A 301 signals a permanent move; a 302 signals a temporary one. The practical difference is what search engines do with the original URL: a 301 tells them to transfer everything to the destination and forget the source, a 302 tells them to keep the source and treat the destination as a detour.
The short answer to the common worry: a correctly implemented 301 does not lose rankings in any meaningful sense. What loses rankings is redirecting to the wrong place, or chaining redirects, or using the wrong type.
Which to use
Use a 301 when the move is permanent
- The URL structure changed during a redesign or migration.
- Two pages were consolidated into one.
- A product was discontinued and a replacement exists.
- You moved to HTTPS or changed www convention.
- A page was renamed and the old slug is retired.
Use a 302 when the move is genuinely temporary
- A page is offline for maintenance and returning.
- You are running a short campaign at a temporary destination.
- A product is out of stock and you are diverting briefly, though keeping the page is usually better. See out-of-stock products.
- You are A/B testing and want the original URL retained.
The test is simple: will the original URL come back? If no, use a 301. If you are not sure, it is probably a 301.
What actually costs you rankings
Redirect type is rarely the real problem. These are.
- Redirecting to an irrelevant page. Sending everything to the homepage is treated as a soft 404 and discards the signal entirely.
- Redirect chains. Each hop adds latency and dilutes. Old to interim to final should be old to final.
- Redirect loops. Nothing resolves and the page is lost outright.
- Leaving internal links pointing at old URLs, so every internal path runs through a redirect.
- Removing redirects too early. They should stay for at least a year, and indefinitely where the old URL still attracts links.
- Client-side redirects via JavaScript, which are slower to be honoured and less reliable.
The 302 mistake that matters
Using a 302 for a permanent move is the more damaging error, because it tells search engines to keep indexing the old URL. The result is that neither URL consolidates properly and the move never fully takes effect. This happens routinely because 302 is the default in some frameworks and server configurations, so it gets used unintentionally.
Check what your redirects actually return rather than what you intended. Tools report this in seconds, and the surprise is common.
Frequently asked questions
Do 301 redirects lose link equity?
Not in any way you should plan around. A correctly targeted 301 passes signals to the destination. Historical concerns about loss are outdated; what causes loss is redirecting somewhere irrelevant, not the redirect itself.
How long should redirects stay in place?
At least a year, and indefinitely where the old URL still receives traffic or has external links. Removing them early reintroduces exactly the loss you avoided at launch.
Is a chain of two redirects a problem?
It works, but flatten it anyway. Each hop is a request, and chains grow over successive changes until they become genuinely damaging. Point every redirect at its final destination.
Should I redirect or use a canonical?
Redirect where the original should no longer be reachable. Canonical where users still need the variant, such as a filtered view. See canonical vs redirect.
Consulting CTA
If your redirects are chained, misdirected, or you are planning a URL change, book an SEO consultation for a redirect review.