Google Hummingbird
Hummingbird was a major rework of Google's ranking systems in 2013 that helped Google understand what a whole search means, beyond its single words.
Google Hummingbird was a major improvement to Google's ranking systems, made in August 2013. Google lists it among its retired systems, because its work has long since been absorbed into how Google ranks everything.
It's remembered as the point where Google moved from matching the words in a search to understanding what the whole search meant. That's the start of a path that leads straight to AI search.
What changed
Google says little more about it. In SEO, Hummingbird is known as the change that made Google pay more attention to the meaning of a whole search, so that pages could match a question even when they didn't repeat its exact words.
This fitted how people were starting to search: longer questions, typed on phones or spoken out loud.
What it meant for marketers
Pages written to repeat one exact phrase lost their edge. Pages that answered the question someone was really asking gained. Writing for a topic, with the related questions answered in plain language, started to matter more than hitting a keyword a set number of times.
From Hummingbird to AI answers
AI platforms take this much further. A language model reads a question as meaning, and it can match a page to a question even when they share hardly any words. For example, a page from Harrow & Pine titled "Our quietest burr grinder" can help answer "Which grinder won't wake the baby?" even though the page never uses the word "baby."
So the advice that started with Hummingbird still holds: write clear answers to the questions your customers ask, in their words, and don't worry about exact repeats.