Search intent and SERP analysis: work out what the searcher wants before you write

What search intent is

Every query has two layers. The first is the interpretation: what the words refer to. [apple] can mean the company or the fruit. The second is the intent: what the person wants to do with that meaning. Someone who means the company may want the official store, a phone review, the share price or the nearest shop.

Intent matters before writing because it fixes the page type. If people searching [trail running shoes] want to buy, Google fills the page with category and product pages. A 3,000-word article on the same words will struggle there, however good it is. If they want to learn how to pick a pair, the reverse is true and a shop category has little chance.

You cannot read intent reliably from the words alone. Words give you clues. The results page shows you which interpretation and intent Google has found most useful for that query in that country and language. This lesson shows how to read both, and how to write the result down so the whole team builds the right page. Collecting the keywords themselves is covered in the lesson Keyword research: sources and method.

What Google officially says

Google's public description of ranking starts with intent:

The Search Quality Rater Guidelines define the term. User intent is “What the user is trying to accomplish when they search.” The same document warns that one query is rarely one need: “Many queries have more than one meaning.” and “A single user may have many different intents for a single query!”

Google also says the query changes how results are weighed: “The weight applied to each factor varies depending on the nature of your query.” Its own example is freshness: “For example, when searching for current news topics, content freshness plays a bigger role than dictionary definitions.” The page layout changes too: “Based on the user's query the search features that appear on the search results page also change.”

What raters do, and what they do not do

The rater guidelines are the most detailed public text Google has on intent, so SEOs read them closely. They are instructions for people who rate result quality, not a list of ranking factors. Google is explicit about this:

The guidelines themselves say the same: “No single rating can directly impact how a particular webpage, website, or result appears in Google Search”. Use them as a model of what a helpful result looks like for a given intent, not as a checklist that moves rankings.

Needs Met: the rater's measure of intent

Raters judge results on a Needs Met scale. “Needs Met rating tasks ask you to focus on user needs and think about how helpful and satisfying the result is for the users.” Two levels are useful for SEO thinking:

• Fully Meets: “A special rating category, which only applies to queries with clear intent to find one specific result and the corresponding specific result the user is looking for.” In practice this is the official site for a brand query, or the exact page someone named.
• Highly Meets: “A very helpful result for any dominant, common or reasonable minor query interpretation/user intent.” The guidelines add: “There is no length requirement for Highly Meets results.”

The last point matches Google's helpful content advice, which asks whether you are writing to a word count and answers: “(No, we don't.)” Depth should come from the intent, not from a target number of words.

Intent types: Google's rater terms and the SEO classification

The rater guidelines say “It can be helpful to think of queries as having one or more of the following intents.” They list four: Know (including Know Simple), Do, Website and Visit-in-person. SEO teams usually work with five labels instead. The two systems describe the same behaviour from different angles.

Know and Know Simple

“The intent of a Know query is to find information or explore a topic.” Examples: [how does compound interest work], [retirement planning]. A special case: “Know Simple queries seek a very specific answer, like a fact, diagram, etc.” Examples: [how tall is mount everest], [boiling point of water]. Know Simple queries are often answered on the results page itself, so expect fewer clicks and plan for being the cited source rather than the destination.

Do

“The intent of a Do query is to accomplish a goal or engage in an activity.” That covers buying, downloading, signing up, watching, calculating or using an online tool: [buy trail running shoes], [bmi calculator], [download invoice template]. The guidelines admit the edges are soft: “Note that the line between Do and Know queries is sometimes blurry, with some queries fitting into both categories.”

Website

“The intent of a Website query is to locate a specific website or webpage that users have requested.” Brand names, brand plus a section ([example login]) and typed URLs belong here. For your own brand, the right result is your page. For someone else's brand, you will rarely outrank the official site, so do not plan content around it.

Visit-in-person

“People use search engines to look for Visit-in-Person information, such as finding nearby coffee shops, gas stations, ATMs, restaurants, etc.” Google's ranking page says the same in simpler words: “Our systems can also recognize many queries have a local intent.” These SERPs usually show a map and local results, so the winning assets are a business profile and a location page, not an article.

