Check this page with an assistantOpens a chat asking it to summarise this article and name the evidence behind each claim.

Claude opens with the prompt on your clipboard: Anthropic does not support prefilled prompts on the web, and we would rather copy it than ship a button that drops it.

The Failure This Explains

A page loads fast, covers the subject thoroughly, has been linked to, and receives almost no impressions for the query it was written for. Nothing in a technical audit will explain it, because nothing is broken. The page is answering a question that the people searching that phrase were not asking.

This is worth naming early because of how much effort gets spent elsewhere first. Teams rewrite titles, chase links, improve speed, add schema, and republish, when the results page for their target query has been quietly showing them a different content type the entire time. Intent mismatch is invisible to every tool that examines your page in isolation, which is most of them.

The diagnosis costs one search. Open the query, look at what ranks, and ask whether your page is the same kind of thing. If the top ten are all product pages and yours is an essay, you have your answer, and no further optimisation of the essay will change it.

The Six Intents Worth Distinguishing

Four classic types plus two that now behave differently enough to track separately. The value of the taxonomy is not the labels; it is that each type implies a different page, and getting the type wrong means building the wrong page well.

IntentQuery shapePage that satisfies it
InformationalWhat is, how do I, why doesA guide that answers directly and early
NavigationalA brand or product name, a loginThe specific destination; rarely worth targeting unless it is yours
CommercialBest, versus, alternatives, reviewA comparison or roundup with criteria stated
TransactionalBuy, hire, book, pricing, near meA service or product page with a clear next step
LocalAnything with a place or proximity impliedA complete profile and location page; see local AI search
GenerativeLong, conversational, several constraints at onceExtractable answers under question-shaped headings

The commercial row is where most small businesses under-invest and most of the money is. Informational queries have the volume, which makes them attractive, and commercial queries have the buyers. A content plan weighted entirely toward the first produces a chart that goes up and a pipeline that does not.

Reading the Results Page as a Brief

The results page is Google's published judgement about what satisfies this query. Treat it as a specification with three parts: the dominant format, the modules present, and the recurring angle across the top results.

  1. Count the page types in the top ten.

    Guides, product pages, category pages, videos, forum threads, tools. If seven of ten are one type, that type is the entry requirement. If the mix is genuinely varied, the query is ambiguous and there is room for more than one approach.

  2. Note which modules appear.

    A map pack means local intent regardless of the words. Shopping results mean transactional. A generated answer at the top means a large share of this query resolves without a click, which changes what winning it is worth, as covered in zero-click searches.

  3. Read the angle, not just the format.

    Are the results aimed at beginners or practitioners? Do they lead with method or with recommendations? Is the framing "best" or "cheapest"? Matching format while missing angle produces a page that looks right and still loses.

  4. Check whether the query is really two queries.

    A results page containing both explainers and product pages is telling you the intent is split. That is an opportunity for two pages, not a mandate for one page that tries to be both.

  5. Do it logged out and unlocalised where possible.

    Your own results are personalised by history and location. A brief drawn from your own logged-in results describes what Google shows you, which is not the same as what it shows your buyer.

Auditing Pages You Already Have

One Search Console pattern reveals most existing mismatches: pages receiving impressions for queries they were not written for. That is Google telling you what it thinks the page is, and where that disagrees with what you intended, you have found an intent problem with data behind it.

Export queries by page and read the top queries for each of your key pages. If a service page is picking up impressions for definitional questions, it is being read as an explainer and will convert badly. If an article is picking up impressions for pricing and hiring queries, there is transactional demand you are catching with the wrong page, and the fix is a new page rather than an edit.

The other signal is high impressions with low click-through. That usually means your title and description promise something different from what the surrounding results promise, so people scan past you. It is a positioning problem rather than a quality problem, and the repair is in titles and meta descriptions rather than in rewriting the page.

Choosing Intent Before You Write

Decide the intent, then the format, then the outline, in that order. Most content briefs invert this: they start from a keyword and a word count, which leaves the most consequential decision on the page to whoever happens to be writing.

A brief that states the intent explicitly changes what gets written. "Commercial intent, comparison format, criteria stated up front, must name when we are the wrong choice" produces a different document from "write 1,500 words about X", and the difference is visible in the first paragraph. If you want a starting structure for a given intent, the outline generator takes the query and produces one.

This also settles the keyword selection question upstream. A keyword whose results page is dominated by publishers with enormous reference profiles is not a good target regardless of volume, and a low-volume commercial query with weak results is frequently the best thing on the list. The demand side of that decision is in keyword research for small business.

Questions People Ask About Search Intent

What is search intent?

Search intent is what a person is actually trying to accomplish when they type a query, as distinct from the words they used. Two people searching the same phrase can want different things, which is why intent is read from the results page Google has already assembled rather than inferred from the keyword itself.

What are the types of search intent?

The four established types are informational, navigational, commercial, and transactional: learning something, finding a specific place, comparing options before deciding, and acting. Two more are worth tracking separately now. Local intent, where the searcher wants somewhere nearby and Google serves a map. And generative intent, where the searcher expects a synthesised answer rather than a list of links.

How do I find the search intent of a keyword?

Search it and read what Google is already showing. The dominant page type in the top results tells you what format wins, the modules present tell you what the query is being treated as, and the recurring angle across results tells you what depth and audience is expected. This is evidence about what satisfied searchers, which no keyword tool's intent label can substitute for.

Why is my page not ranking despite good SEO?

Intent mismatch is the most common explanation for a page that is technically sound and still invisible. If every result for your target query is a comparison page and yours is a how-to guide, you are not competing badly, you are competing in a different category. No amount of speed, links, or optimisation converts a mismatch into a match.

Can one page serve multiple search intents?

Sometimes, and it usually costs you both. A page trying to explain a concept and sell a service tends to satisfy neither reader fully, and the results page will show you whether Google treats those as one query or two. If the top ten mixes formats freely, one page can work. If it is uniformly one type, split them.

Primary Sources

SearchHandled Editorial TeamPublished Nov 26, 2025 · Last reviewed Nov 26, 2025. Every factual claim is checked against the linked primary sources; corrections can be submitted through our contact page.