pressure is not a good reason to write a catch-all rule. Instead, map priority URLs first and leave the rest with an honest error code.

When do plugins that redirect 404 to homepage become dangerous?

Content management systems offer many redirect plugins. Most provide useful features; however, some include a setting that automatically sends every 404 to one destination.

When that setting stays on, every newly broken link also goes quietly to the homepage. In other words, it hides the problem but does not solve it. It also makes it harder to notice a real error later.

• Turn off automatic 404 redirects in the plugin settings.
• Manage redirects in one place; do not mix server and plugin rules.
• Use the plugin's 404 log for monitoring, not for redirecting.

Why should you fix internal links before redirecting?

If a link on your own site points at a broken URL, the real fix is to update that link. Redirects exist to catch traffic coming from outside.

For example, stale links in the menu or in old blog posts create an extra redirect hop on every visit. So linking internally straight to the new URL is cleaner for both crawling and users.

In summary, the order is this: fix internal links first, then map URLs that carry external value, and finally leave the rest with an honest error code.

Also review on-site search. Old product links in search results may point at broken URLs too. In other words, internal link cleanup is not limited to the menu.

Should you handle this yourself or get expert help?

On a small site you can map a few dozen URLs on your own. With a table, a tool and some patience, the job is manageable. On a store with thousands of URLs or a multilingual site, however, things get complicated.

At that point the biggest risk is writing another generic rule in a rush. Our team handles inventory, mapping, testing and monitoring together in such projects. We do not guarantee rankings, but we make sure every decision is reasoned and measurable.

If you want us to review how your site handles 404 errors and redirects, you can request a quote from our SEO consulting page.

At the end of the project we also hand over the mapping table, the test results and our monitoring notes. Then you know where to start the next time you change something.

Which data should support your redirect decisions?

Base each mapping decision on data, not gut feeling. You have three core sources: Search Console, server logs and your analytics tool. Each one answers a different question.

• Search Console shows which URLs Google sees and how it classifies them.
• Server logs show which URLs bots and users actually request.
• Analytics shows how many visits old URLs received in the past.
• Link data shows which URLs earn links from other sites.

When you combine these sources in one table, the priority order appears on its own. For example, a URL with both backlinks and traffic rises to the top. On the other hand, a URL with no value in any source does not deserve mapping time.

In short, data tells you where to spend effort. As a result, a list of thousands of rows can shrink to a few hundred manageable ones.

While merging these sources, standardize URL formats. Trailing slashes, uppercase letters or parameter differences can show the same URL on several rows. This keeps your table free of duplicates.

How do you test redirects before going live?

Pushing new rules straight to production is risky. One bad pattern can send thousands of URLs to the wrong place. That is why we recommend a small testing routine.

1. Pick a random sample from the mapping table.
2. Open each sample URL on staging and check the returned status code.
3. Confirm the destination returns 200 and shows the right content.
4. Check that an unmapped URL lands on the 404 page, not the homepage.
5. Confirm there are no chains or loops.

This checklist looks simple, but in practice it is the most skipped step. Also, run the same samples again after going live. Caching or server configuration can behave differently from staging.

So how many samples are enough? There is no fixed number, but picking a few from each redirect type is a good start. Specifically, test edge cases separately for pattern-based rules.

How should your 404 page guide visitors?

For URLs you do not redirect, a good 404 page is the most honest way to win back a lost visitor. The page must return the correct status code, meaning a real 404.

In addition, the page should explain what happened in plain language. A search box, popular categories and a clear link to the homepage make a good start. That way the visitor makes their own choice.

• Check that the page returns 404, not 200.
• Write a short, clear explanation.
• Add a search box and links to related categories.
• Do not auto-forward the page to another URL.

The last point matters most. A 404 page that bounces people to the homepage after a few seconds repeats the same problem in a different costume.

What changes on multilingual sites?

On multilingual sites, blanket redirect mistakes get even messier. For example, an old German URL that lands on the English homepage is wrong in both language and content.

Therefore, map each language separately. Every old URL should go to its counterpart in the same language. If none exists, it should hit that language's 404 page.

Also check the links between language versions. If a page was removed in one language but other versions still point to it, users and crawlers end up at the wrong address. So review all language versions together after a migration or cleanup.

Core guide: What Is Digital Marketing? Its types, examples and how it is done, on one page with an interactive film that travels through the brain.