Search intent is the reason behind a query: what the searcher wants to have, know or do when they finish. Most SEO guides reduce it to four boxes and a list of trigger words, and most keyword tools assign each term to a box automatically. That is a reasonable start and a poor finish, because the words in a query are only a hint. The pages Google chooses to rank are the evidence.
Get intent wrong and nothing else matters. A thorough guide will not rank for a query where Google shows product pages, and a product page will not rank where Google shows guides.
The six types of search intent
SE Ranking’s guide to search intent by Yulia Deda, published on 12 February 2026, sets out six types. The table below lists each with the modifiers the guide associates with it.
| Type | The searcher wants | Typical modifiers |
|---|---|---|
| Informational | To learn something | who, what, when, where, why, how |
| Navigational | To reach a specific site or product | Brand and product names |
| Transactional | To act now | buy, subscribe, price, cheap, discount, sale, order |
| Commercial investigation | To compare before deciding | review, vs, best, top, alternative, comparison |
| Local | Something nearby | “near me” |
| Generative AI | A tool to create, calculate or generate something | Requests made to AI tools |
The first four are the familiar model. Local is often folded into the others, but it behaves differently enough, with map results and proximity, to treat on its own. The sixth is new, and it changes how you should think about some informational queries.
Commercial vs transactional search intent
The two types most often confused are commercial and transactional, and the distinction is where affiliate sites make or lose money. Commercial investigation is the research stage: “best running shoes for flat feet”, “X vs Y”, “Z alternative”. The searcher has not decided. Transactional is the decision: “buy”, “price”, “discount code”. The searcher has.
That is why a comparison article can rank for a “best” query and fail completely on a “buy” query. Google answers the first with reviews and lists, and the second with shop pages and product listings. An affiliate site belongs in the first set of results, rarely the second.
Navigational queries are the other common trap. A keyword can have high volume because people use it to reach one brand’s site. Ranking second for it is worth very little, because almost everyone wanted the first result.
Local intent deserves its own check because it changes who can compete. A query with “near me” in it, or a service that people buy locally, tends to return map results and business listings. A content site with no physical location has little to offer there, however good the article. If a keyword you like turns out to be local, it is usually a keyword for someone else’s business, not for your blog.
The same keyword can also be local without saying so. Searches for a trade or a service often return map results even when the searcher typed no place name, because Google infers that they want someone nearby. That is another reason to check the results page rather than the modifier list.
Generative AI intent
The newest type covers requests to create, calculate or generate: write this, summarise that, work out the other. SE Ranking cites research that generative intent accounts for 37.5% of ChatGPT queries, while informational queries make up roughly 70% of traditional searches. The two figures come from different platforms, so read them as a contrast in behaviour rather than a like-for-like comparison.
The practical point is that some of what used to be informational search has moved into tools that produce the answer directly. That overlaps with what we cover in AI Overviews and traffic: queries a machine can answer in a paragraph are the ones least likely to send a click.
How to read search intent from the results page
Modifiers are a shortcut. The method is to search the keyword and look at what is there. The result types Google shows are its current best guess about what searchers want, built on far more behavioural data than you have.
- Note the page types. Guides, product pages, category pages, reviews, videos, forum threads. The dominant type is the intent.
- Note the features. A map pack signals local intent. Shopping results with prices signal transactional. Related questions signal informational.
- Check for mixed results. If the first page splits between guides and product pages, intent is mixed, and either format can compete, though neither is safe.
- Read the angle. Within one type, the top pages often share an angle: beginner-friendly, budget-focused, a particular year. That angle is part of the intent.
Our guide to SERP analysis walks through this in more detail, including what to record for each result. Once you have classified a keyword list this way, tag every keyword with its type in your spreadsheet. It stops two pages targeting the same intent and helps you spot gaps where no page of yours fits.
The tags also make the build order clearer. Informational keywords become guides and explainers that earn links and internal authority. Commercial keywords become comparisons and reviews, which is usually where an affiliate site earns. Transactional and local keywords are mostly for businesses that sell directly, and you can often set them aside. If two keywords share an intent and return the same pages, they belong on one page, not two.
Where intent models break
The honest limitation is that every model is a simplification. Many queries carry more than one intent, and the mix changes over time as Google tests what searchers click. A keyword that showed guides last year may show product listings now. A classification is a snapshot, not a property of the keyword.
Modifier lists also miss context. “Apple” is navigational for most people and informational for a few. “Cheap flights” contains a transactional modifier, yet the searcher may well want to compare before booking. Automated intent labels in keyword tools work from patterns like these, so treat them as a first pass and check the results page for any keyword you plan to build a page around.
It also helps to see how successful sites map pages to intent. In the database explorer, filter to your niche and sort by estimated organic traffic, then look at how the top sites split their pages between guides, comparisons and product content. That is a method for studying structure, not a template. Length should follow intent too, which is covered in how long a blog post should be for SEO.
Frequently asked questions
What are the different types of search intent?
SE Ranking lists six: informational, navigational, transactional, commercial investigation, local and generative AI. The first four are the traditional model.
What is the difference between commercial and transactional intent?
Commercial searchers are still comparing options, using terms like “best”, “review” or “vs”. Transactional searchers are ready to act, using terms like “buy”, “price” or “order”.
How do I find the search intent of a keyword?
Search it and look at what ranks. The dominant page type and the results features, such as maps or shopping listings, show the intent Google is serving.
The takeaway Modifiers suggest intent; the results page confirms it. Classify each keyword by what currently ranks, and build the page type Google is already rewarding.
