A 301 redirect is a server response that says a web address has moved for good. When a browser, search crawler or ad crawler requests the old URL, the server answers with the status code 301 Moved Permanently and a Location header naming the new address, and the client requests that address instead. It exists because the web runs on links that a page's owner does not control. Bookmarks, backlinks, printed QR codes and years-old ad campaigns keep pointing at URLs long after a site is reorganised, rebranded or moved to HTTPS. The 301 keeps those references working and tells machines which address replaces the old one.

How the response works

The exchange takes two round trips. A client requests http://example.com/shoes. The server replies with a short message carrying no page content:

HTTP/1.1 301 Moved Permanently
Location: https://www.example.com/shoes

The browser then requests the Location value and renders what comes back, ideally a 200 OK page. Each hop adds a network round trip and, if the new address sits on another host, a fresh DNS lookup and TLS handshake.

The current definition sits in RFC 9110, the Internet Engineering Task Force (IETF) specification of HTTP semantics published in June 2022. A 301 means "the target resource has been assigned a new permanent URI and any future references to this resource ought to use one of the enclosed URIs", and the server "SHOULD generate a Location header field". Since RFC 7231 in June 2014, that header may hold a relative path.

Two properties shape behaviour in practice. RFC 9110 calls a 301 heuristically cacheable, so browsers may store it without explicit instructions; according to http.dev, a reference site on status codes, a 301 without cache headers can persist in a browser indefinitely, and removing the rule on the server does not clear stored copies. The code also permits a method change: "For historical reasons, a user agent MAY change the request method from POST to GET." Browsers following the WHATWG Fetch Standard do exactly that.

The redirect family

MDN Web Docs, Mozilla's developer reference, sorts the main 3xx codes by permanence and method handling:

  • 301 Moved Permanently: permanent; POST may become GET.
  • 308 Permanent Redirect: permanent; method and body preserved.
  • 302 Found: temporary; POST may become GET.
  • 303 See Other: temporary; always switches to GET, typically after a form submission.
  • 307 Temporary Redirect: temporary; method and body preserved.

For ordinary page links, which are GET requests, 301 and 308 behave identically.

Where redirects are configured

Redirects are set wherever a request can be intercepted. In the Apache web server, the Redirect directive defaults to a temporary 302 unless "permanent" or "301" is added, a default that has caught out many migrations. In nginx, a rule such as return 301 https://www.example.com$request_uri carries the original path and query string across. Content delivery networks (CDNs) such as Cloudflare and Akamai answer with edge rules before a request reaches the origin, and content management systems including WordPress, through plugins, and Shopify maintain redirect tables for non-developers.

Typical uses are protocol consolidation (http to https), host consolidation (www and non-www), domain moves after a rebrand or acquisition, and URL restructuring. Two failure modes recur. A redirect chain forms when A points to B and B to C, often because HTTPS and www rules fire separately. A redirect loop forms when rules point back at each other, and browsers give up with an error.

Origin and evolution

The 301 predates formal standards. RFC 9110 notes that in HTTP/1.0, 301 and 302 "were originally defined as method-preserving ... to match their implementation at CERN", where Tim Berners-Lee built the first web server. RFC 1945, published in May 1996 by Berners-Lee, Roy Fielding and Henrik Frystyk, defined 301 as a resource "assigned a new permanent URL" and already recorded that some user agents "will erroneously change it into a GET request."

HTTP/1.1 arrived in RFC 2068 in January 1997 and was revised as RFC 2616 in June 1999, which renamed 302 "Found", added 307 and required an absolute URI in the Location header. Browsers never adopted the method-preserving reading. According to RFC 9110, "prevailing practice eventually converged on changing the method to GET", and the standard was changed to match.

The missing piece was a permanent, method-preserving code. Julian Reschke of greenbytes proposed 308 in experimental RFC 7238 in June 2014, and RFC 7538 made it a Proposed Standard in April 2015, because RFC 7231 "does not define a permanent variant of status code 307." RFC 9110 absorbed 308 in June 2022.

Why it matters for search and advertising

For search engines, a 301 is primarily a canonicalisation signal. Google's redirects documentation, last updated on April 14, 2026, groups 301, 308, instant meta refresh and JavaScript location changes as permanent, and says indexing treats the target of a permanent redirect as the canonical URL. 302, 303 and 307 are followed but carry no such weight.

Whether redirects leak value dominated SEO debate for years. In 2010, Search Engine Land reported some PageRank dilution through 3xx codes. Gary Illyes of Google settled it on July 26, 2016: "30x redirects don't lose PageRank anymore." Google's site move guide, updated on June 17, 2026, repeats that permanent redirects "don't cause a loss in PageRank", says Googlebot follows up to 10 hops and advises keeping redirects for at least a year. The update extended the Change of Address tool, introduced in Webmaster Tools on June 11, 2009, to every www and non-www variant.

Skipping redirects has a measurable cost. In a case reported in October 2025, a company changed the URLs of its top pages without 301 redirects and organic traffic fell 70% within two months, by the dismissed SEO professional's account.

Paid media is stricter. Google Ads disapproves ads under its destination mismatch policy for "redirects from the final URL that take the user to a different domain", and Google AdsBot requires fewer than ten redirects. On June 24, 2026, Google announced that cross-domain redirects would be allowed "in certain circumstances, with prior approval" from early July, according to Search Engine Roundtable, citing consumer goods makers sending shoppers to approved retailers. As of October 21, 2026, the English policy page still lists such redirects as violations without describing the exception. Google Merchant Center says not to "redirect people to another website outside of the domain you claimed."

