RankBrain
Machine learning component of the Google search algorithm that interprets unfamiliar queries and weights results by user behaviour.
Also known as: Google RankBrain, ML ranking signal
RankBrain is a machine learning system that has been part of the Google search algorithm since 2015. It is particularly tasked with interpreting queries the system has never seen before and inferring the intent behind them. To do that it combines word meanings, context and the actual behaviour users show on the result pages.
How RankBrain understands queries
Rather than matching terms verbatim, RankBrain translates words into mathematical vectors and maps them to semantically related terms. An unusual query is therefore linked to known topics for which Google already holds ranking data. Click through rate, dwell time and pogo sticking back to the search feed into the assessment of whether the served results actually fit.
Implications for SEO
For search engine optimisation RankBrain means that rigid keyword density and exact phrase matching have lost importance. What matters is content that fully answers the user's underlying question. Topic clusters, clear structures and semantically related terms support rankings more reliably than repeating a single head keyword.
Practical relevance in everyday marketing
Anyone building landing pages or content for organic search should research real user questions and answer them cleanly. Tools for search intent and semantic analysis help you identify the topics that sit behind a query. Clean headings, well organised paragraphs and useful internal linking provide the signals that RankBrain rewards.