Finding the right SEO keywords means running a short research process before you write anything: generate a list of real candidate terms, check what each offers in volume and competition, match it to what the searcher wants, and choose the ones worth building a page around. It’s research and selection — not a single lookup, and not a guess.
That process is the whole job. Once you’ve found and chosen your keywords, placing them on the page — how much, where, and how to avoid stuffing — is a separate, later step covered in How to Use Keywords for SEO. This page is about what comes first: finding candidates and deciding which are worth targeting.
Where to Find Keyword Ideas
Good keyword ideas come from a handful of real sources — not from guessing what sounds right.
Your own audience’s language. Sales calls, support tickets, reviews, and the questions customers actually ask are a direct line to how real people describe the problem you solve — free, and often the most accurate source you have.
Google Search Console. If your site is live, this shows the actual queries already bringing impressions to your pages, including terms you never intentionally targeted.
Autocomplete and “.” Type a phrase into Google and watch what it suggests, then scroll to the “People also ask” box. Both are drawn from real searches and cost nothing to check.
Keyword research tools. Ahrefs, Semrush, Moz, and Google’s own Keyword Planner (free with a Google Ads account, though the data sharpens once you’re running ads) show volume, a difficulty estimate, and related terms in bulk. Most aren’t free outright, but a limited free tier or a single paid month can generate a solid list for a focused niche.
Competitor pages. What terms are the pages already outranking you built around? Reading what shows up above you regularly surfaces ideas you hadn’t considered.
Forums and communities. Reddit threads, industry forums, and niche Facebook groups show the exact phrasing people use outside a search box — often more natural and specific than what a keyword tool returns.
Pulling from several of these rather than relying on one gives you a fuller list before you evaluate any of it.
How to Research a Keyword Once You Have It
A list of candidate terms isn’t useful until you know something about each one. Two numbers and one judgment call matter most:
— how often people search the term, usually shown as a monthly estimate. Higher volume means more potential traffic, and typically more competition.
Keyword difficulty — a score, usually from a third-party tool, estimating how hard the term is to rank for based on the pages currently ranking. Treat it as an estimate, not a guarantee — the tool doesn’t know your site’s authority or the content you’re about to build.
— what the searcher actually wants. This matters more than the other two combined, which is why it gets its own section next.
None of these numbers means much alone. A high-volume, low-difficulty keyword that doesn’t match your audience’s intent isn’t a good keyword — it’s a number that looks appealing on a spreadsheet.
Understanding Search Intent
Every keyword carries an intent behind it — the actual goal the person had when they typed it. Getting this wrong is a common reason a page can target a seemingly “right” keyword and still underperform.
Informational intent. The searcher wants to learn or understand something — “what is a keyword,” “how does SEO work.” They’re not ready to buy.
Navigational intent. The searcher is trying to find a specific site or page they already have in mind, like a brand name or a login page.
Commercial intent. The searcher is comparing options before deciding — “best keyword research tools,” “Ahrefs vs Semrush.” Close to a decision, but not there yet.
Transactional intent. The searcher is ready to act — “buy,” “hire,” “sign up,” or “near me” phrasing. This is as close as a keyword gets to bottom-of-funnel.
The same root term can carry different intent depending on phrasing: “keyword research” (informational) versus “keyword research tool pricing” (commercial) versus “hire someone to do keyword research” (transactional). Content built for the wrong intent — a sales page targeting an informational query — tends to underperform even when the keyword looks like a strong match on paper.
How to Choose Which Keywords Are Worth Targeting
Once you have a researched list, choosing means weighing several things against each other, not picking whichever number is highest:
Relevance to what you actually offer. A keyword with strong volume that only loosely connects to your product brings traffic that doesn’t convert and won’t build real authority around the topic. Relevance filters out more candidates than any other factor.
Intent match. Choose keywords whose intent matches what the page is meant to do. A page meant to sell should target commercial or transactional keywords; a page meant to educate and build should target informational ones.
