Keyword Research: How to Find Terms You Can Actually Rank For
Most keyword research produces a spreadsheet of high-volume terms the site has no chance of ranking for, sorted by the one column that predicts results least well. The useful version asks three questions instead: what does this person want, can we credibly be the answer, and is that answer worth anything to us.
Intent decides everything else
Before volume or difficulty, work out what the searcher wants. Broadly there are four patterns: informational, where someone wants to understand something; navigational, where they want a specific site; commercial, where they are comparing options before buying; and transactional, where they are ready to act.
You cannot rank a product page for an informational query, however good the page is, because Google already knows what people click for that term. The fastest way to see intent is to search the term and look at what actually ranks. If the first page is all guides and definitions, that is a content page. If it is all product listings and comparison pages, that is a commercial page.
This single check kills more bad keyword targets than any difficulty score. A term with excellent volume and a first page of Wikipedia, Reddit threads and university pages is not a commercial opportunity, no matter how relevant the words look.
Difficulty scores are directional, not factual
Every tool sells a keyword difficulty number, usually derived from the backlink profiles of the current top ten. It is a reasonable proxy and a terrible absolute. It does not know your site authority, your topical relevance, or that three of those top results are outdated pages nobody has touched in four years.
A better assessment takes two minutes per term. Look at the actual results: how many are big national brands versus sites your size, how old and how thorough is the content, and are any of them obviously weak matches that Google is only showing because nothing better exists. That last case is the opening you are looking for.
Be honest about where you are starting. A new site targeting terms dominated by established publishers is choosing a two-year project. The same effort spent on specific, lower-competition terms produces traffic in months, and that traffic builds the authority that makes the harder terms reachable later.
The long tail is where new sites win
Specific multi-word queries have less volume individually but far less competition, clearer intent and better conversion rates. Someone searching a precise question with a location, a constraint or a use case has already narrowed their thinking and is much closer to acting than someone typing a two-word category term.
They also aggregate. Fifty specific pages each attracting modest monthly traffic add up to more than one competitive term you never reach, and they arrive faster. Sources worth mining: the questions in your own inbox and sales calls, the "people also ask" boxes, forum and Reddit threads in your sector, and the query report in Search Console showing what you already rank for on page two.
That Search Console report is the most underused list in SEO. Terms where you sit between positions eight and twenty are already relevant enough to appear; improving those pages is far cheaper than starting new ones from nothing.
Group by topic, then map to pages
Many keywords are the same question phrased differently, and they should not become separate pages. Google resolves them to the same intent, so publishing five thin variations means those pages compete with each other and none of them wins. Group everything that shares an intent, pick the clearest phrasing as the primary target, and build one strong page.
Then map explicitly: one page per group, written down. This is what stops the slow drift into cannibalisation where three pages half-target the same thing. Where you find existing overlap, consolidating into one page and redirecting the others usually produces an immediate improvement.
Finally, sort by commercial value rather than volume. A term with modest traffic that reaches people ready to buy is worth more than a popular term that brings students and job seekers. Our SEO services start from that mapping, and our keyword density checker helps confirm a page reads naturally once written.
Key takeaways
- Check what already ranks before targeting a term. The current first page tells you the intent Google has settled on, and you cannot argue with it.
- Treat tool difficulty scores as directional. Judge the actual results for authority, age and quality instead.
- Specific long-tail terms convert better, arrive faster and aggregate into meaningful traffic.
- Search Console queries ranking between positions eight and twenty are the cheapest wins available.
- Group keywords by intent and map one page per group, or your own pages will compete with each other.
Frequently asked questions
Is search volume the most important keyword metric?
No, and treating it that way is the most common mistake in keyword research. Volume tells you how many people search a term, not whether you can rank for it or whether those people will ever buy anything. A term with a few hundred monthly searches and clear buying intent that you can realistically reach is worth far more than a term with fifty thousand searches dominated by national brands. Rank your list by the combination of achievability and commercial value, and use volume only to break ties.
How do I know if I can rank for a keyword?
Search it and study the first page rather than trusting a score. Ask how many results are large established brands, how strong and how current the content is, and whether any results look like weak matches Google is showing because nothing better exists. If the page is entirely major publishers with comprehensive, recently updated content, that is a long-term project. If there are thin pages, outdated posts or forum threads ranking, there is room for a genuinely good page to break in.
What is search intent and why does it matter more than keywords?
Search intent is what the person is actually trying to accomplish: understand something, find a specific site, compare options, or buy. It matters more than the keyword itself because Google ranks pages that satisfy the dominant intent for a query and will not rank a page of a different type however well optimised it is. Publishing a product page for a question-shaped query, or a long guide for a query where everyone wants a product listing, fails regardless of how much effort goes into it.
Are long-tail keywords still worth targeting?
More than ever. Competition on broad terms has intensified while search results increasingly answer simple questions directly, so the value has shifted towards specific queries where someone has a defined need. Long-tail terms have clearer intent, far less competition, and convert at higher rates because the person has already narrowed their thinking. Individually they look small; fifty of them ranking is a substantial and durable traffic base that also builds the authority needed for harder terms.
How many keywords should one page target?
One primary intent, and as many phrasings of it as naturally fit. If several keywords are just different ways of asking the same thing, they belong on the same page, because splitting them creates thin pages that compete with each other. If two keywords need genuinely different answers, they need different pages. The test is simple: could one well-written page satisfy someone searching either term? If yes, one page. If a reader searching one of them would feel the page was about something else, separate them.
What is keyword cannibalisation and how do I fix it?
Cannibalisation happens when several pages on your site target the same intent, so search engines cannot decide which to rank and often rank the weaker one, or rotate between them. Find it by searching your own domain for the term, or by checking Search Console for a query where the ranking URL keeps changing. The usual fix is consolidation: merge the best content into one strong page and redirect the others to it, which concentrates both the content and the links pointing at them.