1/11/2026
Search intent types classify each query by what the user wants to do: learn, reach a brand, compare or act. This classification decides the page format, the CTA and the KPIs to track. This page covers the four intent types: informational, navigational, commercial, transactional, with their signals in the query and in the SERP, the modifier rule, mixed intent, B2B micro-intents and 20 annotated examples. For the definition, the stakes and the identification method, start with understanding search intent.
The 4 Standard Categories: Signals, SERP, Format and KPIs
Four categories cover the needs of a B2B content plan. Each one is recognised by signals in the query, confirmed by SERP formats, then calls for its own page format, CTA and KPIs. Use the table as your reference grid:
Informational: Get an Answer, a Definition or a Method
The user wants to frame a topic or follow a method. In B2B, the prospect is trying to understand the problem before looking for a solution. The page opens with a direct answer, then expands with steps, examples and a mini FAQ, with a soft CTA. More detail in the signals and formats of informational intent.
Navigational: Reach a Site, a Brand or a Specific Page
The user knows where they want to go: log-in page, help centre, documentation. A brand name combined with an access modifier indicates an immediate expectation. The page stays short and frictionless; a clear architecture helps you earn sitelinks. A brand-only search may also be looking for an evaluation: the SERP decides. See navigational intent and the branded SERP.
Commercial (Investigation): Compare, Evaluate, Shortlist
The user has identified a need and is comparing options: they expect criteria, limitations and proof. A constraint ("for SMEs", "GDPR-compliant") indicates multi-criteria selection. The page structures the comparison in a table and offers an evaluation CTA (demo, pricing) rather than a direct purchase. See comparisons, alternatives and reviews: commercial intent.
Transactional: Take Action (Purchase, Request, Sign Up)
The user wants to act: request a demo or a quote, sign up for a trial. In B2B, the transaction is most often a commercial contact. The page focuses on a single primary CTA, the terms, reassurance and an objections FAQ, without a long introduction. See the pages that serve transactional intent.
The Modifier Rule: The Brand and the SERP Decide
A modifier points the classification in a direction without settling it alone: the same word changes type depending on whether the query names a brand, and depending on the SERP. The grid sets a single rule to apply to every query:
This grid ends contradictory classifications. "Pricing" is only transactional when the user already knows which offer they are looking at; without a brand, they are still comparing. "Reviews" stays commercial even with a brand: the user is evaluating, not trying to log in.
The SERP Settles Borderline Cases
When the grid leaves room for doubt, the top ten results give the answer:
- Standalone pricing pages, product pages, ads: transactional intent.
- Comparisons, "best" lists, "vs" pages: commercial intent.
- Featured snippet, People also ask, AI Overview: informational intent.
- Brand homepage with sitelinks: navigational intent.
- A blend of guides and product pages: mixed intent.
Location Constraints
A city or "near me" adds a local dimension to the type: 46% of Google searches have local intent (Webnyxt, 2026). "SEO training Lyon" stays transactional, but the SERP shows local listings: the page then carries the address, the area served and a direct contact. More benchmarks in the key SEO statistics.
Beyond the 4 Categories: Mixed Intent and Dominant Intent
A mixed intent carries two needs at once, for example understanding and comparing. Rather than a hybrid page, pick the dominant intent, the one the SERP mostly rewards, and handle secondary intents via short sections and internal links. Three signals reveal a mixed intent:
- A scattered SERP: guides, comparisons and product pages share the top 10.
- Heterogeneous People also ask questions: definitions and choice questions ("how to choose", "how much") sit side by side.
- A short query with no modifier or constraint: "AI SEO tool", "CRM for SMEs".
Dominant Intent vs Secondary Intents
Stacking a definition, a comparison and a form on one page dilutes the promise. The page carries the primary promise, then opens onto secondary needs via a contextual link to the comparison or the demo page. It thus keeps a consistent CTA and a measurable KPI. When CTAs become contradictory ("download the guide" and "request a quote"), split the content into two pages.
Ambiguous Short Queries: When the Same Keyword Flips From One Expectation to Another
Long queries carry their own clues: 70% of searches contain more than three words (SEO.com, 2026). Short queries are the exception, and the hardest to classify: "AI SEO tool" may be looking for a definition, a comparison or a product page. Note the dominant formats in the top 10: a homogeneous SERP dictates the format, a heterogeneous SERP justifies two or three distinct pages linked to each other.
B2B Micro-Intents: A Finer Breakdown for Long Cycles
In B2B, the buying journey involves multiple roles and is rarely linear. Micro-intents refine the four types without replacing them: they describe the buyer's stage and the proof they expect:
- Discovery: frame the problem. Guides, glossaries; expected proof: the method.
- Validation: gain confidence in feasibility. Technical pages; expected proof: the usage context.
- Evaluation: compare. Comparisons; expected proof: criteria and limitations.
- Decision: take action. Demo, pricing and quote pages; expected proof: answers to objections.
Technical stakeholders want documentation and integrations during validation, buyers want terms and timelines at decision stage: a single topic may therefore require several pages. Start with the four types, then add micro-intents when your sales cycle demands different proof at each stage.
20 Search Intent Examples, 5 per Type
Each example follows the formula "the user wants… in order to…": rephrasing the query as a need prevents off-target content. If the "in order to…" stays vague, the query is ambiguous and the SERP decides. The examples follow a fictional B2B SEO content-optimisation software vendor, complemented by generic queries:
One Topic, Four Pages: The Running Example
The topic "SEO content optimisation" does not produce one ultimate page, but four distinct pages, each built on its dominant intent:
- Informational: "how to optimize content for seo"; method guide, CTA "download the checklist".