Realistic ranking chance. A brand-new site is unlikely to outrank established, high-authority pages for a competitive, high-volume term in the near term, regardless of how well the page is built. Lower-competition, specific terms are usually the more realistic starting point while a site builds authority.
Long-tail versus short-tail. Short, generic terms (“plumber”) carry high volume and high competition. Longer, specific phrases (“emergency plumber for a burst pipe at night”) carry less volume individually but typically face less competition, with intent that’s narrower and clearer. For most businesses, especially newer or smaller ones, that’s where the realistic opportunity is.
Two honest cautions here. First, difficulty scores are estimates, not a guarantee for your specific page — treat them as a rough filter, not a verdict. Second, chasing volume alone is a trap: a smaller, highly relevant audience that converts beats a large, loosely related one that doesn’t.
Organizing Keywords Into a Content Plan
Finding good keywords isn’t the end of the process — they need to be organized before they’re useful. Group related keywords by topic rather than treating each as a standalone target. A cluster of closely related terms usually belongs on one well-built page rather than spread across several thin ones — targeting the same intent with multiple competing pages just splits your own results against each other. Map each keyword or tight cluster to one page, and check which page already covers which intent before writing something new.
From there, prioritize. Some keywords are realistic quick wins — lower competition, close to your current site authority, a clear intent match. Others are longer-term targets worth building toward as your authority grows. Sequencing which keywords you build for first is what turns a list into an actual plan. For a broader look at running this alongside the rest of an SEO effort without outside help, see How to Do SEO Yourself.
Finding Keywords for an AI-Driven Search World
One newer wrinkle worth understanding: it changes the finding process more than the choosing process.
AI answer engines — Google AI Mode, ChatGPT, Perplexity — respond to full questions and conversational phrasing, not just short keyword strings. Keyword research increasingly means collecting the actual questions your audience asks, in their own words. “People also ask” boxes, forum threads, and the “questions” filters built into most keyword tools are useful sources for this format.
It’s also worth thinking in terms of entities and topics, not only strings of text — search engines and AI systems build understanding around what a business or concept actually is, not just which exact words appear on a page. A well-researched keyword list should cover a topic from multiple angles rather than fixating on one exact-match term. Exactly how AI systems weigh any of this internally isn’t published, so treat this as a direction to lean into, not a formula to follow precisely.
Common Questions
What’s the difference between finding keywords and using them?
Finding keywords is the research and selection process — generating candidates, checking volume and intent, and deciding which are worth targeting. Using them is what happens after: placing them on the page without overusing them. See How to Use Keywords for SEO for that half of the process.
Do I need a paid tool to find good keywords?
No. Google Search Console, autocomplete, “People also ask,” and competitor pages are free and can produce a solid candidate list on their own. Paid tools add bulk volume and difficulty data across hundreds of terms at once, saving time — but they’re a convenience layer on the free sources, not a requirement.
How many keywords should I research before I start choosing?
There’s no fixed number, but casting a wide net before narrowing helps — a shortlist built from only three or four ideas rarely surfaces your best options. Pull candidates from several of the sources above, then narrow using the relevance, intent, and ranking-chance factors covered earlier.
What if a keyword has strong volume but only loosely relates to what I offer?
Skip it, or treat it as a low priority. Traffic from a loosely related keyword rarely converts and doesn’t build genuine authority in your actual area of expertise. Relevance to your business should outweigh volume in almost every case.
Is it worth targeting a keyword my competitors already dominate?
Sometimes, if you can produce something more thorough or current than what’s ranking now — but for a newer or smaller site, that’s usually a longer-term target, not a first move. Prioritize lower-competition terms first, then build toward the harder ones as your site earns authority.
Should I do this research myself, or hand it off?
Plenty of businesses run this process themselves, especially early on — see How to Do SEO Yourself for what that realistically involves. If the research and the ongoing upkeep isn’t something you want to own, What Is an SEO Service covers what a service typically handles instead.