301, 302, 307, 308: Which Redirect Should You Actually Use?

Last update : November 6, 2026

Your new page opens correctly, but that does not prove the redirect is right. You may be telling Google a permanent move is temporary, or sending a form submission to a destination that no longer receives the expected request.

The choice between 301, 302, 307, and 308 redirects starts with two questions: will the original address return, and must the redirected request keep its method and body?

For an ordinary permanent article move, a server-side 301 is a practical choice. A 308 also expresses permanence. For temporary routing, consider 302 or 307. The distinction between those pairs matters most when applications receive submissions rather than simple page visits.

Google Search Central’s “301, 302, 307, 308: Which redirect is the chosen one?” raises these issues. This guide uses Google’s written documentation and the HTTP standard to turn them into decisions you can check on your own site.

Choose the Redirect Around the Move, Not the Number

Write the requirement before configuring a rule.

“Send visitors somewhere else” leaves too much unanswered. A useful requirement says which resource moved, whether the move is permanent, what the replacement contains, and which application requests must still work.

For example: “Our old pricing guide has permanently moved to the updated pricing guide. Existing links should reach that equivalent page, and our navigation should use the new address.”

That gives an editor and developer the same destination and purpose. It also prevents a temporary workaround from quietly becoming the site’s long-term URL strategy.

Our broader guide to HTTP status codes and Googlebot covers response diagnosis. Here, the narrower task is selecting and verifying a redirect.

301 and 308: Permanent Redirects for a Lasting Move

Google’s redirect documentation groups 301 and 308 as permanent redirects. Both signal that the destination should become canonical. Neither code is an automatic ranking upgrade over the other.

Think of a permanent redirect as a forwarding address after you move premises. The old address remains useful for people with outdated directions; your new communications should use the actual location.

Good candidates include an article slug change, a domain migration, or retiring an overlapping page in favor of a relevant replacement.

Before implementing one, ask:

  • Does the destination satisfy the reason someone opened the old URL?
  • Is the move intended to last?
  • Does the destination contain the correct content?
  • Can you update the links you control to point there directly?

A correctly configured status cannot compensate for an irrelevant replacement. Redirecting an old product tutorial to an unrelated sales page may leave visitors unable to finish the task that brought them there.

For editorial consolidation, the useful work happens before the redirect: deciding what to retain and combining the material into a coherent resource. That is the foundation of merging similar blog posts.

302 and 307: Temporary Routing With a Different Intention

Google categorizes 302 and 307 as temporary redirects. They do not send the same canonical replacement signal as permanent redirects. Other signals can still affect which URL Google indexes, so “302 means the original URL always stays indexed” is too absolute.

Use temporary routing when the original address is expected to resume its role. Document the reason, an owner, and a review date. Otherwise, nobody may remember why the rule exists six months later.

Imagine a help resource temporarily moving to an equivalent location during a platform repair. The important question is whether the original location will return. A short-term detour and a permanent replacement are different requirements.

Do not assume every outage needs a redirect. If a resource is briefly unavailable and there is no useful substitute, temporary unavailability may be better represented by a 503 response. Sending every request to a generic maintenance page can also break application workflows.

The Technical Difference: Does a Submission Keep Its Method?

Reading a page usually involves a GET request. Sending a form often involves POST, with information in the request body.

Under RFC 9110’s redirect semantics, 301 and 302 allow POST to change to GET when followed. With automatic redirection, 307 and 308 preserve the request method. Their distinction is temporary versus permanent.

Think of a visitor carrying an envelope. Reaching the new desk is one requirement; delivering the envelope intact is another. A simple page-visit test checks only the first.

For a normal blog migration, most verification concerns GET requests and destination content. A form handler or API endpoint needs a different test: does the receiving application get the expected submission and process it correctly?

Do not change application redirects solely to standardize a status-code report. Ask the developer what each endpoint expects before modifying its behavior.

A 307 Redirect Does Not Guarantee a Successful Checkout

Preserving a request is only one part of a working transaction. The destination must still understand the payload, accept the relevant session, and handle the operation as intended.

An HTTP redirect is not proof that a payment completed, a lead entered the CRM, or an email was sent. Check the application outcome separately.

For a hypothetical enquiry form, an acceptance checklist might require:

  1. The initial submission reaches the intended handler.
  2. Any redirected submission arrives with the expected information.
  3. Exactly one test enquiry is recorded.
  4. The confirmation shown to the visitor matches the recorded result.
  5. Analytics records the intended event without duplicating it.

Run transactional tests in a controlled environment with test data. A live redirect checker is not a substitute for application verification.

Why Some Form Flows Use 303 Instead

Keeping POST is not always the goal. After processing a form, an application may direct the browser to retrieve a confirmation page. A 303 supports that change to retrieval rather than repeating the submission.

This matters because “307 preserves the data” can sound like a universal improvement. It is only appropriate when preserving the request matches the intended workflow.

Have the application owner distinguish a redirect before processing, which may need to carry the submission elsewhere, from one after processing, which may simply display a result. Those are different jobs.

