SMThe SERP MasterAdvanced SEO Tools

AI Keyword Intent Classifier

Paste keywords and find out what searchers actually want — informational, commercial, transactional, or navigational.

✨ AI-Powered

Understanding Search Intent

Paste a list of keywords and this tool labels each one with its search intent — informational, commercial, transactional, or navigational — along with a short reason for the call. It uses AI rather than a keyword blacklist, so it judges what the searcher is actually trying to do rather than looking for the word “buy”.

Intent is the thing most keyword research skips, and it is the reason a lot of well-optimised pages never rank. You can target the right keyword with the right density and the right title and still lose, simply because the page you wrote is not the kind of page that query wants.

The four intent types

Search intent describes what someone wants when they type a query. The standard four-way split has held up well because each type calls for a genuinely different kind of page.

  • Informational — the searcher wants to learn something. “How does compound interest work”, “what is technical SEO”. Best served by guides, explainers and tutorials.
  • Commercial — the searcher is comparing options before deciding. “Best CRM for small business”, “Ahrefs vs Semrush”. Best served by comparisons, reviews and roundups.
  • Transactional — the searcher is ready to act. “Buy running shoes online”, “Notion pricing”. Best served by product, pricing and signup pages.
  • Navigational — the searcher wants a specific site or page. “Gmail login”, “Nike returns policy”. Usually only winnable by the brand itself.

Why intent decides your page format

If a query is informational and you point a product page at it, you will struggle no matter how good the product page is — Google has already learned from click behaviour that people searching that query want to read, not to buy. The reverse is equally true: a 3,000-word guide targeting “buy running shoes size 10” is the wrong shape for someone with a card in their hand.

This is why intent should be settled before you write a word. It determines whether you are writing a listicle or a landing page, whether the page needs a comparison table or a step-by-step walkthrough, and what the call to action should be.

It also determines what success looks like. An informational page that generates no direct sales is not failing — it is doing its job, building awareness and internal link paths toward the commercial pages further down the funnel.

Checking intent against the actual results

The most reliable intent signal is free: search the query and look at what Google already ranks. Google has enormous behavioural data on what satisfies each query, and the top ten results are its answer.

If the first page is all blog posts, the query is informational, whatever the words suggest. If it is all product pages, it is transactional. If it is a mix, the query has split intent and there is room for more than one page format.

Pay attention to the SERP features too. A featured snippet or People Also Ask block signals informational intent. Shopping ads and product carousels signal transactional. A knowledge panel usually means navigational.

Using intent to structure a funnel

Sorting a keyword list by intent produces a rough content funnel automatically. Informational keywords sit at the top, where people are learning. Commercial keywords sit in the middle, where they are comparing. Transactional keywords sit at the bottom, where they are ready.

Most sites over-invest at one end. Ecommerce sites often have only transactional pages and wonder why they capture no early-stage demand; content sites often have only informational pages and wonder why traffic never converts. Seeing the intent split across your keyword list makes the imbalance obvious.

Once you know the split, link the layers together. Informational pages should point to the commercial comparisons, which should point to the transactional pages. The Internal Link Analyzer helps you check whether those paths actually exist on your published pages.

Where intent and clustering meet

Intent and clustering answer different halves of the same question. Clustering tells you which keywords belong on the same page; intent tells you what kind of page that should be. Run a list through the Keyword Clustering Tool first, then classify each cluster's intent here.

A cluster containing mixed intents is the clearest signal you will get that it needs splitting. If half a cluster is “how to choose X” and the other half is “buy X online”, those are two pages, however similar the words look.

Common intent mistakes

  • Assuming “best” always means commercial — “best way to tie running shoes” is informational.
  • Treating every question as informational; “where to buy X near me” is a question with transactional intent.
  • Chasing high-volume navigational keywords for someone else's brand, which you will not win.
  • Writing one page for a query with genuinely split intent instead of covering both angles or picking one.
  • Ignoring intent drift — queries change meaning over time, so a page written three years ago may now be the wrong format.

Frequently asked questions

What are the four types of search intent?

Informational (wanting to learn), commercial (comparing options before deciding), transactional (ready to act or buy), and navigational (looking for a specific site or page). Each is best served by a different kind of page.

How do I verify a keyword's intent myself?

Search it and look at the top ten results. Google's rankings reflect what actually satisfies that query. If the first page is dominated by blog posts, the intent is informational regardless of how commercial the words look.

Can one keyword have more than one intent?

Yes — this is called split or mixed intent, and you can spot it when the results page mixes formats. In those cases either pick the dominant intent or build a page that opens by answering the informational question and moves toward the commercial one.

Is this better than matching keywords against a word list?

Generally yes. Word-list matching flags “best” as commercial and “how” as informational without reading the rest of the phrase, which misfires often. AI classification weighs the whole query, which is why each row also shows its reasoning.

How many keywords can I classify at once?

Up to roughly 80 per run. For bigger lists, split them into batches — and consider clustering first, since classifying one keyword per cluster is usually enough to decide the page format.