Glossary · Technical SEO

What is a 302 redirect?

A 302 redirect is a temporary HTTP redirect. It sends visitors to another URL while indicating that the original address is intended to remain relevant after the temporary change.

Updated

What does a 302 redirect mean?

A 302 redirect says that a resource is temporarily available at a different address. The original address remains the intended reference for future requests. Use this response when the detour matches the actual resource state, and verify that its destination provides a useful, working result under the conditions that trigger the rule.

  • The HTTP standard’s 302 definition, accessed October 8, 2026, describes temporary location and the Location response header.

    A client’s ability to follow that header does not establish why the business selected the temporary destination or how long the detour should remain active.

  • Visitors may simply see another page load.

    That interface hides the resource-state distinction from ordinary browsing. The developer needs to inspect the initial response to confirm whether the move is temporary or permanent. A final screenshot or address-bar observation cannot identify the old route’s response code reliably.

  • Google’s redirect guidance treats temporary redirects as weaker signals that their destinations should be canonical.

    That does not guarantee that the original address will always be indexed. Other page relationships and signals can affect which representation Google chooses.

  • The useful question is whether the original resource is expected to return to normal delivery.

    If the replacement is now its permanent home, review a 301 redirect instead. Choosing an accurate response expresses intent clearly; it does not provide an independent promise about traffic, rankings, or customer conversions.

Compare the concepts

Temporary redirects and request methods

Temporary routing and method preservation are separate decisions.

Temporary redirects and request methods
StatusIntended useMethod behavior
302 FoundTemporary alternate locationClients may change POST to GET
303 See OtherRetrieve another resource after the requestUses a retrieval request to the indicated resource
307 Temporary RedirectTemporary alternate locationPreserves the method when automatically redirected
301 Moved PermanentlyPermanent moveClients may change POST to GET
Choose the status according to the resource state and required request behavior.Conceptual illustration informed by RFC 9110 — 302 Found, 303 See Other and 307 Temporary Redirect.

How is a temporary move different from a permanent move?

A temporary move preserves the original address as the expected future reference, while a permanent move identifies a new lasting location. Decide from the resource’s intended lifecycle rather than the age of a configuration entry alone. A long-running temporary situation still needs investigation, and a newly deployed permanent replacement can justify permanence immediately.

How is a temporary move different from a permanent move?
Point to considerExplanation and application
Illustrative decision: a service guide is temporarily served from an alternate route while its normal publishing system is repaired.If the normal guide address is expected to resume, the temporary relationship is understandable. If the business permanently consolidates that guide into a replacement, the intended resource state changes.
The response code cannot establish that plan.Ask the content or application owner what will return, what the destination supplies, and what condition ends the detour. A rule described only as temporary because nobody reviewed it is not a confirmed resource decision. Clarify the underlying purpose before changing its semantics.
Temporary routing also differs from a missing resource.An unavailable offering with no justified destination may need an accurate 404 error. Sending it elsewhere temporarily does not create a meaningful replacement. The code should not be used as a holding area for unresolved content inventory decisions.

How should the initial response and final page be examined?

Examine the initial response before following its destination, then inspect the final page as its own request. This separates the temporary-routing evidence from destination quality. A functioning 302 can lead to a missing page, another detour, or an unrelated screen, so observing the first response alone cannot establish that the customer task succeeds.

Capture a full GET to the old address using browser network tools or an appropriate command-line request. The HTTP method definitions distinguish GET from HEAD. A header-only response may be useful, but it is not conclusive where the application treats document requests differently.

curl --request GET --dump-header temporary-response.txt \
  --output temporary-body.html https://example.com/service-guide/

This illustrative command saves the initial response without automatically following its Location value. The example hostname is syntax, not audited client evidence. Identify the status, resolve the destination, and request that destination separately. Preserve each step if the path continues through another redirect.

  • In browser tools, match the response to the main document.

    Assets can also redirect, and their statuses do not describe the page’s initial request. Keep the navigation log through the move when possible. Read the final content as a visitor rather than treating successful transport as evidence of relevant information.

  • Record request conditions that influence the result.

    Cookies, authentication, host, and locale can alter routing. A logged-in preview may bypass the public rule entirely. The public anonymous request is often the appropriate starting point for a search-facing page, with other intentional contexts tested separately.

What should the destination provide during the detour?

The destination should provide information or functionality that sensibly serves the original request during the temporary detour. It should work under the same relevant public conditions. A general page that loads successfully can still leave the visitor without the service explanation, selected location, or action that the original address promised.

  • Compare the requested task with the delivered task.