Client-Side Redirects: A Fallback With Additional Dependencies

Google recommends server-side redirects where possible. Its documentation also recognizes meta refresh: immediate refreshes are interpreted as permanent, delayed ones as temporary. JavaScript redirects depend on rendering and may be missed if it fails.

For a migration, the practical concern is how many conditions must succeed before a visitor reaches the destination. A redirect hidden inside a failing script adds another troubleshooting dependency.

If you cannot configure the server, investigate the platform’s supported alternatives before accepting a JavaScript-only solution. Test the actual result rather than assuming that a visible browser jump proves every system sees the same move.

Check the Whole Journey, Not Just the First Response

A redirect can look correct at the start and still end at a broken page. The technical SEO fundamentals apply to the destination as much as the original URL.

1. Record the Intended Result

For each important old URL, save the chosen destination, reason for the move, expected status, and whether application requests are involved.

A useful record might say: “Old delivery guide to the updated delivery guide; permanent; GET page; new address should be used in navigation.”

2. Inspect a Real GET Request

For a public page, this command displays the response headers from a GET request without following the redirect:

curl -sS -D - -o /dev/null https://example.com/old-page

Check the status and Location destination. Using GET here avoids assuming that a server handles HEAD requests identically.

To follow the path and display the response headers along it:

curl -sS -L --max-redirs 10 -D - -o /dev/null https://example.com/old-page

The limit in this example bounds your diagnostic command. It is not a statement about Google’s crawling limits.

3. Read the Destination

Open the final page and check the title, main content, and intended action. A final 200 response can still contain the wrong template or an error message.

For an article, verify that the relevant answer survived the migration. For a product, verify that visitors reach a meaningful equivalent rather than an unrelated category.

4. Update the Signals You Control

Google’s canonicalization guidance explains that redirects, canonical declarations, and sitemap inclusion contribute signals of different strengths. Keep those signals consistent with the intended destination.

Update navigation and contextual links rather than making every visitor travel through the old address. Review sitemap entries and canonical declarations as part of the same change.

If you are unsure which version should remain primary, resolve the canonical URL decision before building a large redirect map.

5. Monitor Search and User Outcomes Separately

Use Search Console to inspect important URLs and observe changes in indexing and performance. Use analytics and application checks for leads, purchases, or other user outcomes.

A successful redirect request proves routing at the time of the test. It does not establish that Google has completed reprocessing or that the destination will inherit a particular ranking.

Three Mistakes That the Correct Status Code Cannot Fix

A redirect with no relevant replacement. If an old page has genuinely gone, decide whether another page meets the same need. Do not force a homepage destination merely to remove an error from a report.

A migration that changes too many things at once. If the URL, content, navigation, and offer all change, a later performance shift becomes harder to diagnose. Record those changes so the team can investigate more than the redirect code.

A rule nobody owns. Temporary rules need review; permanent mappings need maintenance. Keep a clear record of why the destination was chosen, especially when another redesign could remove it.

Existing external links also deserve attention. When recovering lost backlinks, check what a reader now encounters at the linked address. A functioning jump to an irrelevant page does not restore the original experience.

Frequently Asked Questions

Is 301 better for SEO than 308?

Google recognizes both as permanent redirects. Select the behavior your platform and application need, then verify the destination and surrounding signals. Changing a working permanent redirect merely because another code sounds more familiar is not a growth strategy.

Is 302 bad for SEO?

No. It is appropriate for a temporary move. The problem is using it without considering whether the original address should return, then assuming it communicates the same intention as a permanent replacement.

What is the difference between 302 and 307?

Both represent temporary redirection. The key application distinction is request-method handling: 307 preserves it during automatic redirection, while 302 permits a POST-to-GET change.

Should every form redirect use 307 or 308?

No. Some flows intentionally retrieve a confirmation page after processing a submission. Choose the status around the workflow, not a blanket rule that preserving POST is always better.

Can Google process JavaScript redirects?

It can, but rendering introduces dependencies. Prefer an appropriate server-side redirect where available and test the destination. Do not assume client-side redirection is invisible to Google or equally reliable in every implementation.

Does a permanent redirect guarantee the same rankings?

No. Relevance, content changes, technical access, and other signals still matter. Verify routing first, then monitor indexing and performance without treating the response code as a ranking guarantee.

Make the Redirect Decision Testable

There is no single chosen code for every move. Define what changed, choose permanent or temporary behavior, and establish whether a submission must continue or a confirmation should be retrieved. Then test the complete journey.

If you want a second perspective on how redirects, technical fixes, and content maintenance fit into a monthly SEO workflow, connect with Paul Gautheron on LinkedIn. Scale Xpert brings that work together so progress can be reviewed against real site outcomes.

Connect With SEO Professionals and Build Powerful Backlinks

Join Now

Find the right backlink partners and SEO opportunities to grow your website authority

Trusted by SEO professionals

seo growth

4.8 based on 90+ reviews