Understanding it
The most widely used classification separates informational queries (understand, learn), navigational queries (reach a known site) and transactional queries (act, buy). It comes from a 2002 research paper. Practitioners often add commercial intent: comparing before choosing.
The best way to read intent is to type the query into Google. If the results show stores, the intent is to buy. If they show guides and comparisons, the intent is to learn. Google has already decided, based on the behavior of millions of searchers.
With AI assistants, queries get longer and turn into real requests. Intent becomes more explicit, and a page has to address the situation described, not just the keyword.
A concrete example
Illustrative example: “payroll software”
A payroll software vendor publishes a long blog post on “payroll software”, which stays on page 3. On Google, the query shows vendor product pages and comparison sites: the intent is to choose a tool, not to read an article.
The vendor builds a complete product page (features, pricing, integrations, reviews) and keeps the article for another query, “how does payroll software work”, whose intent really is informational. Each page now has a single job.
The common mistake
One page for every intent
Trying to inform, compare and sell on the same page produces a page that does none of the three well. One intent, one page, then internal links to move the reader from one to the next.