    A temporarily relocated pricing explanation should still provide accurate pricing context or clearly explain the limitation. A home page may offer navigation but omit that explanation. Do not call the move successful solely because the browser stopped displaying an error.

  • Check essential content and controls.

    A destination that requires an owner session is not publicly equivalent to the original public guide. A form can appear intact while submitting to an unavailable endpoint. Inspect the actual customer path that the detour is meant to preserve, including any required follow-up action.

  • A soft 404 concern can arise when a destination appears empty or merely states that the requested content is unavailable.

    Google’s HTTP status guidance distinguishes successful delivery from substantive content suitable for processing. A successful final status alone is insufficient.

  • Explain temporary limitations honestly in the destination where they affect the customer’s decision.

    Do not manufacture a full replacement with generic service prose that conceals missing information. The route should help the visitor understand what is available now and how to continue appropriately while the original resource remains temporarily relocated.

How do 302, 303, and 307 differ for submitted requests?

The next request can preserve the original method or retrieve a separate result, depending on the chosen response. Historical handling of 302 can change POST to GET. A 307 preserves the method, while See Other directs the client to retrieve another resource as an indirect response. Choose according to the application’s intended flow.

How do 302, 303, and 307 differ for submitted requests?
Point to considerExplanation and application
The HTTP standard documents these distinctions.They matter beyond SEO. A contact form, booking action, or upload endpoint can behave differently from an ordinary document visit even when both end at an address that loads successfully in a browser.
Illustrative form flow: an application accepts a submission and then directs the visitor to a separately retrievable confirmation.That is a different relationship from temporarily moving the endpoint that accepts the submission itself. The confirmation resource need not be an equivalent copy of the original form-processing endpoint.
The 302 definition allows historical POST-to-GET handling. 303 See Other directs retrieval of an indirect response through GET or HEAD. 307 Temporary Redirect preserves the request method.These differences matter when a submitted form should produce a separate confirmation rather than repeat the submission.
Do not select 307 automatically without assessing the destination.Preserving a method also preserves the need for an endpoint designed to accept it. A general confirmation page may expect a retrieval request rather than a repeated submission. The appropriate code follows the intended flow, not a universal preference for preserving every method.
Google’s redirect documentation discusses search-facing temporary redirect signals.That treatment does not remove the application-level method distinction. Test navigation and controlled submission paths separately, using test inputs and the application’s safe test arrangement rather than creating real customer actions for verification.

Can a 302 destination still become Google’s preferred address?

Google evaluates multiple page relationships, so the temporary response alone cannot determine which address becomes the selected representative. The response expresses a temporary move, but it cannot force a particular indexing representation by itself. Review the destination, equivalent pages, and current inspection evidence before assuming that a different chosen address proves a technical failure.

  • Google’s redirect guidance describes temporary redirects as weaker canonical signals.

    It also discusses other signals that can influence the destination’s representation. Avoid statements that 302 always preserves the source in search or prevents the target from being selected.

  • A canonical tag can identify an equivalent-page preference, while the redirect controls the visitor’s route.

    These mechanisms are related but distinct. Review whether metadata expresses the same intended relationship, rather than adding another instruction without identifying what the source and destination actually represent.

  • Current internal links may heavily favor the alternate destination after an extended change.

    Sitemap entries can also describe a different inventory from the temporary rule. That inconsistency deserves investigation. It does not justify pretending that one isolated signal must override everything else in Google’s processing.

How should temporary redirects interact with canonicalization?

Temporary redirects should fit the intended canonical relationship without introducing contradictory preferences casually. Determine whether the alternate address is an equivalent representation, a temporary delivery route, or a different resource entirely. Then review the source and destination signals appropriate to that relationship, rather than applying the same canonical template to every temporary destination.

  • A canonicalization review concerns equivalent representations.

    A login screen is not simply an equivalent copy of a public guide. A regional variant may differ in material content. Those relationships need their own interpretation before any preferred-page instruction is inserted into a shared template.

  • Inspect the destination’s metadata and response headers.

    An inherited exclusion can make a temporary public replacement unavailable for normal indexing. A canonical instruction can point to a page that no longer works. These are actual destination conditions, not assumptions that can be inferred from the temporary status at the source.

  • Avoid adding noindex simply because a temporary alternate URL exists.

    Exclusion and equivalent-page preference have different functions. If the destination is meant to be publicly discoverable, an accidental exclusion can conflict with that purpose. Establish whether the resource should appear in search before selecting an indexing instruction.

  • Retest metadata after the detour ends.

