User Intent & Query Types
Search intent is the job a person is trying to get done when they type a query: to learn something, to reach a specific site, to choose, or to buy. A page reaches the top when it does that job, not when it repeats the words of the query. This lesson covers the query types a user intent model is built on: four base ones plus the fifth and most profitable sub-type, and how to read intent off the results page.
That Google judges the meaning of a query rather than word matching is set out in the guide to ranking systems and the ranking section of «How Google Search works».
Search intent is the task a person opened Google for. Not the query itself — the query is just the words they pick to describe the task. Google tries to read the task behind the words and pick a page that solves it.
Why intent is the master filter
If your page's intent doesn't match the query's intent, no amount of technical SEO will save it. That's a hard ranking ceiling. A few common mismatches:
| Query | User intent | Common mistake |
|---|---|---|
| buy nike sneakers | Transactional — needs a catalog | Writing a blog review that won't rank |
| what is EEAT | Informational — needs a guide | Building a landing page for "EEAT services" |
| best CRM 2026 | Comparison — needs a roundup | Pointing it at one product page |
How Google frames the need behind a query is visible in its own instructions for raters: the Search Quality Rater Guidelines, in the Needs Met section. It also explains why a formally relevant page scores low when it does not do the job the query implies. The requirements for people-first content are in the helpful content guidance.
The 4 base intent types
| Type | What the user wants | Trigger words |
|---|---|---|
| 🧭 Navigational | Reach a specific site. Already knows where they're going | brand name, login, sign in, app store |
| 📚 Informational | Learn, understand, figure out. No purchase | what is, how, why, guide, tutorial |
| 🛒 Transactional | Take action: buy, download, sign up | buy, order, download, price, shipping |
| 📍 Local | Find a place nearby or in a specific city | near me, in [city], address, on map |
The 5th critical sub-type: commercial investigation
Between "informational" and "transactional" sits a huge class of queries that's worth treating as its own bucket: commercial investigation. The user wants to buy — but is still choosing.
| Sub-type | Example query | What the page should do |
|---|---|---|
| Best-of category | "best wireless headphones 2026" | Ranked roundup with reasoning |
| Comparison | "Notion vs Obsidian" | Side-by-side comparison table |
| Review | "iPhone 17 Pro review" | In-depth review with pros/cons |
| Alternatives | "alternatives to Photoshop" | List with pricing and features |
These queries are especially valuable: the user is already close to buying, and competition on commercial roundups is usually lower than on raw "buy X" terms.
The query is the tip of the iceberg. Below it is a pyramid of needs
The same query can hide very different expectations. The classic example is a query like "Ford Focus":
👤 One person wants the hatchback. Another wants the sedan. A third wants the electric version. A fourth cares about mileage. A fifth cares about leather seats and seat warmers. A sixth is just reading Wikipedia.
All typed the same two words. All want different things.
In situations like this Google produces a hybrid SERP: model catalog + Wikipedia + comparisons + video reviews + local dealers. That's the signal that intent on this query is mixed. The ideal page is either very broad (Wikipedia-style — covers ~40% of intent) or very narrow ("Ford Focus 2024 electric review") — surgically targeting one sub-segment.
Messy middle: the journey is not linear
Thinking "user is informational first, then transactional later" oversimplifies things. In reality, between "I have a need" and "I bought" the person bounces around in a loop:
Between exploration and evaluation the user ping-pongs: read three reviews → compared two models → went back to read about the category → checked some user reviews → bounced back to comparisons. Google calls this stretch the messy middle.
Takeaway: a strong SEO strategy covers queries at EVERY stage, not just transactional. Informational content builds trust — research consistently shows that users who first read useful content on a site are significantly more likely to come back and buy.
How to detect intent in 60 seconds (working process)
- Type the query into Google in incognito mode (no personalization)
- Read the SERP features:
- Map pack → local intent
- Shopping carousel / products at the top → transactional
- Featured snippet with a definition → informational
- Knowledge panel on the right → navigational/encyclopedic
- Video carousel near the top → process intent ("how to...")
- Look at the top 10 organic results: what are they? Catalogs, articles, reviews, comparisons, landing pages?
- If the top is one consistent type → intent is clear, build the same kind of page.
If it's a mix → intent is hybrid, pick a sub-genre and fight for it.
Intent → page type (cheat sheet)
| Intent | Right page type |
|---|---|
| Informational ("what is X") | Guide, explanatory article, FAQ, glossary entry |
| Informational process ("how to X") | Step-by-step tutorial with screenshots/video |
| Commercial (comparison) | Roundup ranking, "X vs Y" page |
| Transactional (product) | Product / service page, landing page |
| Transactional (category) | Category / collection page with filters |
| Local | Location page + Google Business Profile |
| Navigational (own brand) | Homepage or relevant section |
🔥 2026 lifehacks: what actually works for intent today
The 4-type intent classification is foundational — and 25+ years old. Between 2024 and 2026, new tools and habits reshaped day-to-day intent work in ways the textbooks haven't caught up with yet:
1. AI Overview is the ground truth on intent
When Google shows an AI Overview at the top of the SERP, it's literally telling you "this is how I read this query." Use it:
- Expand the AI Overview and look at who it cites. Those are the "right" page formats for the query. Cites Wikipedia + 3 articles → informational. Cites Reddit threads → wants opinions. Cites product pages → commercial.
