Once you understand that SEO is fundamentally about matching pages to what people search, the next logical question is: how do you actually figure out what people search? That's what keyword research answers, and it's less mysterious than most explainers make it sound.
What are SEO keywords, exactly?
An SEO keyword is any word or phrase a real person types into a search engine when looking for something. That's the whole definition. "Best running shoes for flat feet," "how to fix a leaking faucet," and "CRM software" are all keywords, they just represent very different searchers at very different points in their decision.
Optimizing a page for a keyword means structuring the content, the title, and the headings around what someone searching that exact phrase actually wants to find. Get that match wrong, and no amount of technical polish will make the page rank.
Where keywords actually go on a page
Knowing your keyword only helps if the page reflects it in the places search engines and readers look first. The primary keyword, or a close natural variant, belongs in the title tag, the main heading (H1), and the opening paragraph, plus a couple of subheadings where it genuinely fits the section. It also helps in the URL slug, and in image alt text when the image really does show that topic.
The meta description is worth writing around the keyword too, though for a different reason: it doesn't directly influence rankings, but it shapes whether someone clicks your result over the one above it. Beyond those placements, the rule is simple. If a sentence sounds like it was written for a robot, rewrite it for a person.
The main types of keywords, with examples
Keywords sort along two useful axes: length, and intent.
Short-tail keywords are one or two words, high search volume, and brutally competitive. "SEO," "shoes," "insurance," these get searched constantly but rarely tell you what the person actually wants, and ranking for them takes serious authority.
Long-tail keywords are longer, more specific phrases. "SEO agency for Shopify stores" or "running shoes for flat feet under $100" get searched far less often individually, but they're easier to rank for and convert better, since the searcher's intent is much clearer.
Layered on top of length is intent, why someone is searching:
- Informational: "what is SEO," "how does keyword research work." The searcher wants to learn, not buy yet.
- Navigational: "Zeronovax contact page." The searcher already knows where they want to go.
- Commercial: "best SEO agencies 2026." The searcher is comparing options before deciding.
- Transactional: "hire SEO agency," "SEO packages pricing." The searcher is ready to act.
Matching a page to the right intent matters as much as matching the topic. A commercial-intent page written like an informational blog post will rank for the wrong reasons, if it ranks at all.
What search volume and keyword difficulty actually mean
Two numbers show up in every keyword tool, and both get misread constantly. Search volume, visible in free tools like Google Keyword Planner, is roughly how many times a term gets searched per month, it tells you the size of the opportunity, not its quality. Keyword difficulty (KD) is an estimate of how hard that term is to rank for, based on the authority of pages currently ranking there.
The mistake we see most often: chasing the highest-volume term available while ignoring difficulty entirely. A term with 10,000 monthly searches and a KD of 70 is usually a worse target for a newer site than a term with 500 searches and a KD of 15, the second one is actually winnable.
How to actually analyze keywords
Stripped of tool-specific jargon, keyword analysis is four steps:
- Start with seed terms, the obvious words describing what you do or sell
- Expand them using a keyword tool to surface related searches and their volume
- Check difficulty against your site's current authority, not the term's popularity alone
- Group results into clusters by shared intent, so one page can target a primary keyword plus its close variants, rather than splitting them across competing pages
That last step, clustering, is where most beginner keyword lists go wrong. A spreadsheet of 200 loosely related keywords isn't a strategy; a dozen intent-based clusters mapped to specific pages is. Google's own SEO Starter Guide covers the fundamentals of how it evaluates content relevance, worth reading once you've got your clusters mapped.
A worked example: from seed term to cluster
Say you run a small online store selling running shoes. Your seed term is "running shoes," which any keyword tool will show has enormous volume and a difficulty score a new store has no realistic chance against. Expanding it surfaces hundreds of related searches, and a few stand out: "running shoes for flat feet," "best running shoes for overpronation," and "running shoes for flat feet under $100."
Those three share the same searcher: someone with a specific foot problem who is ready to compare options. Instead of writing three thin pages, you build one strong page targeting "running shoes for flat feet" as the primary keyword, and cover the other two as supporting variants within it. The exact numbers will differ by tool and market, but the pattern holds: a specific, winnable cluster beats a famous term you will never rank for.
Check the search results before you commit
A keyword tool tells you what a term looks like on paper. The search results page tells you what Google actually rewards for it. Before you commit a page to a keyword, search it yourself and study the first page.
Notice what kind of pages rank: product pages, comparison lists, long guides, or videos. If every result is a how-to guide and you're planning a sales page, the intent doesn't match and you'll be fighting the wrong battle. Notice who ranks, too. If the top spots belong to major publishers and household brands, the realistic target for a smaller site is a narrower variant of the term. Five minutes on this step regularly saves weeks of writing the wrong page.
Common mistakes beginners make with keywords
Stuffing a keyword into a page unnaturally, repeating it every other sentence, does more harm than good; search engines read that as manipulation, not relevance. Ignoring search intent is just as common, ranking a product page for an informational query (or vice versa) means it'll bounce even if it briefly ranks. And chasing only the biggest, most obvious terms while ignoring long-tail variants leaves easier, still-valuable traffic on the table.
A fourth mistake is treating keyword research as a one-time task. Search behavior shifts, new competitors appear, and your own site gains authority, which makes terms that were out of reach last year realistic today. Revisiting your clusters every few months keeps the plan honest. It also helps to map one primary keyword to one page before you publish anything, because two of your own pages targeting the same term end up competing with each other instead of with your rivals.
Where to go from here
Keyword research is the input; what you do with it is the actual strategy, which pages to build or fix first, which clusters to prioritize given your site's current authority, and how to sequence the work so early wins fund the harder ones. That sequencing is exactly what a complete SEO strategy engagement is built to figure out.
Frequently asked questions
SEO keywords are the words and phrases people actually type into search engines. Optimizing for a keyword means structuring a page so it clearly answers what someone searching that term is looking for, in a way search engines can recognize.
One primary keyword per page, plus a handful of closely related secondary terms that share the same search intent. Targeting unrelated keywords on the same page usually dilutes it rather than expanding its reach.
Yes, arguably more relevant. AI Overviews and chatbots still have to identify what a query is actually asking before generating an answer, and that identification runs on the same underlying language patterns keyword research maps out. The targets are shifting toward fuller questions, but the discipline behind it hasn't gone away.