Search intent is the goal behind a query: the job a person is actually trying to do when they type or speak it. Two people can search the identical phrase and want opposite things, which is why intent is about purpose, not wording. Match it and even a modest page satisfies; miss it and even a brilliant page fails, because it answers a question nobody asked.

This is the single most common reason good content underperforms. The writing is fine, the topic is relevant, the keyword is right there in the title, and it still goes nowhere, because it delivers the wrong kind of answer. Diagnosing intent correctly is the first decision in any content plan, and getting it wrong invalidates everything downstream.

What are the types of search intent?

There are four classic intent types. Most queries lean primarily on one, even when they touch others.

Intent	What the reader wants	Typical signals
Informational	To learn or understand	"what is," "how to," "why," "guide," "examples"
Navigational	A specific site or page	a brand name, a product name, "login," "pricing page"
Commercial	To compare before deciding	"best," "vs," "review," "alternatives," "top"
Transactional	To act, buy, or sign up	"buy," "price," "near me," "discount," "free trial"

Informational queries dominate the open web. The reader wants knowledge: a definition, an explanation, a how-to. The right response teaches clearly and leads with the answer.
Navigational queries are looking for a known destination. You usually win these only when you are the destination; trying to intercept someone's navigational query with a generic page rarely works.
Commercial queries sit just before a decision. The reader is weighing options and wants comparison, evidence, and an honest verdict. Not a hard sell, and not a neutral encyclopedia entry either.
Transactional queries signal readiness to act. The reader wants the path to the action (the product, the price, the button) with minimal friction in the way.

A query like "best project management software" is commercial: the reader wants a vetted shortlist, not a history of project management. "What is project management" is informational. "Asana pricing" is navigational: the reader wants Asana's own pricing page. Same domain, three different jobs, three different right answers.

How do you diagnose intent?

Intent has two readable sources: the query itself, and what the engines already return for it. Use both, and trust the second when they conflict.

Read the query for signals

Start with the words. Signal terms ("how to," "best," "vs," "buy," "near me") flag the likely intent before you look anywhere else. Note the phrasing too: a question implies informational; a comparison implies commercial; an action verb implies transactional. The query is your first hypothesis, not your conclusion.

Read the SERP and the AI answers

Then look at what's actually ranking (and what the AI answer engines surface) for that query. The results are a map of consensus intent, because they reflect aggregated behavior over time. The engine has already learned what satisfies people for this query, and it's showing you:

Format. Are the top results listicles, step-by-step guides, definitions, product pages, videos? That format is the one the engine judges most satisfying. Match it.
Depth. Are they short answers or 3,000-word deep dives? The prevailing length tells you how much the reader expects.
Angle. Are results neutral and educational, or comparative and opinionated? That reveals whether the reader wants to learn or to decide.
Answer-engine treatment. Does an AI answer resolve the query in two sentences, or does it cite several sources for a nuanced topic? That tells you whether you're competing to be the quick answer or to be one cited source among several. (See Generative Engine Optimization.)

When your read of the query and the read of the results disagree, the results win. They encode what real people chose, repeatedly. Your intuition about the phrasing is just a guess against that data.

How do you match format and depth to intent?

Once intent is clear, the format and depth should follow almost mechanically. The job is to give the reader the shape of answer their intent calls for.

Informational → a clear, well-structured explainer or how-to. Lead with the answer, define terms, use headings and lists, go deep enough to be genuinely useful but not padded.
Commercial → a comparison or buying guide. Offer real criteria, honest trade-offs, a table, and a verdict. Hedging on everything is its own kind of failure here; the reader came to decide.
Transactional → a focused page with a clear path to action. Strip friction. The reader doesn't want an essay; they want to proceed.
Navigational → the actual destination, fast and unambiguous. Don't bury what they came for under content they didn't ask for.

Depth is part of the match, not a separate dial. Over-writing a simple query buries the answer the reader wanted in two lines; under-writing a complex one leaves them unsatisfied and sends them back to the results. Calibrate length to the question, not to a word-count target.

Which intent mismatches tank performance?

A mismatch is when your format or depth doesn't fit the intent, and these fail even when the writing is excellent, because relevance is judged by whether the reader's job got done, not by prose quality. The usual culprits:

Selling to a learner. A product or sales page for an informational query. The reader wanted to understand; you tried to convert. They leave.
Lecturing a buyer. A neutral encyclopedia entry for a commercial query. The reader wanted a verdict; you gave them a textbook. They go find the shortlist elsewhere.
Thin answer to a deep need. A 200-word definition for a query whose results are all comprehensive guides. The reader needed depth; you gave them a snippet.
Bloating a quick answer. A 2,000-word essay for a query that wanted one clear sentence. The reader bounces before reaching the point.
Wrong format entirely. A wall of prose where every competitor offers a comparison table or a numbered process. You made the reader work for what others handed over.

The tell is almost always the same: high impressions, poor engagement. People find the page, recognize within seconds it doesn't fit their job, and leave. The fix is rarely better writing; it's the right format for the actual intent.

An intent-diagnosis checklist

Run any target query through this before you commit to a format:

What are the signal words in the query, and what intent do they imply?
Could this phrasing hide a second intent? Which one dominates?
What format do the top results and AI answers use: list, guide, definition, product page?
What depth is the norm: quick answer or comprehensive treatment?
Is the prevailing angle educational or comparative; does the reader want to learn or to decide?
Does your planned format match that consensus, or are you fighting it?
If query and results disagree, are you trusting the results?
Will a reader know within five seconds that this page does their job?

If the format you're planning doesn't survive this check, change the format, not the topic. The topic is fine; the job is what you have to get right.

Where to go next

Intent is the first decision a brief settles, and it shapes everything after it:

Editorial Planning for AI Content: how intent becomes the foundation of a complete brief.
Internal Linking Strategy: how to connect content that serves related intents into a coherent cluster.
Generative Engine Optimization: how matching intent helps your content get surfaced and cited by AI answer engines.

Read the query, read the results, and build the format the reader's job demands. Relevance isn't about covering the topic; it's about finishing the reader's task.

Search Intent: Matching Content to What People Actually Want