Google will follow a reasonable number of hops in a redirect chain, but its documentation consistently recommends against chains altogether, since they add crawl overhead and delay signal consolidation, without ever publishing a specific numeric figure for how much ranking signal is lost or degraded at each individual hop. Any claim of a precise “X% signal loss per hop” is not something Google has verified or disclosed; the honestly stated, practically observed pattern is that longer chains correlate with slower reprocessing and messier, less confident consolidation, not a quantified decay rate that can be cited as fact.
The mechanism: why chains slow consolidation without a disclosed decay formula
Google’s site-move-with-URL-changes documentation recommends redirecting each old URL directly to its closest equivalent new URL in a single hop, explicitly discouraging redirect chains as part of that guidance. The stated reasoning is practical rather than about a specific mathematical penalty: each additional hop in a chain requires Google’s crawler to make another request and follow another redirect before it can resolve to the final destination, adding processing overhead to every single crawl of that URL. At the scale of a full domain migration, potentially thousands or millions of URLs each requiring this resolution, that overhead compounds across the whole recrawl-and-reconsolidation process the migration already depends on, plausibly slowing how quickly Google works through confirming and consolidating signals across the full migrated URL set.
Beyond the pure overhead, chains introduce additional complexity and potential fragility, more hops means more opportunities for an individual link in the chain to be misconfigured, to point somewhere unintended, or to eventually break if any intermediate step in the chain is later removed or changed, particularly likely during a domain migration where multiple technical systems and teams may be involved in different parts of the redirect implementation. A single, direct hop from old URL to final new URL removes this entire category of risk, since there’s no intermediate step that can independently fail or be later disrupted.
Why no verified per-hop decay figure exists, and why that matters
It’s worth being direct about this: no specific, Google-verified percentage or formula describing signal loss per additional redirect hop has ever been published or confirmed by Google. Any number quoted for this, “you lose X% of link equity per hop,” circulating in SEO commentary is not sourced to anything Google has stated or documented, and repeating such a figure as established fact means treating an unverified, likely fabricated number as if it were real. The genuinely defensible claim is qualitative and directional: chains correlate with slower, less certain consolidation and added crawl overhead, not a quantified per-hop decay rate.
This distinction matters practically because it shapes what kind of guidance is honest to give a team asking about redirect chain risk during a migration. Telling a team “each hop costs you roughly 10-15% of your ranking signal” sounds authoritative and actionable, but it’s inventing a precision Google has never disclosed and that no independent, verifiable study has established either. The accurate and still genuinely useful guidance is structural: avoid chains, redirect directly to the final destination in one hop, because chains slow and complicate the process Google needs to complete anyway, not because a specific quantified penalty per hop has been confirmed.
What “a reasonable number of hops” actually implies
Google’s crawlers will follow chains up to some practical limit, beyond which continued resolution becomes unlikely or unreliable, though Google hasn’t published an exact maximum hop count as a hard rule site owners can rely on. The practical takeaway from this isn’t “so a chain of two or three hops is basically fine since Google will follow it anyway”; it’s that following a chain successfully is different from following it efficiently and with full confidence, and even chains well within whatever practical following-limit exists still carry the overhead and consolidation-delay costs described above relative to a direct single-hop redirect, independent of whether Google technically manages to resolve the chain at all.
A hypothetical illustration of how chains accumulate
Consider a hypothetical mid-size retailer, “Northwind Outdoor Gear,” migrating domains for the third time in six years. Each prior migration redirected old URLs to whatever the then-current URL structure was, and the newest migration’s redirect map was built by pointing old URLs to the previous migration’s URLs rather than tracing all the way back to a true final destination. A product page from the original site could plausibly end up two or three hops deep before landing on its current equivalent, without anyone on the current team realizing the chain existed, since each individual redirect looked correct in isolation. Auditing the full redirect map and collapsing every chain down to a single hop from the oldest URL straight to the current one would likely reduce crawl overhead across the affected URLs and could speed up how quickly Google finishes reconciling signals for those pages, though no specific percentage or timeline could be promised from that cleanup alone.
Practical implication
During any domain migration, implement redirects as direct, single-hop mappings from each old URL straight to its final new-domain equivalent, auditing the redirect map specifically for accidental chains, cases where an old URL redirects to an intermediate URL that itself redirects again before reaching the true final destination, which can occur inadvertently when migration work happens in stages or when multiple past redirect implementations stack on top of each other over a site’s history. If a chain is discovered, whether from a previous migration or introduced during the current one, update it to point directly to the final destination rather than leaving the intermediate hop in place. When explaining this to stakeholders, frame the recommendation around avoiding unnecessary crawl overhead and consolidation delay, which is the genuinely documented and defensible mechanism, rather than citing any specific per-hop percentage figure, since no such figure exists in verifiable form and presenting one as fact undermines the credibility of otherwise sound technical guidance.