    Temporary templates and routing middleware can persist even when content returns. The recovered original should provide its intended information and correct instructions. Removing the redirect rule alone does not demonstrate that every related preference or error-state exclusion was also cleared.

What can go wrong with geographic or language-based routing?

Geographic or language-based routing can prevent a requester from reaching the version they intended if the rule relies entirely on inferred location or browser settings. A crawler’s request conditions may also differ from the business owner’s visit. Test the actual public paths and provide a usable way to select the appropriate version when the site supports alternatives.

  • Do not assume a visitor’s physical location establishes the service location they need.

    A customer can research work for another property. Browser language does not necessarily establish the desired commercial region. A redirect that repeatedly overrides an explicit version choice can make otherwise available information inaccessible to that visitor.

  • Check whether an explicit regional address remains directly accessible.

    If the rule sends every request to another region, inspect its exceptions and selection logic. The presence of correct regional content in the repository does not prove that anonymous requesters can receive it through the public delivery path.

  • Test with recorded conditions rather than changing several variables at once.

    Compare the host and path, request headers, session state, and routing output. Browser emulation alone may not reproduce edge geolocation. Describe what the test actually changed so its result is not overstated as complete evidence of all regional behavior.

  • If temporary localization rules remain necessary, verify that the destination accurately describes its applicable market.

    A working page with the wrong service area can mislead a customer despite correct transport. The technical success criterion should include usable access to intended information, not merely an address change that matches an inferred setting.

Why can cookies or authentication make tests disagree?

Session-specific routing can produce different destinations for requests that appear to use the same public address. A returning customer, authenticated owner, and fresh public visitor can receive different destinations. Identify those conditions before calling one result incorrect, and verify that each intentional branch provides the information or access behavior appropriate to its actual audience.

  • An owner session can bypass a temporary public landing screen.

    A saved preference can select an alternate region. A signed-in account can route to a private dashboard. These observations do not prove that an anonymous crawler receives the same result. Test the relevant public request separately from the owner’s convenient preview.

  • Capture the initial response in each relevant state.

    A cached client route can also hide a fresh server request. Use the browser’s network evidence to establish what was requested and returned. Do not infer server behavior merely because two tabs display different final pages.

  • Where the destination is private, distinguish an access flow from a public content move.

    Sending an anonymous visitor to login is not evidence that the original guide has an equivalent public replacement. The expected authentication boundary should be explicit, with public information and private functions reviewed according to their actual purposes.

  • Avoid conclusions based on an unidentified user-agent spoof.

    Search crawler identity and session conditions are separate questions. A copied user-agent string does not prove the requester is Google. If crawler-specific behavior is central to the diagnosis, use verified evidence and do not present one simulated browser request as an observed crawler visit.

When does a temporary redirect become a stale configuration?

A temporary redirect becomes stale when its triggering condition or original purpose no longer applies while the rule continues to route requests away. Determine whether the resource returned, became permanently relocated, or was removed. The elapsed time alone cannot settle that question, but an unowned rule with no remaining purpose needs an explicit resource decision.

  • Locate the rule’s actual producer.

    It may live in an edge configuration, hosting platform, application middleware, or content record. Editing the original page cannot remove a rule generated before that page runs. Identify the delivery layer instead of repeatedly changing content that visitors never receive at the source address.

  • Compare the source’s intended current state with the destination.

    If the original guide is restored, test whether it can be served directly after the rule is removed. If the alternate address is now permanent, review the move accordingly. If neither resource fulfills the intended task, solve the content problem before changing only the code.

  • Check overlapping conditions.

    A former temporary rule can remain beneath a new permanent mapping or normalization rule. The observed public sequence may therefore include both. Trace the complete redirect chain and identify which condition still sends the request along the obsolete branch.

How should short-lived alternate routes appear in site references?

Keep owned links consistent with the intended public resource plan instead of automatically advertising every alternate delivery address. A genuinely temporary detour can leave the original address as the stable reference. A lasting relocation calls for a different decision. Confirm the plan before changing navigation and inventory metadata broadly.

  • The HTTP standard says clients should continue using the original reference for a temporary move.

    That does not prescribe every CMS workflow. It provides the resource relationship against which the site’s current linking and publishing choices can be assessed.

  • Review which address navigation advertises and why.

    If staff have replaced every original reference with the alternate route, ask whether the move is still intended to end. The links may reveal a de facto permanent change, but they do not establish the correct response until the resource owner confirms that plan.

  • An XML sitemap should reflect the intended current indexing inventory.

