Search Intent
Search intent is the reason someone typed a query at all, not the words they used to type it. The same string can hide four different goals: learning something, finding a specific site, comparing options, or buying now. Search engines and AI assistants both try to infer which goal is in play and match content to it, so intent, not keyword density, decides what actually gets shown or cited.
Why search intent matters
Two people can type the exact same phrase and want completely different things. Someone searching "Golf" might mean the car, the sport, or the body of water; someone searching "best CRM for small teams" might be starting research or ready to sign up. Ranking and citation systems have gotten good at reading which of these a query implies, and they favor content built for that specific need over content that merely mentions the topic. This matters more, not less, with AI assistants: an AI Overview or a ChatGPT answer has to pick a small number of sources to summarize, and it picks the ones that resolve the actual question cleanly. Content aimed at the wrong intent can be well written, accurate, and still invisible, because it is answering a question nobody asked.
How search intent works
Most practitioners sort queries into four buckets. Informational queries want an explanation ("how does a heat pump work"). Navigational queries want a specific destination ("Chase online banking login"). Transactional queries want to complete an action, usually a purchase ("buy running shoes size 10 wide"). Commercial investigation sits between the last two: someone is comparing before they commit ("best budget e-bikes 2026"). Search engines and language models infer the bucket from the wording, surrounding context, and any earlier turns in a conversation, then favor sources whose format, depth, and tone match that bucket — a comparison table for commercial investigation, a direct numeric answer for a quick informational lookup, a homepage or login page for navigational intent. Matching the format to the intent matters as much as matching the topic.
Common mistakes
The most common error is treating the keyword and the intent as the same thing: stuffing a phrase into a page without asking what the searcher actually wants once they arrive. A second is mismatched format — writing a 2,000-word guide for a transactional query when the person wants a price and an "add to cart" button, or the reverse. A third is trying to serve several intents on one page, which usually ends up serving none of them well. And a fourth, increasingly common one, is guessing at intent instead of checking it: looking at what actually ranks or gets cited for a query is a faster and more reliable signal than assumption. Google's own guidance for AI Overviews and AI Mode is consistent with this — it explicitly warns against writing content specifically "for AI" rather than for the underlying human need, and states that no special markup or AI-specific file is required to be eligible for citation.
Relation to AI recommendations
In traditional search, getting intent right affects your ranking position. In AI search, it affects whether you appear in the answer at all — a much smaller, more binary outcome. AI assistants break a question down to its underlying need and then draw on sources that resolve that need precisely and at an appropriate length, rather than sources that merely rank well for the same keyword. That's part of why the overlap between AI citations and traditional rankings is so low: one Ahrefs analysis found only around 6-8% of URLs cited by ChatGPT also appear in Google's top 10 for the same query, and roughly 80% of ChatGPT-cited URLs don't rank in Google's top 100 at all. AI citation is a distinct selection process, and reading intent correctly is one of the few levers that transfers between the two.
Example
Take a person who searches "changing tires." The intent is genuinely ambiguous and splits into several needs at once. One driver wants step-by-step instructions to do it themselves. Another wants a nearby garage that can do it for them today. A third just wants to know the legal deadline for switching to winter tires. A garage's website that uses "changing tires" as a heading but serves none of these needs clearly will struggle to be found or cited by either a search engine or an AI assistant. A site that separates a how-to guide, an appointment booking page, and a page on seasonal tire deadlines gives each intent a page built for it — and each one becomes independently findable.
Common questions
How do I figure out the search intent behind a term?
Type the term itself into a search engine and into an AI assistant, and look at what kind of content dominates the results: step-by-step guides, product listings, comparison tables, or a single brand's page. That pattern tells you which need the system has assigned to the query, and you build content to match it rather than guessing.
Is search intent the same thing as a keyword?
No. A keyword is the literal string someone types. Search intent is the goal behind that string. The same keyword can carry different intents depending on who's typing it and in what context, which is exactly why optimizing for the word alone is never enough.