Google classifies queries by intent before applying ranking algorithms. This classification determines which page types are eligible to rank, not just which pages rank highest. A product page can be perfectly optimized but will never rank for an informational query if intent classification excludes transactional pages.
The Intent Classification Layer
Query intent classification operates as a filter before ranking.
The process:
- User submits query
- Google classifies query intent
- Classification determines eligible page types
- Ranking algorithms score eligible pages
- Top eligible pages appear in results
Intent categories (Google’s Quality Rater Guidelines, March 2024):
- Know (Informational): User wants to learn something
- Know Simple: User wants a specific fact
- Do (Transactional): User wants to accomplish something
- Website (Navigational): User wants a specific website
- Visit-in-person (Local): User wants to go somewhere physically
- Multi-intent: Query has multiple common intents
The eligibility mechanism:
A “Know” query might exclude:
- Product pages
- Service pages
- E-commerce category pages
- Promotional landing pages
A “Do” query might exclude:
- Wikipedia-style informational articles
- Definition pages
- Educational content without action capability
Observable Intent Patterns
SERP analysis reveals intent classification through result patterns.
Method:
Search a query and analyze what types of pages rank:
| Query | SERP Content Types | Likely Classification |
|---|---|---|
| "what is machine learning" | Wikipedia, educational sites, explainers | Know |
| "machine learning tutorial" | Tutorial sites, course platforms, guides | Know + Do |
| "machine learning course" | Course platforms, pricing pages, universities | Do (transactional) |
| "tensorflow" | TensorFlow.org, docs, GitHub | Website (navigational) |
Intent signals in queries:
- Question words (what, how, why) → Know intent
- Action words (buy, download, subscribe) → Do intent
- Brand names → Website intent
- Location modifiers → Visit-in-person intent
- Review, best, vs, compare → Mixed Know/Do intent
The Site Type Dimension
Beyond page type, site type affects eligibility.
Observable site type classifications:
The 2024 API leak (Rand Fishkin, SparkToro, May 2024) revealed “siteType” as a ranking attribute. While specific values weren’t fully documented, observable patterns suggest categories:
- Brand sites (official company/product sites)
- Retailers (e-commerce sellers)
- Publishers (news, blogs, content sites)
- UGC platforms (forums, Q&A, social)
- Video platforms
- Reference sites (Wikipedia-style)
Site type eligibility:
| Query Type | Eligible Site Types | Often Excluded |
|---|---|---|
| Brand query + informational | Publishers, reference | Brand site (considered biased) |
| Brand query + transactional | Brand site, retailers | Publishers |
| Product category | Retailers, review sites | Brand sites (for generic terms) |
| How-to | Publishers, UGC, video | E-commerce |
| News query | Publishers, news sites | Everything else |
The brand query paradox:
Brand sites often don’t rank for their own brand + informational queries:
- “Nike running shoe reviews” → Review sites, not Nike.com
- “Apple iPhone problems” → Forums, tech sites, not Apple.com
Google considers brand sites potentially biased for informational queries about their own products.
Intent-Optimized Page Strategy
Align page types with query intent to ensure eligibility.
Strategy 1: Create intent-specific pages
For each important keyword cluster, create pages matching the dominant intent:
| Query Cluster | Dominant Intent | Page Type Needed |
|---|---|---|
| "best CRM software" | Know + Do | Comparison/review article |
| "CRM software pricing" | Do | Pricing page |
| "what is CRM" | Know | Educational article |
| "buy Salesforce" | Do | Product/purchase page |
A single product page won’t rank for all these queries. Each requires its own intent-matched page.
Strategy 2: Intent bridging
Create pages that satisfy multiple intent layers:
Comparison article (Know intent)
├── Explains CRM options (satisfies Know)
├── Includes buying recommendations (bridges to Do)
└── Links to product pages (enables Do completion)The article ranks for Know queries while funneling to Do-oriented pages.
Strategy 3: SERP analysis for intent understanding
Before creating content, analyze SERP to understand intent classification:
- Search target query
- Note page types ranking (articles, products, categories, tools)
- Note site types ranking (publishers, retailers, brands)
- Note SERP features (featured snippets suggest Know; shopping results suggest Do)
- Create content matching observed intent patterns
Misalignment Diagnosis
Identify when poor rankings result from intent misalignment versus other factors.
Diagnosis protocol:
- Search your target query
- Analyze top 10 results for page type pattern
- Compare your page type against pattern
- If your page type doesn’t match, intent misalignment is likely cause
Example diagnosis:
Query: “how to use Google Analytics”
Top 10: Tutorials (8), official docs (1), video page (1)
Your page: Product page for analytics consulting service
Diagnosis: Intent misalignment. Product pages don’t rank for tutorial queries.
Solutions for misalignment:
| Misalignment Type | Solution |
|---|---|
| Product page for Know query | Create separate educational content |
| Article for Do query | Create transactional page or add clear CTA |
| Category for specific product query | Create specific product page |
| Brand page for informational brand query | Accept or create separate editorial content |
Query Intent Volatility
Intent classification isn’t static. It shifts based on:
Temporal factors:
“iPhone” query intent shifts with product cycle:
- Pre-announcement: Informational (rumors, speculation)
- Launch: Mixed (news + purchase intent)
- Post-launch: Transactional (purchase dominant)
- Mature: Split (troubleshooting, accessories, purchase)
Current events:
“Tesla” during regular period: Company info, stock price, products
“Tesla” during controversy: News about controversy
“Tesla” during product launch: Product information + purchase
Seasonal shifts:
“Halloween costumes” in July: Ideas and inspiration (informational)
“Halloween costumes” in October: Purchase intent (transactional)
Strategic response:
Monitor SERP changes for important queries. Adapt content strategy when intent classifications shift.
Multi-Intent Queries
Many queries have legitimate multiple intents.
Multi-intent handling:
Google may show mixed SERP with different intent segments:
- Featured snippet (Know Simple)
- Informational articles (Know)
- Product listings (Do)
- Local pack (Visit)
Opportunity identification:
Multi-intent queries allow various page types to rank. Create content for each intent segment:
Query: “running shoes”
- Educational article about choosing running shoes (Know segment)
- Product category page (Do segment)
- Store locator page (Visit segment)
Each captures a segment of the multi-intent query.
Intent Classification Testing
Test intent hypotheses before major content investments.
Testing approach:
- Create minimal viable content for hypothesized intent
- Monitor ranking response
- If content indexes but doesn’t rank despite quality, intent may be misaligned
- If content achieves positions, intent alignment confirmed
A/B intent testing:
- Create two versions targeting same keyword with different intents
- Version A: Product-focused (transactional intent)
- Version B: Educational (informational intent)
- Monitor which version achieves rankings
- Scale the winning intent approach
Intent monitoring:
Track SERP composition for important queries monthly:
- What page types appear?
- What site types dominate?
- Have patterns shifted?
- Do your pages match current patterns?
Query intent classification is an often-invisible filter that determines ranking eligibility before ranking quality assessment begins. Pages excluded by intent classification never compete, regardless of other optimization factors. Understanding and matching query intent ensures ranking eligibility, making all other optimization efforts worthwhile.