TF-IDF
TF-IDF is a formula that scores how important a word is to a page: frequent on that page, rare across others.
TF-IDF, short for term frequency times inverse document frequency, is a formula that scores how important a word is to a page compared with a collection of other pages. A word scores high when it appears often on this page and rarely elsewhere. Common words like "the" score low everywhere, because every page uses them.
It's one of the oldest ideas in information retrieval. In SEO it turns up in content tools that compare your page with the pages that rank and suggest words you've left out.
How it works
Term frequency counts how often a word appears on a page. Inverse document frequency measures how rare the word is across the whole collection. Multiply the two and you get a score that picks out the words that make a page distinctive.
For example, across a hundred coffee blogs, "coffee" appears everywhere, so it scores low. "Cascara" appears on only a few pages. On Fernhill Coffee's page about coffee cherry tea, it appears ten times, so it scores high there, and marks the page as being about something specific.
Is it useful for SEO?
As a way to spot gaps, yes. If every page that ranks for "French press" mentions grind size and yours doesn't, you've probably missed something readers want.
As a target, no. Modern search engines understand meaning as well as word counts, and padding a page with words to match a score reads badly and can tip into keyword stuffing.
TF-IDF and AI search
AI platforms find and compare text using embeddings, which capture meaning, so a page can match a question without sharing its words. That makes covering the ideas in a question more useful than repeating its terms. TF-IDF is still a simple way to see which ideas the pages that already get cited have in common.