SpamBrain is Google's AI-based spam-prevention system and the named part of its spam detection systems. Google says it launched in 2018 and is regularly improved through spam updates. Since the December 2022 link spam update it also detects sites buying links and sites used to pass outgoing links, and neutralises those links. For a site, the practical consequences are that unqualified paid or affiliate links, scaled low-value content and hacked pages are all within its scope.
What is SpamBrain?
SpamBrain is Google's AI based spam-prevention system. Google's ranking systems guide lists it as one of a range of spam detection systems that deal with content and behaviour that break Google's spam policies.
Google's spam updates page describes a spam update as a notable improvement to these systems, and uses SpamBrain as its example.
We employ a range of spam detection systems, including SpamBrain, to deal with content and behaviors that violate our spam policies.
Google, A guide to Google Search ranking systems, Google Search Central
What does SpamBrain do with links?
The July 2021 link spam update was announced on 26 July 2021 as an algorithm change for "identifying and nullifying link spam more broadly". That post does not mention SpamBrain, so we do not claim SpamBrain powered it.
The December 2022 link spam update, announced on 14 December 2022, was explicitly built on SpamBrain. Google said SpamBrain could now detect both sites buying links and sites used to pass outgoing links, and that credit passed by those links would be lost. In its 2022 spam report, published 11 April 2023, Google said this detected 50 times more link spam sites than the previous link spam update.
Today, we're leveraging the power of SpamBrain to neutralize the impact of unnatural links on search results.
Google, December 2022 link spam update releasing for Google Search, Google Search Central Blog
What evidence is there that Google uses SpamBrain?
This is a Confirmed system. Google names SpamBrain in its ranking systems guide, its spam updates documentation and its annual spam reports. The 2021 spam report, published 21 April 2022, gives the launch year and says SpamBrain identified nearly six times more spam sites in 2021 than in 2020.
Google does not publish SpamBrain's features, model type or thresholds. Claims about specific signals it uses are speculation.
SpamBrain was launched in 2018 and we've been continuously improving its performance.
Cody Kwok, Google, How we fought Search spam on Google in 2021, Google Search Central Blog
What does SpamBrain mean for sites?
Two things follow from Google's own statements. First, a site does not need to buy links to be affected: a site that sells or passes unqualified links is also a detection target. Second, neutralised links simply stop counting, so a ranking drop after a link spam update can mean you had been relying on links that no longer pass value.
The falsifiable test for your own site: every paid, sponsored, affiliate or user-generated outbound link should carry rel="sponsored", rel="ugc" or rel="nofollow" as described in Google's guide to qualifying outbound links. If they do not, the site looks like a link seller.
- Tag affiliate and paid links with rel="sponsored" at template level so new content cannot slip through.
- Tag comment, forum and profile links with rel="ugc".
- Monitor for hacked pages and injected links, which Google also reports SpamBrain detecting.
Which Laurelin audit checks test for SpamBrain risk?
The new check, Paid, sponsored or affiliate outbound links are not qualified, finds commercial outbound links without the right rel value. It tests for a documented SpamBrain target; it cannot show whether SpamBrain has flagged the site.
Related checks:
- Inauthentic bought mentions or links: the buying side of the same link spam.
- Outbound links to spam: outgoing links that associate a site with spam networks.
- Auto-generated content with no value: scaled content is within the spam policies SpamBrain enforces.
- Expired domain or parasite abuse: another spam policy category.
- Cloaking: a long-standing spam policy breach.
- Toxic or spammy backlink profile: inbound links that may already be neutralised.
What are the key dates for SpamBrain?
- 2018: SpamBrain launched (year given by Google) (source)
- 2021-07-26: Link spam update announced (no mention of SpamBrain) (source)
- 2022-04-21: Google's 2021 spam report names SpamBrain and its 2018 launch (source)
- 2022-12-14: December 2022 link spam update uses SpamBrain to neutralise unnatural links (source)
- 2023-04-11: Google reports 50 times more link spam sites detected than the previous link spam update (source)
Frequently asked questions about SpamBrain
When did SpamBrain launch?
Google says SpamBrain was launched in 2018 and has been continuously improved since.
Did the 2021 link spam update use SpamBrain?
Google's July 2021 announcement does not mention SpamBrain. The December 2022 link spam update is the first one Google explicitly said used SpamBrain.
Can I get a SpamBrain penalty removed?
SpamBrain is automated, so there is no reconsideration request for it. Fix the policy issue; Google says sites may improve once its automated systems see they comply. Manual actions are separate and do have a review process.