What Redirects Are and How to Fix Chain Problems Easily
Learn what 301 and 302 redirects are, why redirect chains and loops damage your SEO, and how to find and fix them. Step by step for complete beginners.
What you will learn in this guide
- What a 301 redirect is and when to use one
- The difference between 301 and 302 redirects
- What a redirect chain is and why it hurts SEO
- What a redirect loop is and how to spot one
- How to check your redirect chains using the free tool
- How to fix a redirect chain in under 5 minutes
1 What is a redirect?
A redirect tells the browser and Google that a URL has moved. When someone visits the old URL, they are automatically sent to the new one. The most common type is a 301 redirect — which signals a permanent move and tells Google to index the new URL in place of the old one. It does pass ranking signals — and so does a 302: Google has said that no PageRank is lost through either. The real difference between them is which URL ends up in the index, not how much value survives the hop.
| Type | Meaning | When to use |
|---|---|---|
| 301 | Permanent redirect | Page has moved permanently — use for site migrations, URL changes |
| 302 | Temporary redirect | Page is temporarily unavailable. It does not "leak authority" — that is a myth. What it does is tell Google to keep the old URL indexed, which is wrong if the move is permanent |
| 307 | Temporary (HTTP/1.1) | Same as 302 — rarely needed for most sites |
2 What is a redirect chain?
A redirect chain is when one URL redirects to a second URL, which then redirects to a third URL. Hops do not lose ranking authority — Google has stated that no PageRank is lost through a 301 or a 302. What chains genuinely cost is latency for the user, wasted crawl fetches, and fragility: Google abandons a chain after roughly ten hops, and chains grow silently over years of site changes.
/old-page → /temp-page → /new-page/old-page → /new-page3 How to check your redirects
- 1Open the Redirect Chain CheckerGo to audit-tools.html#redirect-checker and enter any URL you want to check.
- 2Review the redirect chainThe tool follows every redirect hop and shows you each step — the URL, the status code, and the final destination. A chain with more than one hop should be fixed.
- 3Fix chains by updating to point directly to the final URLIn your CMS, .htaccess file or NGINX config, update the original redirect to point directly to the final destination, skipping all intermediate hops.
4 The myth that drives most redirect advice
Nearly every article about redirects tells you that each hop bleeds away a percentage of your ranking authority. It is not true, and correcting it changes what you should actually be doing.
Google has said plainly that no PageRank is lost through a 301 or a 302 redirect. The familiar "each hop loses 10–15%" figure describes an old model and does not describe current behaviour. Nothing is evaporating in your chains.
This matters because the myth sends people after the wrong thing. Teams audit their chains looking for leaked equity, find none they can measure, and conclude the work did not matter. Meanwhile the costs that are real go unaddressed — and they are all costs you can observe directly:
- Latency. Each hop is another round trip for a person on a mobile network, on a page they wanted instantly. This one your users feel.
- Crawl budget. Each hop is a fetch spent on a redirect rather than on a page you wanted indexed — acutely expensive right after a migration, when Googlebot has thousands of new URLs to work through.
- Fragility. Google abandons a chain after roughly ten hops. Chains grow silently across years of site changes until one crosses that line, and the destination becomes unreachable with nobody noticing.
5 301 versus 302: what the choice really decides
Since neither code loses value, what does the choice between them actually control? Which URL ends up in the index.
| 301 Permanent | 302 Temporary | |
|---|---|---|
| What you are telling Google | The move is permanent — index the destination instead | The original will be back — keep indexing the original |
| Passes ranking signals? | Yes | Yes — the "302 leaks authority" claim is a myth |
| The mistake it causes | Using it for a genuinely temporary move — the old URL drops out of the index and you wanted it kept | Using it for a permanent move — Google keeps indexing a URL you have abandoned |
The practical rule follows straightforwardly. Ask one question: is this URL coming back? If it is — a product out of stock for a fortnight, a page down for maintenance — 302, and Google keeps the original where it belongs. If it is not, 301, and Google moves on.
The common failure is a 302 left in place after the move turned out to be permanent, which happens because 302 is the default in a great deal of framework and server code. Check what your platform emits by default; a surprising number of "temporary" redirects on the web are temporary only in the sense that nobody chose them.
6 Loops: two correct rules in the wrong order
A redirect loop is two or more URLs pointing at each other in a circle. The browser gives up with an error, and Google indexes none of them — which makes it the most severe redirect problem and, thankfully, the most visible.