The SEO classification and how it maps

The common SEO labels are informational (learn something), commercial investigation (compare before buying), transactional (act now), navigational (reach a known site) and local (find a place nearby). Commercial investigation has no single rater term: it sits between Know and Do, because the person is still learning but already has a purchase in mind. Queries with best, vs, review or alternatives usually fall here.

Modifiers are useful clues: how to and what is lean informational, best and vs lean commercial, buy, price and order lean transactional, a brand plus login is navigational, near me or a city name points to local. Most keywords have no modifier at all, which is exactly why you need the SERP.

Dominant, common and minor interpretations

When a query is ambiguous, raters sort its meanings by how many people have them in mind:

• Dominant: “The dominant interpretation of a query is what most users mean when they type the query.” But “Not all queries have a dominant interpretation.”
• Common: “A common interpretation of a query is what many or some users mean when they type a query.” A query can have several.
• Minor: meanings that fewer people have. Reasonable minor ones still deserve a place in the results; unlikely ones do not.

The guidelines' own example is [apple] in English (US): the computer company is the dominant interpretation, the fruit a common one, and a city named Apple an unlikely minor one. For [mercury] they show no dominant meaning at all: the planet and the chemical element are both common interpretations. Brand queries can also split by intent rather than meaning. The guidelines put it plainly: “Many queries have more than one likely user intent.” For a big retailer, some people want the nearest store, others the online shop, others company news.

What this means for you:

• Clear dominant interpretation: build one page for it. Do not dilute it to cover minor meanings.
• Several common interpretations: decide which one your business serves. If you sell thermometers, the [mercury] SERP is not your market, however big the volume.
• Your meaning is minor: the head term is a poor target. Add the word that removes the ambiguity ([mercury thermometer disposal]) and target that instead.

Mixed and fractured intent

A SERP has mixed intent when Google shows different page types for one query: some product categories and some buying guides, or some local results and some national articles. It is Google hedging because different people want different things. A fractured SERP is the extreme case, with several unrelated meanings each getting one or two results.

How to decide what to build:

1. Count the page types. As a working rule, when most of the top 10 are one type, that is the dominant intent. When the split is close to even, the intent is mixed.
2. Serve the dominant intent with the main content. For the query in the figure, that is a category page at https://example.com/trail-running-shoes/ with filters, prices and stock.
3. Cover a common second intent lightly on the same page if it is small: a short “How to choose” block under the product grid.
4. Make a second page if the second intent is large and needs a different page type: a guide at https://example.com/guides/best-trail-running-shoes/ that links to the category. Target each page at its own wording so they do not compete (the lesson Keyword cannibalisation covers this).
5. Do not merge a shop and a long guide in one main content area. The page ends up weaker than the pure category pages and the pure guides it is competing with.

Freshness intent

Some queries need current information. The rater guidelines name four kinds: breaking news ([tornado]), recurring events such as elections, sports seasons and tax forms, current information such as prices and populations, and product queries where people want the latest model. For these, the rule is strict: “When a query demands recent content, only pages with current, recent, or updated content should get high Needs Met ratings.”

Meaning itself moves with time. Raters are told: “Remember to think about the query and its current meaning as you are rating.” For [iphone] the guidelines note: “Most users now are looking for the most recent or upcoming iPhone model.” A page about an old model can be accurate and still be the wrong answer.

How to spot freshness intent in a SERP: visible dates in snippets, a Top stories block, years in titles, and results that are mostly recent. How to serve it:

• Update facts, prices, versions and screenshots for real, then show the updated date.
• For recurring topics, keep one evergreen URL (https://example.com/guides/tax-deadlines/) and update it each year instead of publishing a new URL every year. Links and history stay on one page.
• Do not fake it. Google's helpful content page asks: “Are you changing the date of pages to make them seem fresh when the content has not substantially changed?”

Many queries need no freshness at all. [how does a heat pump work] does not change every year; spend your updating effort where the SERP shows it matters.

How to analyse a SERP step by step

SERP analysis takes a few minutes per keyword and saves you from building the wrong page. Do it for every keyword that will get its own page, and record the result.

