Google SpamBrain: neural detection at the network level.
SpamBrain is Google's neural detection system for spam and link schemes, introduced internally around 2018 and publicly confirmed in 2022. The system scores at the network level rather than per-link. The March 2024 link spam update was substantially a SpamBrain enforcement cycle that targeted marketplace placements, scaled guest-post networks, and parasite-SEO patterns. White-hat methodology meets the threshold by rendering organic-outreach fingerprint across the source-domain footprint.
Four moves the SpamBrain detection surface runs against.
The system is neural and operates on the link-graph network. The scoring is at the network level. The March 2024 update intensified the enforcement surface. White-hat methodology renders the same fingerprint as organic editorial outreach across every detection dimension.
SpamBrain is Google's neural detection system for spam and link schemes.
SpamBrain was introduced internally at Google around 2018, publicly mentioned by Google in 2020 and 2021, and confirmed as a named system in 2022. The system applies neural models to score web content and link patterns against trained spam signals. The off-page surface SpamBrain operates against is the link-graph network: which source domains link to which destination domains, what anchor patterns appear, what the source-domain footprint looks like, what timing pattern the links land in. The system runs continuously as part of Google's core infrastructure and intensifies on named update cycles.
SpamBrain scores at the network level rather than per-link.
The system reads link-graph patterns at the network level. A single link from a source-domain to a destination-domain is not what SpamBrain scores against; the system reads the source-domain's broader linking pattern, the destination-domain's broader inbound pattern, and the network-level structure those patterns sit inside. Source-domains that link out to many destinations following the same anchor-pattern footprint get flagged as scheme-coordinated. Destination-domains whose inbound profile shows a high concentration of links from one such source-domain cluster get downstream-affected. The network-level scoring is why white-hat methodology focuses on the source-domain footprint rather than the per-link decision.
The March 2024 link spam update intensified the SpamBrain enforcement surface.
The March 2024 core update and the parallel link spam update were substantially a SpamBrain enforcement cycle. The update targeted entire surface categories at once: marketplace-sourced placements, scaled guest-post networks selling placements without editorial vetting, source-domain clusters with shared analytics IDs or shared themes, and parasite-SEO patterns where established domains hosted external content with link-injection signals. Sites placing through these surfaces saw inbound profile drops on the update cycle. The drops did not require a manual action because the discounting was algorithmic at the SpamBrain layer.
White-hat methodology renders the same fingerprint as organic editorial outreach to SpamBrain.
White-hat off-page methodology meets the SpamBrain threshold by rendering the same fingerprint as organic editorial outreach. The fingerprint includes: mailbox identity segmented per campaign cohort so the outreach pattern does not cluster, outreach paced against the 30-day moving average the link-graph evaluates over, prospect networks vetted against TrustRank seed-set proximity rather than by raw DR alone, no shared C-class IP clustering across the source set, no whois-history overlap, no content-template overlap across the source set, no shared analytics IDs or AdSense IDs across the source set. The SpamBrain detection surface is the boundary between white-hat methodology and link-scheme apparatus regardless of intent.
Methodology questions we get during the audit conversation.
What is Google SpamBrain?
SpamBrain is Google's neural detection system for spam and link schemes. The system was introduced internally at Google around 2018, publicly mentioned by Google in 2020 and 2021, and confirmed as a named system in 2022. SpamBrain applies neural models to score web content and link patterns against trained spam signals. The off-page surface SpamBrain operates against is the link-graph network: which source domains link to which destination domains, what anchor patterns appear, what the source-domain footprint looks like, what timing pattern the links land in. The system runs continuously as part of Google's core infrastructure.
How does SpamBrain detect link schemes?
SpamBrain scores at the network level rather than per-link. The system reads the source-domain's broader linking pattern, the destination-domain's broader inbound pattern, and the network-level structure those patterns sit inside. Source-domains that link out to many destinations following the same anchor-pattern footprint get flagged as scheme-coordinated. The detection signals include shared C-class IP clustering, whois-history overlap, content-template overlap across the source set, shared analytics or AdSense IDs across the source set, and timing patterns where many links land in a short window. White-hat methodology meets the threshold by rendering the same fingerprint as organic editorial outreach across all of these dimensions.
What was the March 2024 link spam update?
The March 2024 core update and the parallel link spam update were substantially a SpamBrain enforcement cycle. The update targeted entire surface categories at once: marketplace-sourced placements, scaled guest-post networks selling placements without editorial vetting, source-domain clusters with shared analytics IDs or shared themes, and parasite-SEO patterns where established domains hosted external content with link-injection signals. Sites placing through these surfaces saw inbound profile drops on the update cycle. The drops did not require a manual action because the discounting was algorithmic at the SpamBrain layer.
How can I tell if SpamBrain affected my site?
SpamBrain demotions do not surface in Search Console the way manual actions do. The diagnosis is inferred from the ranking-and-traffic pattern coinciding with a confirmed Google update. The audit pulls the inbound profile, segments by acquisition pattern, identifies placements fingerprinting as scheme-coordinated (anchor pattern, source-domain footprint, content-template overlap), and reads whether the traffic drop aligns with the demotion pattern. A confirmed Google update date plus a traffic-drop start date plus a placement-pattern match across the inbound profile is the standard diagnostic triangulation. Manual-action recovery has a defined remediation path; algorithmic SpamBrain recovery runs until the next update incorporates the cleaned profile.
How does white-hat methodology meet the SpamBrain threshold?
White-hat off-page methodology renders the same fingerprint as organic editorial outreach. The fingerprint includes: mailbox identity segmented per campaign cohort, outreach paced against the 30-day moving average the link-graph evaluates over, prospect networks vetted against TrustRank seed-set proximity rather than by raw DR alone, no shared C-class IP clustering across the source set, no whois-history overlap, no content-template overlap, no shared analytics or AdSense IDs across the source set. The SpamBrain detection surface is the boundary between white-hat methodology and link-scheme apparatus regardless of intent. Methodology that requires interpretation around these signals is grey-hat by definition. An off-page SEO company running campaigns inside the SpamBrain boundary scopes against every dimension above as part of the prospect-list workflow.
What's the difference between SpamBrain and Penguin?
Penguin and SpamBrain are complementary surfaces. Penguin (Penguin 4.0 launched September 2016, integrated into the core algorithm) evaluates anchor-text distribution per-URL and discounts anchors that over-concentrate against the per-vertical baseline. SpamBrain (introduced 2018, confirmed 2022) evaluates link-graph network patterns at the source-domain and network level and discounts links from networks that fingerprint as scheme-coordinated. A site can fail Penguin (anchor-text over-optimization on a single landing page) without failing SpamBrain, and vice versa. White-hat methodology meets both thresholds: per-URL anchor distribution stays inside the per-vertical baseline, and the source-domain footprint reads as organic editorial outreach.
SpamBrain is the boundary. White-hat methodology stays inside it by design.
The audit reads the inbound profile against the SpamBrain detection patterns, surfaces the placement gap against the per-vertical baseline, and scopes the campaign mix that renders organic-outreach fingerprint across the source-domain footprint.