A permanent 301 redirect passes ranking signals to the destination URL, and a temporary 302 does not, according to Googlers Gary Illyes and Cherry Prommawin. The two Googlers made the case on the company’s Search Off the Record podcast, as covered by Search Engine Journal on Thursday. For anyone running a migration, the choice of status code decides which URL carries the ranking history.
Prommawin opened with reassurance. “First of all, relax,” she said, adding that a page stays indexed whichever redirect a site picks. “It’s really just a matter of which URL will show up.” That framing matters: the question is not whether a page survives, but which address represents it.
Illyes separated crawling from consolidation. Asked whether Google cares if a 301 and a 302 both send users to the new page, he answered, “Google’s crawlers don’t care at all.” The difference appears later, in what he called “the real battlefield here, the deduplication process.”
Deduplication works in two steps. First, Google groups pages with similar main content, which Prommawin called clustering. Then the index picks one representative URL from each cluster, a step known as canonicalization. Prommawin said both redirect types influence clustering, but only the 301 has a canonicalization effect.
That split explains the signal handling. Illyes said that with a 301, “Google takes all the ranking signals of the old page and hands them to the new page.” A 302 produces the reverse outcome. In his words, Google “assumes you are going back eventually, keeps the old URL in the search results, and basically ignores the new URL.”
Search Engine Journal’s Roger Montti drew a wider conclusion from the discussion. He argued that Google’s clustering behavior undercuts the idea of content cannibalization, the belief that similar pages dilute signals until none rank. In his reading, what SEOs label cannibalization is usually another problem, such as thin or non-unique content. That is Montti’s interpretation, not a statement from the podcast guests.
The pair also covered the newer status codes. Illyes positioned the 308 as the newer counterpart to the 301, and the 307 as the newer counterpart to the 302. The main practical difference concerns POST data, the form submissions a browser sends to a server. He used a Black Friday checkout failover as the example: a 307 keeps the submitted data intact and forwards it to a backup payment processor.
On implementation, Google prefers redirects issued by the server to meta refreshes or JavaScript. Illyes called the server approach “a clean, instant instruction.” Client-side methods, he said, make the browser fetch the page, parse it, and wait for a script before moving the visitor along.
The discussion contains no measurement. No data showed how quickly signals move after a 301, or how long a lingering 302 keeps the old URL in results. Practitioners should treat the guidance as Google’s stated model, not a tested one.
The timing is useful, with the holiday season close. Teams planning URL changes before peak trading should audit their redirect rules for temporary codes left over from launches and tests, and convert any permanent move to a 301 or 308 before they depend on the new URLs to rank.
Search Engine Journal, “Google Explains How Redirects Affect Ranking Signals” by Roger Montti, published November 5, 2026.