Measurement is where redirects bite hardest. Third-party click trackers historically inserted their own hop between ad and landing page, which is why Google moved that tracking into a background request with parallel tracking, announced in October 2017. Any hop that drops the query string loses click identifiers and UTM parameters. Google's Tag Diagnostics began flagging redirects that strip the gad_ parameter on July 30, 2026, noting that a parameter can survive one hop and vanish at a later locale redirect. Short links, vanity URLs, QR codes and many deep links are redirects by design.

Limitations and disputes

Cache stickiness is the most common operational complaint. A mistaken 301 can strand returning visitors long after it is fixed, so many engineers deploy a 302 first and switch once a move is confirmed, accepting a weaker canonical signal meanwhile.

Signals are hints, not commands. John Mueller of Google said in April 2025 that redirects are one part of canonicalisation and that "just redirecting and leaving everything else alone" may not convince Google when internal links, sitemaps and canonical tags disagree. lintlab's October 2026 crawl found 23% of top-1,000 sites with a sitemap listed URLs that redirected, the most common fault it measured.

Audit effort is contested. SEO tools flag chains of two or three hops as errors, yet Mueller said in February 2026 that he could not recall redirect problems hurting SEO without also being visible to users.

Abuse is the other side. Google's spam policies define sneaky redirects as showing users and search engines different content, such as diverting mobile visitors to a spam domain, which can draw manual actions. Redirect chains are also a staple of malvertising, where a clean landing page hands off to a malicious one.

Not the same as

302 redirect. A temporary move; crawlers keep the original URL as canonical.

308 Permanent Redirect. Equivalent to 301 for search, but forbids changing POST to GET. Fabrice Canel of Microsoft confirmed in September 2024 that "Bing treats 308 redirects the same as 301 redirects."

Meta refresh and JavaScript redirects. These run in the browser after a page loads. Google treats an instant meta refresh as permanent and a delayed one as temporary, and recommends JavaScript redirects only as a last resort, since rendering can fail.

Canonical tag. A rel="canonical" link, or its HTTP header form, suggests a preferred URL while still serving the duplicate page. A 301 removes the old page from view.

URL rewrite. An internal rewrite maps a public address to a different file on the server; the address bar does not change and no 3xx is sent.

Recent developments

This month's change comes from browsers. Chrome 154, which Google's Chrome security team scheduled for October 2026 in a post on October 28, 2025, turns on "Always Use Secure Connections" for public sites by default: Chrome tries HTTPS first and asks permission before loading a public site without it. That alters the old HTTP-to-HTTPS 301. An http:// link on a printed QR code or legacy campaign only works smoothly if the same hostname answers on HTTPS, because the browser may warn before it ever sees the HTTP redirect, according to monitoring firm Semonto.

Timeline

  • Early 1990s: CERN's HTTP implementation uses 301 and 302 as method-preserving redirects.
  • May 1996: RFC 1945 (HTTP/1.0) defines 301 Moved Permanently and 302 Moved Temporarily.
  • January 1997: RFC 2068 publishes HTTP/1.1, adding 303 See Other.
  • June 1999: RFC 2616 renames 302 "Found", adds 307 and requires absolute Location URIs.
  • June 11, 2009: Google adds the Change of Address tool to Webmaster Tools.
  • 2010: Search Engine Land reports PageRank dilution through 3xx redirects.
  • June 2014: RFC 7231 permits relative Location values; RFC 7238 introduces 308 as experimental.
  • April 2015: RFC 7538 makes 308 Permanent Redirect a Proposed Standard.
  • July 26, 2016: Gary Illyes says 30x redirects no longer lose PageRank.
  • October 2017: Google announces parallel tracking for ad clicks.
  • June 2022: RFC 9110 becomes the current HTTP semantics specification.
  • September 6, 2024: Microsoft's Fabrice Canel confirms Bing treats 308 like 301.
  • October 28, 2025: Chrome announces HTTPS-by-default for Chrome 154.
  • April 14, 2026: Google last updates its redirects documentation.
  • June 17, 2026: Google's site move guide extends Change of Address to www and non-www variants.
  • June 24, 2026: Google announces approval-based cross-domain redirect exception in Google Ads, effective early July.
  • July 30, 2026: Google Tag Diagnostics begins flagging redirects that strip gad_ parameters.
  • October 2026: Chrome 154 enables Always Use Secure Connections for public sites by default.

Summary

Who: Web servers, CDNs and content management systems issue 301 responses; browsers, search crawlers such as Googlebot and Bingbot, and ad crawlers such as Google AdsBot follow them. The IETF defines the code, and Google, Microsoft and ad platforms set rules for how they treat it.

What: A 301 Moved Permanently is an HTTP status code telling a client that a resource has a new permanent address, given in the Location header. Search engines treat it as a strong signal to index the new URL and consolidate ranking signals there.

When: It dates from CERN's early web software, was standardised in RFC 1945 in May 1996, revised through RFC 2616 (1999) and RFC 7231 (2014), and is currently defined in RFC 9110, published in June 2022. Google said in July 2016 that 3xx redirects no longer lose PageRank.

Where: In the HTTP response, configured at the web server, CDN edge or CMS, and encountered on site migrations, HTTPS upgrades, short links, QR codes, ad final URLs and product feeds.

Why: Links outlive the pages they point to. The 301 keeps old references working, tells search engines which address replaces the old one, and, for advertisers, determines whether clicks, tracking parameters and policy checks survive the journey to the landing page.