Google’s search engine algorithm, the secret code behind how websites are ranked on Google Search, has been leaked in a 2,500-page document called the “Content API Warehouse.” While Google has not confirmed the legitimacy of the leak, SEO experts like Rand Fishkin believe it to be legitimate. Fishkin, the CEO of Sparktoro and founder of Moz, received the document from an unnamed source and has shared some of its interesting details.
One surprising revelation from the leaked document is that Google uses Chrome to rank pages. This contradicts Google’s previous denial of using Chrome for ranking websites. It appears that Google tracks how many clicks a webpage receives from users in Chrome to determine which pages of a website to include in its search query sitemap. While this doesn’t affect the overall ranking of a website, it does influence which internal pages are shown in search under the website’s homepage.
Another intriguing finding is the existence of a tag that Google attaches to “small personal websites.” The purpose of this tag is unknown, as is how Google determines what constitutes a “small” or “personal” website. SEO experts speculate whether this tag is meant to promote or demote these websites in search rankings.
The leaked document also confirms the significance of user clicks for search rankings. Google relies heavily on click data to improve search results through a ranking factor called NavBoost. NavBoost includes a specific module focused on click signals, with short clicks versus long clicks playing a role in determining a website’s ranking. This contradicts Google’s previous denials and confirms the long-held speculation of SEO experts.
Contrary to popular belief, exact match domain names may not be beneficial for search rankings. While they were favored by Google’s algorithm in the past, evidence from the leaked document suggests that Google now actively demotes these websites in search. The algorithm views exact match domain names as potential spam and treats them similarly to keyword stuffing practices.
The leaked document also reveals that Google maintains whitelists for certain topics. Websites related to COVID information, politics queries, and travel appear on these whitelists, requiring manual approval for inclusion in search results. This sheds light on how Google manually curates search results for specific topics, deviating from the usual algorithmically ranked search factors.
Overall, the leaked document has exposed some discrepancies between Google’s public statements and the actual workings of its search algorithm. SEO experts like Fishkin and Mike King feel that Google has not been entirely truthful about how its algorithm functions. As the industry continues to analyze the leaked document, more interesting details about Google’s search algorithm may emerge, challenging our current understanding of SEO practices.