- Read the sub-questions it answers. That's a free sub-intent map — cover them as H2/H3s on your own page.
- If there's NO AI Overview, that's also a signal: the query is either purely transactional or Google didn't find enough trustworthy sources. That's an opportunity window.
2. SERP feature fingerprint (extended map)
| What you see in the SERP | What it tells you about intent |
|---|---|
| Reddit / forums in the top 10 | "I want real human opinions, not marketing." Add direct quotes, user experience, real cases |
| AI Overview at the top | Informational/research intent. The goal shifts: get cited, not just rank in the top 10 |
| Video carousel in top 3 | Process intent. Text alone won't close it — you need a video or embedded YouTube |
| Shopping / product carousel | Transactional. An article won't get in — you need Merchant Center / a filtered category |
| Map pack with no organic results above | Pure local. You need Google Business Profile, not a blog |
| PAA + related searches block | Hybrid/broad — user isn't sure what they want. Cluster page with sections |
3. People Also Ask = a free sub-intent map
Below the main results, expand 5–7 PAA questions (they lazy-load — keep clicking to reveal more). These are the questions Google considers related to your query. The 2026 top-results pattern:
- Main query → in the title and H1
- 5–7 PAA questions → as H2/H3s in the same article
- Each opened with a short direct answer + expanded context
This pattern almost guarantees cluster ranking: one page lands in the top 10 for 10–20 related queries simultaneously, instead of just one.
4. GSC "Search Appearance" — your intent-fit X-ray
In Google Search Console: Performance → Search Appearance. You'll see breakdowns by feature type — AI Overviews, Featured Snippet, Web Stories, Recipes, Videos. This shows which intent Google associates with your pages. If a page is supposed to be transactional but it shows up in Featured Snippets with a definition, Google reads it as informational. Either restructure it or build a separate transactional version.
5. GEO (Generative Engine Optimization) — the new discipline
ChatGPT Search, Perplexity, Claude, and other LLM-based search engines read the web differently from Google. To get cited in AI answers, format your content for their patterns:
- Specific numbers and statistics with source and year — LLMs love pulling these out
- Direct answer in the first paragraph (TL;DR / featured-snippet style)
- Numbered lists instead of bullet markers — they parse better
- Author byline with expertise (E-E-A-T in its purest form)
- Freshness: publish date and last-updated date — critical for being included in AI answers
Lifehack: run your top 10 pages through Perplexity and ChatGPT Search on their head queries. Cited? Great. Not cited? Look at who is, and reformat to match their structure.
6. Reddit primacy since February 2024
After the Google ↔ Reddit deal ($60M/year for training data), the forum jumped sharply in rankings for queries like "best X", "honest review", "is X worth it", "real experience with X". If Reddit is in the top 3 for your topic, the user wants a human voice — not a vendor's landing page. Tactics:
- Quote real community discussions with attribution
- Run a topical subreddit yourself or via a community manager — Reddit positions feed back into Google positions
- A page with a real author bio outperforms anonymous corporate copy
7. Seasonal intent shifts
The same query changes intent across the year. "iPhone" in September = new model (info/commercial); in March = repair or trade-in (service). "Taxes" in April = transactional (filing); in September = informational. Once a quarter, re-check the SERPs for your top queries — intent can shift quietly, and a page that ranked for six months may start bleeding positions.
- Open the query in Google in incognito
- Is there an AI Overview? If yes — note who it cites and which sub-questions it covers
- Which SERP features sit in the top 3 (video / shopping / map / Reddit)?
- Expand all PAA questions — that's your H2/H3 outline
- Reddit/forum in the top 10? If yes — add a "human" layer of content
- What page type dominates the top 10: catalog / article / comparison / landing?
Practice: classify the intent of your own queries
Do not guess intent at your desk — read it off the results page and out of your own data. Five steps:
- In Search Console, open the performance report and write down the 20 queries with the most impressions.
- Label each one: informational, navigational, commercial, transactional or local.
- Check which of your pages shows for each query. An article showing for a transactional query is an intent mismatch.
- Inspect the results page for the doubtful queries in the SERP analyzer: which page types already hold the top ten, and which features the page carries.
- Run the list through the intent classifier and compare it with your own labels. The disagreements are the interesting cases.
Then decide per mismatch: rewrite the page for the intent, build a separate one, or hand the query to another page. The top pages report shows where the upside is. The glossary covers the terms: search intent, keyword intent, intent alignment, SERP.
How to tell you matched the intent
Record three numbers for the page before you change it: impressions, CTR and average position for its main query. Three to four weeks after the rewrite, look at the same three.
A match looks like this: that page is the one showing for the target queries, CTR grows at the same position, and nearby queries of the same type start appearing. If the position improved but CTR did not, the snippet promises something other than what people want.
What not to expect: an instant effect, or the same reaction across every query. Intent behind a phrase shifts over time, so the results page for important queries is worth re-reading every few months.
What comes next
We have covered why people type a query. The next lesson is how users search: phrasing, refinements and the path from the first query to a decision. The previous lesson on the stages of Search is here.