Free URL Redirect Checker: Every Hop, Not Just the Destination

Across the sites we monitor, 15.0% of all crawler requests were answered with a 3xx in July 2026 (1–21 July) — roughly one request in seven spent arriving somewhere else instead of reading a page. Chains of two or more hops turn up on 24.3% of the sites we audit (H1 2026), which makes them the tenth most common failure our audit engine finds.

That is the real cost, and it is not the extra 200 milliseconds a visitor waits. Every hop is a request a crawler makes and does not get a page for. On a large site with a stale redirect map, a meaningful share of the crawl allowance is spent re-learning routes that were settled years ago.

The second cost is diagnostic. A browser resolves a chain silently: you type an old URL, you land on a new page, and everything looks correct. The 302 in the middle that should have been a 301 is invisible unless something refuses to follow the chain for you.

There are five status codes a server uses to redirect, and this tool reports whichever one it actually received rather than flattening them into the word "redirect".

For a page that has moved for good, use a 301 — or a 308, which search engines also treat as permanent. For a page you genuinely intend to bring back, use a 302. If you want the difference in one sentence: 301 and canonicalization is about telling search engines which URL is the real one, and 302 explicitly says "don't".

The method column is the part almost every redirect checker throws away. Clients have historically rewritten a POST into a GET when following a 301 or 302; 307 and 308 forbid that. On a marketing page nobody notices. On a form endpoint or an API path, it is the whole bug.

This checker reads the HTTP status line, which means there is a category of redirect it will honestly report as "no redirect". A <meta http-equiv="refresh"> tag and a JavaScript location.replace() both run after the response has arrived: the server returns 200, the body is delivered, and only then does the browser move on.

So if a URL clearly sends people somewhere else but traces clean here, you have already found your answer — the redirect is in the page, not in the protocol. That matters because search engines treat the two very differently. An HTTP 301 is an instruction; a meta refresh is a hint that a crawler may or may not honour, and a JavaScript redirect only exists for a crawler that renders the page.

The fix is almost always the same: move it to the server. If you need to confirm what is happening first, the website crawl test runs a real headless render and shows the difference between the raw HTML and the rendered DOM.

The output is one numbered row per hop. Each row carries the raw status code, the URL that returned it, and — underneath — the Location header that URL handed off to, written exactly as the server wrote it. The line above the table counts the redirects it took to reach the end.

Read it in this order. First, the last row. It should be a 200. A chain ending in 404 or 410 is a redirect pointing at nothing, and it is worse than no redirect at all, because it consumes the equity of the old URL and delivers it nowhere.

Then the hop count. One redirect is correct. Two is tolerable and extremely common. Three or more means separate rules are stacking on top of one another and should be collapsed into a single hop from the original URL to the final destination.

Then the Location values. This is where the subtle failure lives, and the next section is about it.

Chains. Rule A sends /old to /new, and a later rule sends /new to /newer. Nobody wrote a chain; two people wrote one redirect each, months apart. Fix it by repointing the first rule at the final destination, not by deleting the second.

Loops. Two rules point at each other, usually a trailing-slash rule fighting a canonical-host rule. Your browser gives up with a generic "too many redirects" page; here you see the same two URLs alternating, which tells you exactly which pair to look at.

Broken redirects. A 301 landing on a 404. Common after a second migration, when the destination of an old redirect gets retired without anyone checking what still pointed at it.

Redirect-to-homepage. Every retired URL sent to /. The chain looks perfect — 301 then 200 — and it is the most damaging of the four, because it tells search engines the old page's topic no longer exists anywhere on the site. Only the Location value reveals it.

A migration is the one time redirects are written in bulk, which is why it is the one time they go wrong in bulk. Work from the list, not from memory: export your top entry pages, then trace them.

Check the four root variants first — http://, https://, the www and the non-www form. You want exactly one hop from each into the canonical address. Two hops here is the chain we see most often, and because every visitor and every crawler passes through it, fixing it is the single highest-leverage change on this list.

Then work down the list by traffic. For a whole column at once, the bulk HTTP status checker traces up to 25 URLs in one pass and shows every hop of each chain; come back here when one row needs looking at properly. The rest of the sequence — timing, sitemaps, when to retire the old URLs — is in our website migration checklist.

Chains rarely appear on their own. They appear when a new rule lands on top of an old one, and the place they do the most damage is inside your own internal links, because those are the links you control and the ones crawlers follow most often.

A link written three years ago to /old-product still works — it redirects — so nothing looks broken and nobody fixes it. But every crawl of that page spends a request on the hop, and broken internal links are already the most common Links failure in our audits, at a median of 14 per site (H1 2026).

Pull a page's outbound links with the internal link checker, then trace the ones that look old. Updating the link itself is strictly better than relying on the redirect: it removes the hop permanently instead of paying for it on every request.

Tracing one URL answers one question well. Three things extend it, in order of effort:

• A list instead of a URL. The bulk HTTP status checker takes up to 25 URLs at once, with the response headers and timing at each hop — the right tool for a migration sheet.
• The links pointing into your chains. The internal link checker lists a page's outbound links with their anchor text, so you can fix the source rather than the symptom.
• Every page, not the ones you thought to check. A full site audit runs all 251 rules across the site and finds the chains on pages you would never have pasted in one at a time.

All three are free, and all of them sit alongside the rest of SEOmator's free SEO tools.