- Commercial: "seo platform comparison"; criteria-based comparison, CTA to the demo.
- Transactional: "seo platform demo"; demo page, CTA "book a slot".
- Navigational: "incremys login"; log-in page, CTA "log in".
Internal linking connects the pages in journey order: from the guide to the comparison, from the comparison to the demo, from the log-in page to the help centre. Two pages never target the same query with the same intent, which prevents cannibalisation.
When "Price" Is Used to Compare, Not to Buy
"SEO software price" looks like a buying query. If the SERP shows standalone pricing pages, create a pricing page; if it shows comparisons, a criteria table that includes price answers it better. A brand-only search follows the same logic: it may target the customer account or discovery of the offer, and the homepage then gives access to log-in, support and the demo.
Classifying Your Own Queries Without Getting It Wrong
The three-check method (wording, SERP, data) is detailed in the main guide. Classifying a real list adds two traps: averages that hide different behaviours, and trigger words taken at face value.
Segments to Isolate Before Classifying
The same page can serve different intents depending on the segment. Before drawing conclusions about a query, separate at least:
- Branded and non-branded: branded traffic inflates overall CTR.
- Device: mobile may favour short formats.
- Country: expectations and SERPs differ from one market to another.
- New and returning visitors: returning visitors more often have navigational intent.
The Performance report in Google Search Console filters queries and pages by device and by country, which prevents you from over-interpreting an average.
Trigger-Word Mistakes
Classification errors come from a shortcut taken on the query:
- Classifying on one word without checking the SERP: "tool" may call for a definition or a benchmark.
- Confusing intent with journey stage: a contact form on a discovery page slows users down, a long guide on a pricing page delays action.
- Automatically classifying a brand-only search as navigational: it may be looking for an evaluation.
- Freezing the classification: when the SERP changes formats, the intent changes with it.
Reading Classification Signals
Some signals show that a classification needs revisiting:
Which Types Do Generative Engines Cite?
At the time of writing, ChatGPT, Perplexity, Gemini, AI Overviews and AI Mode (available in France since summer 2026) mainly answer informational and comparison queries. Yet each type expects something specific from a page before it gets reused:
- Informational: a direct answer at the top, explicit headings, lists and named entities.
- Commercial: criteria tables and a visible selection method.
- Transactional: extractable information (scope, terms, steps) in a table or a mini FAQ.
- Navigational: consistent brand pages that point to the right page.
The AI Overview is also a SERP format: its presence signals informational or commercial intent and pressure on organic CTR. A direct answer and citability then matter as much as position. Correlation is not causation: a cited page is not necessarily the one that converts.
Depending on Your Situation
Once a query's type is settled, the next step depends on your starting point:
- You start from a keyword list and want to know which ones to tackle first: choosing high-intent keywords.
- You are organising pages of different types around the same topic: building a semantic cocoon.
- You are writing the page once the type is settled: the rules of SEO copywriting.
- You want to check a classification through queries, CTR and position: making the most of Google Search Console.
To classify your queries by intent and spot the pages to create or realign, the Incremys GEO & SEO platform offers opportunity analysis: keywords, intents and pages to create.
FAQ
The most frequent questions about classifying queries, with a short answer.
How Many Search Intent Types Are There?
Four: informational (understand), navigational (reach a brand or a page), commercial (compare before acting) and transactional (act: demo, quote, sign-up). Mixed intents come on top, when a query carries two needs: you then answer the dominant intent and open onto the secondary one. Micro-intents refine this framework without creating a fifth type.
Commercial or Transactional: How Do You Decide?
Through the SERP. If the top results are comparisons, "best" lists or "vs" pages, the user is still evaluating: the intent is commercial. If they show demo, pricing or quote pages and ads, the user wants to act: the intent is transactional. In borderline cases, keep one dominant intent and connect the page to its neighbour with a contextual link.
Is "Pricing" a Commercial Query?
Without a brand, yes: "seo software price" is used to compare price levels, and a comparison that includes price as a criterion answers it. With a named brand, the query becomes navigational or transactional: the user wants this offer's pricing page. The SERP confirms the classification: standalone pricing pages on one side, comparison lists on the other.
What Should You Do With a Query That Has Two Intents?
Identify the dominant intent in the SERP and build the page around it. Cover the secondary intent with a short section and a link to a dedicated page. If the two needs call for contradictory CTAs, for example a guide to download and a quote request, create two distinct pages and connect them with a contextual link.
Do Micro-Intents Replace the 4 Types?
No. The four types describe what the user wants to do; micro-intents (discovery, validation, evaluation, decision) describe the stage they have reached in a B2B cycle. They help you choose a page's proof and CTA once the type is settled. Start with the four types, then add this breakdown when several roles take part in the decision.
How Do You Classify 500 Queries Quickly?
Start with modifiers: a first sort by word (how, comparison, demo, brand name) classifies a large part of the list. Then separate branded and non-branded queries. Keep SERP reading for short, ambiguous or high-stakes queries, then check the result against the CTR and position recorded in Google Search Console.
Does Intent Change Over Time?
Yes. SERPs evolve: comparisons replace guides, an AI Overview appears, new players start ranking. A query classified as informational can become commercial as a market matures. Re-read the SERP as soon as a page loses CTR or position stability, and reclassify the query if the dominant formats have changed.
.png)
.jpeg)

.jpeg)
%2520-%2520blue.jpeg)
.avif)