The 301 Redirect Myth That’s Still Costing Sites PageRank
Somewhere around 2012, the SEO industry locked onto a clean story: 301 redirects pass ~99% of PageRank, 302s pass less, and that’s basically the whole picture. Still being taught in guides published this year. Still incomplete enough to cause real damage on large migrations and consolidation projects.
Here’s what Google’s indexing pipeline actually does with a redirect chain — and why the percentage framing was always a simplification that obscures the mechanisms that matter.
What Happens Inside the Pipeline When Googlebot Hits a Redirect
Googlebot fetches a URL and gets a 3xx response. Three things happen in sequence, and each one introduces a potential cost that has nothing to do with whatever PageRank percentage is being passed.
1. The crawl budget gets consumed at the source URL
Googlebot spends a crawl slot reaching the redirecting URL before it even knows it needs to follow somewhere else. On a site with thousands of redirected legacy URLs still being internally linked or sitting in sitemaps, this is pure waste — the crawl cost is paid at the old address, then paid again at the destination. For sites where Google is already crawling conservatively (new domains, slow servers, very large URL spaces), redirect hygiene directly translates to indexing lag on pages that actually matter.
2. Chain length determines whether Google follows the redirect at all
Google’s documentation notes following up to 5 hops in a redirect chain. In practice, longer chains don’t always generate a crawl error in Search Console immediately — they quietly stop being followed. Intermediate URLs may stay in the index as soft 404s or remain uncrawled, and the destination URL may never receive the link signals from the source. A three-hop chain built across two site migrations isn’t unusual, and it’s rarely visible in standard audits unless you’re crawling live URLs against actual server responses rather than checking your redirect rules file.
3. Link signal consolidation is not instantaneous
When Google processes a 301 and determines the destination is the canonical, the link signals from the source get re-attributed through the link graph. This is not a real-time operation. On large sites there’s a meaningful window — sometimes weeks — where the destination is indexed but hasn’t fully inherited the authority the redirect is supposed to transfer. This is the gap that produces post-migration traffic drops that look like ranking losses but are actually processing lag. The mistake is responding to that lag with more changes.
Why the “99% PageRank” Framing Is Too Simple to Be Useful
The percentage framing treats link equity like water through a pipe with a small leak. The real mechanism is closer to a re-evaluation of the destination URL’s authority from scratch, informed by the old URL’s signals but not identical to them. A few things break the clean transfer story:
- Topical relevance between source and destination matters. If the redirected URL had strong signals for a specific topic cluster and the destination sits in a different cluster, the re-attribution is not 1:1. Google’s topical co-occurrence layers influence how link signals from the old URL get applied to the new one.
- Anchor text context moves with the redirect, but the destination page’s existing context can dilute it. A redirect doesn’t merge two pages’ relevance signals cleanly — it points the source’s inbound links at the destination’s existing content. If that content targets something different, you may be pointing authority at the wrong page rather than transferring it precisely.
- Redirect chains accumulate signal loss across multiple re-evaluation steps, not a single continuous leak. Each hop is another point at which Google may decide the destination is different enough to not fully inherit the source’s authority.
The Case That Actually Breaks This: Site Consolidations
We’ve worked through keyword cannibalization between competing properties under the same ownership — exactly the scenario where you’d redirect one domain to the other or merge subdirectories. The instinct is to treat the 301 as a clean consolidation move.
When both properties have existing authority in the same topical cluster, redirecting one to the other can initially create a signal conflict. Google’s indexing pipeline has to reconcile two previously separate entities into one, and during that reconciliation both ranking histories become temporarily unreliable predictors of where the combined entity will settle. In our experience the merged site often underperforms both predecessors for several weeks, then recovers — but only if the destination’s content actually justifies the consolidated authority it’s now carrying.
The sites that don’t recover are the ones where the redirect moved authority to a thinner or less topically coherent destination. Google’s quality classifiers don’t care that the 301 was technically correct. They evaluate the destination on its own merits after consolidation stabilizes.
What a Redirect Audit Should Actually Check
Most audits stop at confirming the redirect returns a 301 and points at the right destination. That’s table stakes. What actually drives ranking outcomes:
- Chain length on live URLs — crawl the live redirects, not the rules file. Rules files can obscure chains built across multiple layers (CDN, server, CMS).
- Internal links pointing at redirected URLs — every internal link hitting a redirect is a crawl slot spent unnecessarily and a PageRank transfer happening through an extra hop. Fix the internal links, not just the redirects.
- Sitemaps listing redirected URLs — this signals to Google that the redirected URLs are supposed to be canonical, which creates a contradiction that slows resolution.
- Redirect destination topical alignment — especially for category or directory-level redirects during migrations. A product category redirected to a homepage because the new site dropped that category is not a clean transfer; it’s a signal that the destination is wrong.
- Time since migration vs. Search Console coverage data — if indexed count dropped post-migration and hasn’t recovered after six or more weeks, the redirect implementation probably isn’t the problem. The destination content is.
The One Rule Worth Keeping From the Old Mental Model
Use 301s for permanent moves. Use 302s when you genuinely intend to restore the original URL. That part holds because the signal consolidation behavior is genuinely different — Google is more willing to consolidate authority through a 301 because a 302 signals the source URL should stay in the index as the canonical.
But a correctly implemented 301 is just a precondition, not a solved problem. What determines whether the transfer succeeds is whether the destination deserves what the source was carrying — and Google evaluates that independently.
Before your next migration: identify which redirected URLs carry real inbound link equity from sources Google actually trusts, and confirm those destinations are your strongest content, not placeholders or category shells. That’s the step most migration checklists skip — and the one that decides whether the traffic comes back.