Google Says Search Got Much Better at Identifying Spam Sites Last Year

In a Webspam report, Google says it caught six times more spam sites in 2021 than in 2020.

Imad Khan Senior Reporter
Imad is a senior reporter covering Google and internet culture. Hailing from Texas, Imad started his journalism career in 2013 and has amassed bylines with The New York Times, The Washington Post, ESPN, Tom's Guide and Wired, among others.
Expertise Google | Internet Culture
Imad Khan
2 min read
Getty Images

Google Search caught six times more spam sites in 2021 than it did in 2020, according to a Webspam report published by Google on Thursday. The Alphabet subsidiary lauded its AI-based spam-prevention system called SpamBrain, which it said has helped keep "more than 99% of searches spam-free."

SpamBrain has helped prevent users from clicking to harmful sites that might inject malware and to scam sites that aim to trick people into revealing personal information or wiring money , the search giant said. AI-based systems are necessary given that spammers are constantly finding ways to circumvent the company's filters, Google said. 

Search is Google's most significant revenue driver, making it important for the company that high quality and safe sites rise to the top of its results. Spam sites use search engine optimization, or SEO, techniques to artificially improve their search rankings. Google filters out spam sites designed to trick its algorithm.

SpamBrain led to a 70% reduction in hacked spam, Google said. Hacked spam involves hackers infecting websites with harmful code that can record people's keystrokes and steal their login credentials. Other types of hacked spam can inject malware onto people's computers and redirect them to malicious sites. SpamBrain helps keep these websites out of Google Search, Google said. 

"Scammers often attempt to create low quality websites with keyword stuffing, logos of brands they're imitating, and a phone number they want you to call," Ned Adriance, policy communications manager at Google, said in an email. "Our algorithmic solutions work to make sure that scam and fraud are very unlikely to show up in your search results."

Google said it also reduced gibberish spam by 75%. Gibberish spam involves malware or scam sites that use keyword-stuffed nonsensical sentences in an attempt to boost their ranking. Spam sites can also use link spam to game Google's quality signal, known as PageRank. This is done by having other sites link to them, sometimes by offering money, to boost ranking. 

"We have many solutions to identify and nullify link spam, and sites engaging in link spam are wasting valuable time and money that could be better invested in better content or user experience," said Adriance.

SpamBrain launched in 2018. Google says search was able to catch 200 times more spam sites last year compared with when it first started weeding out malicious sites nearly two decades ago. Still, it's an ongoing battle between those aiming to abuse or narrowly skirt the company's quality guidelines and Google trying to elevate valuable content.