1. Set the locale. Search in the country and language you target. Results for the same words differ between countries and cities. Use a private window without signing in to limit personalisation; for local queries, check from the city you care about.
2. Note the page types in the top 10: category, product, service page, guide, comparison, tool, video, forum thread, local listing.
3. Note the formats: step-by-step how-to, list, comparison table, review, calculator, template.
4. Note the angle: for beginners, for small businesses, cheapest, fastest, this year, for a specific country.
5. Judge the depth. Short answers or full guides? Open the top three and skim how much they cover.
6. Check freshness: dates in snippets and on the pages, years in titles.
7. List the SERP features: featured snippet, People also ask, local pack, images, videos, shopping results, Top stories, AI features. Each one tells you what format people engage with.
8. Note who ranks: large brands, marketplaces, publishers, niche sites, forums. A SERP full of small niche sites is easier to enter than one held entirely by household names.
9. Write a one-line verdict, for example: “Category pages with filters; buyers in the UK; prices visible; no freshness need.”

Two cautions from the rater guidelines apply to you as much as to raters. First: “Do not assign a high rating to a webpage just because it appears at the top of a list of search results on Google.” The SERP shows what Google rewards today, not the best possible page; your aim is to meet the need better, not to copy the top result. Second: “Just because a result matches several key words in the query does not mean it is relevant or helpful to the user.” Repeating the keyword is not matching the intent.

A useful final test, from Google's helpful content questions: “Does the content provide substantial value when compared to other pages in search results?”

Matching the page type to the intent

Page type comes first. Quality, depth and links decide the order among pages of the right type, but a page of the wrong type is usually not in the contest. The table shows typical matches; confirm each one against the real SERP before you build.

Matching is not only about the type. Within the type, match the format (a comparison wants a table), the angle (a “for small house” query wants that audience in the title and the content) and the depth. Google's helpful content page frames the goal as a question: “After reading your content, will someone leave feeling they've learned enough about a topic to help achieve their goal?” The opposite warning sign: “Does your content leave readers feeling like they need to search again to get better information from other sources?”

When intent shifts over time

The SERP you analysed is a snapshot. Intent shifts for predictable and unpredictable reasons:

• New versions. Product and software names start to mean the latest release, as in the [iphone] example.
• Seasons and events. Gift, holiday, tax and sports queries change before, during and after their season.
• News. A brand or topic in the news can turn an informational SERP into a Top stories SERP for a while.
• Google's own changes. After ranking updates or new SERP features, Google may favour a different page type, for example forum threads or videos where articles used to rank.

Re-check the SERP for your priority keywords on a regular schedule (quarterly works for most sites) and every time a page loses positions without a technical cause. If the page types in the top 10 have changed, the fix is usually a different page type or a rewritten angle, not more links or more words.

Recording intent in your keyword sheet

Intent that lives only in someone's head gets lost. Add intent columns to the keyword sheet next to volume, so every target URL has a reason behind it. A practical set of columns: keyword, volume, rater intent, SEO intent, dominant page type, format, angle, freshness need (none, low or high), SERP features, who ranks, target URL, verdict.

Example rows on example.com. With intent recorded, two later jobs become simple: grouping keywords that share one page type and intent into one URL (the lesson Clustering and the content map), and spotting two pages built for the same intent before they compete.

Common mistakes

• ✕Guessing intent from the wordsModifiers are clues. Most keywords have none, and the SERP is the proof.
• ✕A blog post for a shopping SERPIf the top 10 are categories, an article is the wrong page type, however good it is.
• ✕One page for two incompatible intentsA shop grid and a long guide in one main content area lose to pure versions of each.
• ✕Copying the top resultMatch the type and format, then meet the need better. Ranking first is not proof of being best.
• ✕Checking the wrong localeA SERP from another country or city can point to a different intent.
• ✕Ignoring SERP featuresA map, a video row or a shopping block tells you what people use for that query.
• ✕Treating rater guidelines as ranking factorsRaters do not control rankings; the guidelines describe helpful results.
• ✕Fake freshnessA new date on unchanged content. Update the substance or leave the date alone.
• ✕Writing to a word countGoogle has no preferred word count. Depth should follow the intent.
• ✕Analysing once and never againIntent shifts. Re-check priority SERPs on a schedule and after drops.

FAQ

Sources

• Google — Search Quality Rater Guidelines (PDF): Understanding user intent, Needs Met, Freshness
• Google — How Search works: ranking results
• Google Search Central — In-depth guide to how Google Search works
• Google Search Central — Creating helpful, reliable, people-first content

Quotes checked against the official pages on 30 October 2026.