Search intent types every marketer should know (and how to actually use them)
Open your keyword tool. Type in a seed term. You get 400 rows back, sorted by volume, and about a third of them are questions people will never convert on, a third are people looking for your homepage, and a third are genuinely worth your time. Most marketers sort by volume and start writing. That is the mistake that eats budgets.
A query is not a keyword. It is a person, mid-task, typing a fragment of a thought into a box. The job of search intent analysis is to reverse-engineer that thought before you write a single word. Get it right and a page can hold position 3 for two years. Get it wrong and you rank, get clicks, and watch people bounce in eight seconds because you gave them a definition when they wanted a price.
I have made this specific mistake more times than I want to count. The worst one: a 2,400-word comparison page targeting a phrase that looked commercial but was actually informational. It ranked. It got 1,100 visits in a month. It produced three leads. I rewrote the same page into a short, blunt buying guide with prices and it converted at roughly eight times the rate on a fraction of the traffic. Same topic. Different intent. Completely different outcome.
Key takeaways
- Search intent is the reason behind a query, not the query itself.
- The four canonical types are informational, navigational, commercial, and transactional.
- The three C's (content type, content format, content angle) are the practical checks that keep you aligned with intent.
- Commercial intent is where the money usually is, and where most teams under-invest.
- Generative AI intent is the newest type, and it behaves differently from the classic four.
- Intent maps to funnel stage, so your content plan should follow the same order.
What search intent actually means (and why the definition is not enough)
Search intent is the underlying goal behind a search query. Also called user intent, audience intent, or query intent. That is the textbook answer, and it is technically correct, and it tells you almost nothing you can act on.
Here is what actually matters. A query is a compressed signal. Someone typing "best crm for small business" has already done a category search, already read one or two listicles, and is now trying to narrow a shortlist. Someone typing "what is a crm" is at the very start and does not yet know what problem they have. Same product. Two entirely different pages required.
Why keyword volume misleads you
Volume measures how many people ask a question. It says nothing about what they want back. A 40,000-search phrase can be worth less than a 900-search phrase if the second one carries buying signals.
I keep a running note of every query our site gets that led to a demo request. Almost none of them are the high-volume terms. They are longer, uglier, more specific. Specificity is the tell. When someone adds three qualifiers to a query, they are close to deciding.
Intent is not the same as topic
Two pages can cover the same topic and need completely different structures because of intent. A guide answering "how does X work" needs explanation, diagrams, and patience. A page answering "X pricing" needs a number in the first screen. Reusing one piece of content for both is how sites end up ranking well and converting poorly.
What are the different types of search intents? The four main types in SEO
Strip away the marketing language and there are four. Everything else is a variation.
| Type | What the searcher wants | Example query | Content that wins |
|---|---|---|---|
| Informational | To learn or understand something | "how does search intent work" | Guides, explainers, definitions |
| Navigational | To reach a specific site or page | "HubSpot login" | Your homepage or a clean brand page |
| Commercial | To compare options before buying | "best crm for small business" | Comparisons, alternatives pages, reviews |
| Transactional | To complete an action now | "buy crm subscription" | Product pages, pricing, signup flows |
The temptation is to treat this as a taxonomy exercise. It is not. It is a routing problem. Each type sends the visitor to a different part of your site, and mixing them up within a single page is what produces the classic "ranks but does not convert" pattern.
Commercial search intent deserves its own paragraph
Commercial intent is the most valuable and the most commonly botched. These searchers are not ready to buy and not clueless either. They are in the comparison phase, and they want to be shown differences, not benefits.
If you answer a commercial query with a generic feature list, you lose. If you answer it with an honest comparison that names a competitor and explains when that competitor is the better choice, you win trust fast. Counter-intuitive, but it works. Our best-performing page on our own site is a comparison piece that openly says a rival is a better fit for one specific use case.
Navigational search intent is not a waste
Plenty of marketers dismiss navigational queries because the traffic does not look new. That is a mistake. Someone searching your brand name plus "login" or "pricing" is already aware of you, and the friction is purely technical. If a competitor buys ads on your brand term, you have a problem you can measure and fix. Branded search is often the cleanest signal that everything upstream is working.
What are the three C's of search intent?
The three C's are content type, content format, and content angle. Together they form the practical test for whether a page matches the intent behind its target query.
Content type is what the piece fundamentally is: a definition, a list, a comparison, a tutorial, a review. Format is how it is presented: video, long-form text, checklist, calculator, image gallery. Angle is the specific viewpoint or promise you bring, which is what separates your page from the nine other results answering the same question.
- Type answers "what is this page for?"
- Format answers "how is it consumed?"
- Angle answers "why would someone pick this one over the others?"