What is worth knowing is where loops come from, because it is almost never one obviously wrong rule. It is two perfectly reasonable rules, written months apart by different people, evaluated in an order nobody thought about:
- A rule forcing HTTPS, and a rule forcing a trailing slash — each undoing the other's rewrite.
- A rule stripping
www, and a CDN rule adding it back. - An application-level redirect and a server-level redirect that disagree about the canonical form of a URL.
So debug a loop by reading the rules in the order your server evaluates them, not by looking for the broken one. There usually is no broken one, and the fix is nearly always to consolidate the two rules into one that does both jobs in a single hop.
7 Two habits that prevent most of this
Point internal links at the destination. A redirect is a safety net for links you do not control — external ones, bookmarks, old emails, printed material. An internal link is one you control completely, and pointing it at a URL you know has moved means every visitor pays a round trip you could have saved them, and Googlebot spends a fetch on a redirect instead of a page. Keep the redirect for the outside world; stop relying on it for your own navigation.
Collapse chains when you create them, not later. When you add a new redirect for a URL that is already the target of an old one, update the old rule to point at the final destination as well. This takes thirty seconds at the time and is the difference between a site with a hundred single-hop redirects and one with a hundred chains that nobody wants to touch.
And when you delete a page, choose deliberately. A genuine successor gets a 301, in one hop. Content that is permanently gone with no equivalent gets a 410 and comes out of the sitemap — or a 301 to the most specific relevant category if that truly serves the visitor. What never works is redirecting everything to the homepage: it is not a relevant destination for somebody who wanted a specific page, Google frequently treats such redirects as soft 404s regardless, and the visitor is simply lost.
8 Frequently asked questions
No. Google has said that no PageRank is lost through a 301 or a 302 redirect, and the widely-quoted figure of ten or fifteen percent lost per hop describes an old model that does not reflect current behaviour. This matters because it changes what you are optimising for. Chains are still worth collapsing — for latency, for crawl efficiency, and because a chain that keeps growing will eventually break — but not because value is evaporating along the way. It is not.
Which URL ends up in the index. A 301 says the move is permanent, so Google should index the destination in place of the original. A 302 says the original will be back, so Google should keep the original indexed. Both pass ranking signals. The mistake that costs you is using a 302 for a permanent move — not because authority leaks, but because you have told Google to keep indexing a URL you have abandoned, and it will.
Three concrete reasons, none of them about equity. Latency: every hop is another round trip a person on a mobile connection waits through, on a page they wanted immediately. Crawl budget: every hop is a fetch that could have gone to a page you want indexed, which matters most right after a migration when Googlebot has thousands of new URLs to get through. And fragility: Google gives up after roughly ten hops, and chains grow across years of site changes until one day the destination is unreachable and nobody notices.
For as long as anything still points at the old URL, which for any URL with external links means indefinitely. Redirects are cheap to serve and expensive to remove: the day you delete them, every backlink pointing at an old URL starts returning 404, and you have discarded the value of a migration you paid for. The one thing worth doing over time is collapsing chains, because successive site changes turn a two-hop chain into a five-hop one without anybody deciding to.
Two or more URLs redirecting to each other in a circle, so nothing ever resolves. The browser gives up and shows an error; Google indexes neither page. It is usually caused by rules that were written independently and now contradict each other — a canonical HTTPS rule and a trailing-slash rule, for instance, each undoing the other's work. Read the rules in the order the server evaluates them, because a loop is almost always two correct rules in the wrong sequence rather than one wrong rule.
The destination, always. A redirect is a safety net for links you do not control — external ones, bookmarks, old emails. An internal link is one you do control, and pointing it at a URL you know has moved means every visitor who clicks it pays a round trip you could have saved them, and Googlebot spends a fetch on a redirect instead of a page. Keep the redirect in place; just stop relying on it for your own navigation.
It depends on whether a genuine successor exists. If it does, 301 to it in a single hop. If the content is permanently gone with no equivalent, return 410 Gone and remove the URL from your sitemap — or 301 to the most specific relevant category if that genuinely serves the visitor. What never works is redirecting everything to the homepage: it is not a relevant destination for somebody who wanted a specific page, and Google frequently treats such redirects as soft 404s anyway, so you gain nothing and confuse the reader.