The most expensive mistake in SEO is not a technical one. It is building the wrong kind of page for a query. A two-thousand-word guide for a query where people want a product listing. A product page for a query where people want to understand their options first. Both pages can be excellent and both will lose, because Google is matching the type of result to the intent of the query, not the quality of the page in the abstract.
This sub-guide is part of the SEO basics guide. It is the step between keyword research, which tells you what people type, and keyword mapping, which decides which page answers it.
The four intent types, and two that get left out
The classic taxonomy has four categories. I use it, with two additions that matter in practice.
Informational
The person wants to know or understand something. "What is GST composition scheme", "how to reduce LCP", "symptoms of vitamin D deficiency". The best result explains. The page type is usually a guide, an article, a definition, a video or a how-to. Monetisation, if any, is indirect.
Navigational
The person wants to reach a specific site or page. "IRCTC login", "Zerodha Kite", "HDFC credit card customer care". There is one right answer and it is almost always the brand's own property. If the brand is not you, do not chase these; if it is you, make sure your own pages win cleanly.
Commercial investigation
The person intends to buy or choose, but is still comparing. "Best term insurance plan in India", "Zoho vs Freshworks CRM", "Tata Nexon EV review". The best result helps them decide: comparisons, reviews, best-of lists, buying guides. This is the intent most businesses misread, because the query contains commercial words so they send it to a product page. The SERP usually says otherwise.
Transactional
The person is ready to act. "Buy running shoes online", "book cab to Mumbai airport", "Canva Pro pricing". The best result lets them do the thing: a product page, category page, pricing page, booking flow. Guides lose here.
Mixed intent
Many queries do not resolve to one intent, especially short ones. "Mutual funds" could be informational (what are they), commercial (which should I buy), navigational (a specific fund house) or transactional (invest now). Google handles this by serving a mixed SERP: a definition, a few investment platforms, a news box. For mixed queries, you choose which slice of the intent to compete for and build the page type that wins that slice, rather than trying to be everything.
Local intent
Local is a modifier that overrides the others. "Dentist near me", "CA firm in Indiranagar", "best biryani Hyderabad" all expect a map pack and location-specific results, regardless of whether the underlying intent is informational or transactional. If the SERP shows a map, you need a local presence and often a location page, which the local SEO guide covers.
The SERP is the ground truth
Keyword tools will label intent for you. Treat those labels as a first guess. The only authoritative source on what Google thinks a query wants is what Google is currently showing for it, because that result page is the product of everything Google has learned from how people behave on it.
Reading a SERP for intent is a quick, repeatable exercise. I do it for every target query before deciding what to build.
Search the query in a clean context
Use an incognito window, set the location to the market you care about, and use the device your users use, which in India is overwhelmingly mobile. Personalisation and location change the result.
Note the SERP features
A featured snippet, People Also Ask box or video carousel suggests informational intent. Shopping results, product grids or a pricing-heavy layout suggest transactional. A map pack means local. A knowledge panel for a brand means navigational.
Classify the top organic results by page type
For each of the top eight or ten, is it a guide, a comparison, a product page, a category, a tool, a forum thread, a video? Count them. If seven of ten are comparison articles, the intent is commercial investigation whatever the keyword tool said.
Look at the titles and angles
The way winning pages frame the topic tells you what sub-intent dominates. "Best X for beginners" and "X pricing explained" are different jobs even if the query is the same.
Decide the page type, then the angle
Match the dominant type. If the SERP is mixed, pick the slice where your site can credibly win and note which features you are not competing for.
Intent shifts over time
Intent is not fixed to the query. It moves, and a page that was correct two years ago can be the wrong type today.
Markets mature: when a product category is new, most queries about it are informational. As it becomes mainstream, the same queries turn commercial and then transactional; early "what is UPI" demand became "best UPI app" and then "UPI payment" as adoption spread. Events move intent too, and so do Google's own layouts: as AI-generated summaries answer simple definitions directly, the organic results underneath often shift toward deeper or more opinionated content.