The reason the three C's matter is that most intent mismatches are not conceptual, they are format mismatches. You understood the intent correctly and then delivered the wrong shape of content. A three-minute video when the searcher wants a table of prices. A 4,000-word essay when the searcher wants a template they can copy in 30 seconds.
How to read intent off the SERP in under two minutes
You do not need a tool for the first pass. Search the query, ignore the ads, and look at what the organic results have in common.
- Note the dominant format. Are seven of ten results lists? Then lists win.
- Check the freshness. If most results are recent, the topic moves fast and old content will not hold.
- Look for the feature blocks. A shopping carousel on page one means transactional. A featured snippet means informational.
- Read two results closely. What do they promise in the first three lines?
That gives you the type, the format, and the angle most competitors are using. Then you decide whether to match the pattern or break it deliberately. Matching is safer. Breaking it only works if you can genuinely be more useful, and occasionally that pays off enormously.
What is the 80/20 rule in SEO?
The 80/20 rule in SEO is the observation that a small share of your pages drives the large share of your organic traffic. Roughly, a fifth of your content generates four-fifths of the results, while the rest sits there doing very little.
The mechanism is not mysterious. Most published pages are either chasing volume without intent, duplicating a topic already covered, or answering a query nobody actually cares about. A handful of pages hit a real need at the right moment, and they carry the site.
The practical response is not "publish less." It is "audit more." Every few months I pull our top pages by organic sessions and top pages by assisted conversions. The overlap between those two lists is usually smaller than people expect, and that gap is where the money is hiding. Pages with traffic but no conversions almost always have an intent mismatch, and fixing an existing page is far cheaper than writing a new one.
Applying it without getting precious about it
Sort your content into three buckets: pages that bring traffic and convert, pages that bring traffic but do not convert, and pages that do neither. Rewrite the middle bucket first. Delete or consolidate the third. Leave the first alone unless something is clearly broken.
I spent about two weeks on this once across roughly 90 pages and moved our conversion rate from around 1.4% to just under 2.6%. Nothing was published. Nothing was rewritten from scratch. Pages were reordered, retitled, and re-angled to match what searchers actually wanted. That is the entire trick.
Local intent: the type that does not fit the four
Searchers looking for something near them behave differently enough to warrant separate treatment. Queries with "near me," a city name, or a postcode carry strong urgency and almost zero patience. They want hours, an address, a phone number, and directions. Long-form content barely registers.
If you serve multiple locations, each one needs its own page with genuinely distinct information. Duplicated location pages with the city name swapped are easy to spot and rarely rank. The businesses that do this well have real differences between pages: different staff, different stock, different reviews.
Generative AI intent: the newest type
Here is where it gets interesting for anyone reading this in 2026. A growing share of searches are not people looking for a page at all. They are people asking an AI assistant a question and expecting a synthesized answer, not a link.
This behaves like its own intent type. The searcher has no patience for navigation. They are not comparing brands by clicking between tabs. They want a direct answer, with the option to verify. That changes what wins.
What this means for how you write
Three things shift. First, clear factual statements beat rhetorical build-up, because a synthesis engine rewards statements it can lift cleanly. Second, structured data and clean headings matter more, since machine readers lean on them. Third, brand mentions across the web carry more weight than a single high-ranking page, because being cited is the new visibility.
I am not going to pretend I have this fully figured out. Nobody does yet. But the teams that treat this as a sixth intent type and adapt their formatting accordingly will be ahead of the ones still writing for ten blue links only.
Mapping intent to funnel stage
Intent and funnel stage are two views of the same thing. Informational queries sit at the top. Commercial queries sit in the middle, where the decision is being made. Transactional queries sit at the bottom, and navigational queries run alongside the whole thing for people who already know you.
The gap most content plans have is in the middle. Teams produce plenty of top-of-funnel explainers because they are easy to write and plenty of bottom-of-funnel product pages because they already exist. The comparison and evaluation content in between is where a decision gets made, and it is usually thin.
A quick audit you can run this week
- Pick your ten highest-traffic pages.
- Write down the intent type you think each one targets.
- Search the query it targets and note the dominant format on page one.
- Flag every page where your format and the SERP's format disagree.
- Rewrite the worst offender first.
One page, one afternoon. That is enough to see whether the problem is real on your site, and on most sites it is.
The part tools cannot do
Intent classification tools get you about 70% of the way. They are useful for sorting a thousand keywords into buckets. They cannot tell you whether the searcher wants reassurance or a number, and that distinction decides whether your page works.
That judgment comes from looking at what actually ranks, reading the queries that land on your site, and paying attention to what people ask in the sales process. The most reliable intent data I have ever collected did not come from a tool. It came from reading 200 support tickets and noticing that half of them were asking a question our top-ranked page answered in paragraph nine.
Rank for the query. Answer the person. The gap between those two things is where almost all underperforming content lives, and closing it is less about writing better and more about listening more carefully to what was asked.