TL;DR
In August I found forty links in my own site’s header and footer that each fired a redirect. The site strips trailing slashes and every navigation link had been written with one. Nothing was broken and nothing looked wrong. Internal links pointing at redirects is the real cost and the cheapest fix.
In August I found forty links in my own site’s header and footer that were each firing a redirect.
The site strips trailing slashes, so /seo-audit/ 301s to /seo-audit, and every navigation link had been written with the slash. So every page on the site was linking to forty URLs that immediately bounced somewhere else. Nothing was broken. Nothing looked wrong. I had been looking at that menu for months.
That is the whole character of this problem. Redirects do not fail loudly. They work, which is why nobody looks at them.
It belongs to the same family as everything else a site accumulates without anyone deciding: individually harmless, never surfaced by a tool as urgent, and only visible to somebody who went looking.
What actually costs you
Let me separate the real from the folklore, because this topic attracts a lot of the second.
- Internal links pointing at redirects. The real one, and the cheapest to fix. Every internal link to a URL that redirects is a link you placed deliberately at an address you have decided is not the right one. Google follows it and consolidates, so nothing dramatic happens, but you are spending crawl on the round trip and pointing your own votes at a URL you have deprecated. On my site it was forty links on every page.
- Chains of three or more hops. Each hop is another request and another chance for something in the middle to break. Google will follow a reasonable number and then give up, and the pages at the end of long chains are the ones that quietly stop being visited.
- Redirect loops. Rare, fatal, usually created by two well-intentioned rules meeting.
- Redirects to irrelevant destinations. The one that actually loses value. Pointing forty retired product URLs at the home page is not consolidation. Google is likely to treat those as soft 404s, because the destination does not answer what the original did. The nearest sensible category is the right target, and if there is no sensible category the honest answer is a 410.
Google describes that outcome directly in its documentation on soft 404s: a URL returning a success code while the content signals the page does not exist. A redirect to a home page that has nothing to do with the request produces exactly that mismatch, which is why the tidy-looking fix so often shows up later as a new row in the indexing report.
What does not cost you much
A single clean 301. It is the correct tool and it does its job. There is no meaningful penalty for having redirects, and a site with a long history should have plenty.
The “link equity loss” figure. People still quote a percentage lost per hop. I would not build a business case on it. The measurable costs are crawl waste, latency and breakage, and those are enough of an argument without inventing a number.
The one that takes a site down
HTTP to HTTPS, done incompletely.
On an agency site I audited, http://www.example.com/ was not redirecting to the https version at all. The site was running as two sites, and had been for years, with no one noticing because everyone typed or clicked their way to the secure one.
The 2025 Web Almanac found 91.5% of mobile pages served over HTTPS, so roughly one in twelve still is not, more than a decade after Google called it a signal. The ones I find are rarely sites that never migrated. They are sites that migrated and left something behind: a canonical on http, one host variant unredirected, an internal link hard-coded in a template.
Reading it in a crawl
Look at the proportion, not the count. One crawl I ran on a small manufacturer’s site returned 174 URLs, of which 68 were redirects and 16 were broken. Sixty-eight redirects on a large site is housekeeping. Sixty-eight out of 174 is a site that has been rebuilt more than once with nobody cleaning up after.
Then check two things specifically: the maximum chain length, and how many internal links point at redirecting URLs. The second number is the actionable one and most reports bury it.
Fixing it
Update the internal links to the final destination. Do this first. It is free, it is safe, and on most sites it is a find-and-replace in a template rather than a project. It also removes most of the crawl cost immediately.
Collapse chains to a single hop. A→B→C becomes A→C and B→C. Keep both rules. Deleting the middle one breaks anything still linking to B.
Do not delete old redirects to tidy up. They are load-bearing. A redirect from 2019 is carrying whatever links that URL earned, and removing it converts a working page into a 404 for no benefit.
Check the host and protocol variants once, properly. With and without www, http and https, trailing slash and not. Five minutes, and it is the check that catches the serious version.
The short version
- Redirects do not fail loudly. I had forty on every page of my own site.
- Internal links pointing at redirects is the real cost and the cheapest fix.
- Redirecting everything to the home page is not consolidation. Expect soft 404s.
- A single clean 301 costs you nothing. Ignore the per-hop equity percentages.
- Incomplete HTTPS migration is the version that hurts. One page in twelve still is not on HTTPS.
- Read the proportion. 68 redirects in 174 URLs is a different finding from 68 in 20,000.
- Never delete old redirects to be tidy. They are carrying something.
If a site should be ranking and it isn’t, that’s the work I do. Technical SEO covers redirects, canonicals and the crawl layer. If you’re not sure which of several plausible problems is costing you, that’s what an SEO audit is for.