The practical implication is that intent should be rechecked, not assumed. I re-read the SERP for core queries when a page's performance changes without an obvious cause, and on a schedule for the queries a business depends on.
Mapping intent to page types
Once the intent is read, the page type follows. This is the rule of thumb I use, with the caveat that the SERP always wins over the rule.
| Example query | Intent | What the SERP usually shows | Page type to build |
|---|---|---|---|
| what is a demat account | Informational | Definitions, featured snippet, PAA, broker explainers | Explainer guide with a clear definition up top |
| how to file ITR online | Informational, procedural | Step-by-step guides, government portal, videos | Step-by-step guide with screenshots |
| groww app login | Navigational | The brand's own pages, app store listings | Not a target unless you are the brand |
| best health insurance for parents in India | Commercial investigation | Comparison articles, aggregator listings, insurer guides | Comparison or buying guide with criteria and a shortlist |
| razorpay vs cashfree | Commercial investigation | Head-to-head comparisons, review sites, vendor pages | Honest comparison page; a vendor can own this with a fair treatment |
| buy office chair online | Transactional | Marketplace category pages, shopping results, D2C brands | Category page with filters, not a blog post |
| crm software pricing | Transactional, SaaS | Pricing pages, comparison tables | Pricing page, supported by a plan-comparison guide |
| ca near me | Local, transactional | Map pack, directory listings, firm websites | Business profile plus a location page |
| seo | Mixed | Definitions, guides, news, service providers | Pick a slice: a definitive guide or a service page, not both on one URL |
| mobile under 15000 | Commercial, price-bounded | Listicles with specs, marketplace category pages | Curated list with specs and a clear update date |
Two patterns in the table deserve comment. Price-bounded and "best" queries are commercial investigation even though they sound transactional, and the pages that win them are lists and comparisons. And vendors can rank for "X vs Y" queries that name themselves, provided the comparison is fair enough that the SERP accepts it; a page that only lists the vendor's strengths rarely holds the position.
What happens when you get it wrong
The symptom is a page that ranks on page two or three indefinitely, or gets impressions but a low click-through rate, or ranks briefly and falls back. In Search Console, the page appears for the query, but the position never stabilises in the top five. The usual fix is not to improve the page but to build the correct type and let the two pages point at each other: a guide that links to the category, a category that links to the guide. That is also how you avoid the two pages competing, which keyword mapping addresses in detail.
Google is not ranking the best page. It is ranking the best answer to what the query meant.
Frequently asked questions
Can one page rank for queries with different intents?
Sometimes, when the intents are adjacent, such as a how-to guide that also picks up "what is" queries. It rarely works across the informational-to-transactional gap. If the SERPs for two queries show different page types, you need two pages.
How do I find the intent of a query if the SERP is mixed?
Count the page types in the top ten and look at the SERP features. The majority type tells you the dominant intent. Then decide which slice you can win: a strong brand might take the transactional slice, a publisher the informational one.
Do keyword tools classify intent accurately?
They are a reasonable first pass and are useful for sorting thousands of queries. They are wrong often enough on the queries that matter that I always verify the top targets against the live SERP, in the right country and on mobile.
Does search intent differ for Hindi or Hinglish queries in India?
The intent categories are the same, but the phrasing and the SERP can differ. Hinglish queries such as "sasta phone kaunsa hai" are often commercial and are served by Hindi-language lists and videos. Check the SERP for the language variant separately; do not assume the English result carries over.
How often should I recheck intent for my main queries?
Whenever a page's performance changes without an obvious cause, and on a routine basis for the queries the business depends on. Markets mature and Google's layouts change, so an annual check is a minimum for core terms.
Where this connects
This sub-guide belongs to the SEO basics guide, and intent reading is the first step of my SEO content framework, where every brief begins with the SERP shape. The topical authority cluster test shows what happens when a cluster covers the full range of intents on a topic, and the SaaS organic growth case study is largely a story of separating informational, comparison and pricing intents onto the right pages. If you want this done across a whole site, it is part of my SEO strategy service.