usdigitalmarketingindex.com SEO
When should a removed URL return HTTP 410 instead of using a 301 redirect?
Updated November 15, 2026
When the content is gone for good and no page on the site does the same job. A 301 says the content has moved and names where. A 410 says the content has been removed and is not coming back. Which is true decides which to send. If a real successor exists, a page a visitor arriving for the old content would accept as its replacement, redirect to it. If none exists, an honest gone response is the accurate answer, and a redirect to something unrelated is a false statement that the old content lives there. On the difference between 410 and 404, Google’s documentation is plain: “All 4xx errors, except 429, are treated the same” (Google’s documentation on HTTP status codes). The choice of 410 is about saying what is true, not about a faster effect.
Truly gone content
Content is truly gone when all of these hold:
- it has been removed deliberately, not lost by accident;
- it will not be republished;
- no other page covers the same subject for the same purpose;
- visitors arriving for it cannot be helped by any single page on the site.
Examples:
| Removed content | Truly gone? |
|---|---|
| A product line the company has left entirely, with no successor | Yes |
| A past event's registration page, for an event that will not recur | Yes |
| A job posting that has been filled | Yes |
| Legal or regulatory content withdrawn as no longer correct, with no replacement | Yes |
| An expired promotion | Yes |
| User-generated content deleted by its author or for policy reasons | Yes |
| A blog post merged into a longer guide | No; the guide succeeds it |
| A page whose URL changed in a restructure | No; it moved |
| A discontinued product with a successor model | No; redirect to the successor |
What Google’s documentation says happens with a 4xx response: Google does not index URLs that return a 4xx status code, and URLs that are already indexed and return one are removed from the index. That applies to 404 and 410 alike. The documentation lists both under the same heading and gives them the same treatment.
So why choose 410 over 404? The HTTP standard distinguishes them in meaning: 404 says the server found nothing at the address and makes no statement about permanence, and 410 says the resource is intentionally and permanently gone. A 410 is the more exact statement when the removal is a decision. Other clients and tools may use the distinction. This guide makes no claim that Google removes a 410 URL sooner than a 404 one.
Both codes are answers from the server. A request that times out gives a crawler no answer at all, which is a different situation, described in the guide to how an HTTP 404 response differs from a network timeout for search crawlers.
Meaningful successor content
A redirect is right when a successor exists. The test: would a visitor who wanted the old page consider the new one an answer to the same need?
| Old page | Candidate destination | Meaningful successor? |
|---|---|---|
| Guide, 2022 edition | Guide, current edition | Yes |
| Product model A | Model B, which replaces it | Yes |
| Three short articles | One consolidated article that contains their content | Yes |
| Service no longer offered | A different service | No |
| Article on a specific topic | The blog index | No |
| Discontinued product, no replacement | Its category | Only if the category helps someone who wanted that item; otherwise no |
| Any page | The home page | No |
Google’s guidance on site moves addresses the last rows. It says not to redirect many old URLs to one irrelevant destination such as the home page, and that this “might be treated as a soft 404 error” (Google’s guidance on site moves with URL changes). It also confirms the consolidation case: where content from multiple pages has been combined on a new single page, the older URLs can redirect there.
Handling backlinks and user intent
Links from other sites are the argument made for redirecting everything. Look at them case by case.
| Situation | Decision |
|---|---|
| The removed page has links, and a true successor exists | 301 to the successor |
| The removed page has links, and no successor exists | 410; a redirect to an unrelated page does not serve the people following those links |
| The removed page has links because it was a reference others cited | Consider keeping the page, marked as archived, instead of removing it |
| The links point to content removed for being wrong or harmful | 410 |
Scenarios, with the reasoning:
| Scenario | Response | Why |
|---|---|---|
| A retailer stops selling a brand entirely; 40 product pages | 410 for each | No equivalents; the category would mislead |
| A model is replaced by its successor | 301 to the successor's page | Same need, current answer |
| A 2019 conference page, for a one-off event | Keep as an archive if it holds outcomes or is cited; 410 if it has neither | Nothing replaces it |
| An annual conference's 2025 page, with a 2026 page live | 301 to the event's current page, or keep 2025 as an archive linking forward | The event continues |
| A blog post retracted as inaccurate | 410, with a retraction note on a live page if readers need to know | Redirecting would hide the retraction |
| All of the above sent to the home page | Poor in every case | The home page answers none of those needs; visitors land confused, and the redirect may be treated as a soft 404 |
For visitors, the body of the error response matters. A 410 can carry a helpful page: a plain statement that the content has been removed, a search box, and links to the main sections. The status code tells machines the truth, and the page helps people.
Testing correct status
| Test | Method | Pass condition |
|---|---|---|
| Status code | Request the URL with a command-line client | 410, or 301, as decided |
| Not a soft error | Check that a removed URL does not return 200 with a "not found" message | True error status |
| Redirect target | For 301s, request the destination | 200; relevant content; one hop |
| Variants | Request with and without trailing slash, http and https | Consistent result |
| Error page content | Load a 410 URL in a browser | Helpful page, with the 410 status preserved |
| Internal links | Crawl the site for links to removed URLs | None |
| Sitemap | Search for removed URLs | Absent |
| Search Console | Page indexing report over time | Removed URLs listed as not found; no unexpected soft 404 entries |
| Caching | Request through the CDN | Same status as origin |
Implementation notes:
- A CMS may answer 404 for a deleted page unless told otherwise. Sending 410 takes a rule for the specific URLs or patterns.
- Keep a list of intentionally removed URLs, with dates and reasons. It distinguishes decisions from accidents when not-found errors are reviewed later.
- A 410 is a statement of permanence. If the content might return, use 404, or keep the page.
- For removals that must take effect in search at once, such as content published in error, Search Console’s removals tool hides a URL temporarily while the status code does the permanent work.
Decision summary:
| Condition | Response |
|---|---|
| Moved or replaced by a true equivalent | 301 |
| Merged into another page | 301 to that page |
| Removed permanently, no equivalent | 410 |
| Missing, status uncertain or possibly temporary | 404 |
| Still useful as a record | Keep, marked as archived |
A separate fault, a page that reports success while showing an error or empty content, is examined in the guide to why a page returning HTTP 200 can be treated by Google as a soft 404.
The criterion: redirect when you can honestly say “it is now here”, and return 410 when the honest statement is “it is gone”.
Add your comment to this answer