    Do not submit an alternate route solely because the source returns a temporary response. Determine which resources are meant to be discoverable, whether they work, and how their equivalent relationships are expressed.

  • Check restoration when the detour ends.

    Current references, temporary navigation notices, and relevant metadata may need updates. Scope those changes to what actually differed. A site-wide replacement operation can miss context or alter unrelated routes, so verify the rendered links and surrounding explanation that customers use.

How can temporary redirects affect performance and availability?

The extra request before the final resource arrives can influence the loading experience and expose another delivery dependency. The cost depends on the actual delivery path, caching, network conditions, and destination behavior. Measure that path rather than claiming a universal delay, and distinguish the redirect’s contribution from the performance of the destination itself.

  • A same-origin detour and a move to another host can have different network requirements.

    Additional redirects can extend the sequence. The important evidence is the request waterfall for the actual navigation, not a generic statement that every 302 adds a fixed amount of time under all conditions.

  • The destination can be less reliable than the source’s normal delivery route.

    A temporary publishing workaround may depend on another platform or service. Test essential content and controls there. A correct response at the source does not protect the final page from an unavailable dependency or broken public access policy.

  • Do not use a redirect as a universal answer to an outage.

    A meaningful alternate service can justify a temporary move, while a temporary server failure needs an accurate availability response when no such replacement exists. Google’s HTTP guidance treats those states differently.

  • When evaluating a repair, compare the same entry address and relevant conditions before and after the change.

    Opening the final URL directly omits the detour and answers a different performance question. It can help isolate destination behavior, but it should not be presented as the complete customer navigation measurement.

How should the end of a temporary move be verified?

Verify the end of a temporary move by requesting the original public address and confirming its intended current response, substantive content, and usable customer path. Review any conditions that formerly triggered the detour. Removing a configuration entry is implementation evidence, but only delivered behavior demonstrates that the public request now reaches the intended resource.

  • Start with the initial full GET and then inspect the rendered document.

    Confirm that the original resource does not redirect through another lingering layer. Edge and application rules can overlap, so a successful local test does not necessarily establish what the public host currently delivers.

  • Check restored content and metadata.

    Temporary screens can leave old titles, exclusion instructions, or preferred-page references. The ordinary page should not inherit them merely because its main content returned. Review the output actually received by an anonymous visitor rather than assuming a template’s default values are active.

  • Google’s Page indexing documentation and URL Inspection guidance help separate historical observations from current delivery.

    Later processing may differ from an immediate live test. Retain that distinction instead of repeatedly changing an accurate current response to chase an older label.

Continue the public-page review

Use our website SEO checker for preliminary public-route observations. Check temporary responses and their destinations through the procedures above. Our technical SEO services connect confirmed routing inconsistencies with an appropriate repair.

Questions about 302 redirect

When should I use a 302 instead of a 301?

Use 302 when the change is temporary and the original URL should remain the intended reference. Use a permanent redirect for a genuine lasting move.

Google Search redirects ↗
Does a 302 preserve a POST request?

Not reliably: historical behavior permits clients to change POST to GET for 302. Use a method-preserving status when preserving the submitted request is required.

HTTP standard ↗
Can a temporary destination still be indexed?

Yes. Google follows temporary redirects, and other canonicalization signals can still cause the destination to be indexed.

Google Search redirects ↗
How do I remove a test redirect after the experiment?

Remove the temporary redirect and obsolete variation URLs or scripts once the experiment ends. Verify the original page and chosen experience on the public route.

Google Search guidance for website experiments ↗

Continue learning

Try a relevant tool

  • Website SEO checker →

    Start a public-response review, then test the temporary rule under its triggering conditions.

Sources

RFC 9110 — 302 Found, 303 See Other and 307 Temporary Redirect ↗Accessed October 8, 2026Redirects and Google Search | Google Search Central  |  Documentation  |  Google for Developers ↗Accessed October 8, 2026RFC 9110 HTTP Semantics: GET and HEAD ↗Accessed October 8, 2026How HTTP Status Codes Affect Google's Crawlers | Google Crawling Infrastructure  |  Crawling infrastructure  |  Google for Developers ↗Accessed October 8, 2026URL Inspection tool - Search Console Help ↗Accessed October 8, 2026Page indexing report - Search Console Help ↗Accessed October 8, 2026Google Search guidance for website experiments ↗Accessed October 8, 2026

Published . Definitions and examples link to their supporting sources. Our SEO methodology →

SEO · Content · Local · Web Design

Connect the website work to your business.

We assess the pages, search demand, and customer actions that matter to your business, then explain where to focus the work.