Keyword research is figuring out which words your target audience types into Google and other search engines, so you create content that matches what people actually search for. You map out which questions exist, how big the demand is, and what intent people search with. Then you link those insights to specific pages.
Keyword research in short
Keyword research is the phase where you decide what your content is about, before you write a single word. You collect the terms and questions your audience uses, look at how many people search for them, and work out what they're trying to achieve. Based on that, you decide which topics deserve their own page and which belong together.
So it's not a list of separate words you cram into a text. It's the logic behind your entire site. Do keyword research properly, and you build a structure that works for both visitors and search engines. Skip it, and you're writing blind and hoping for traffic.
Why keyword research is the foundation of SEO
Without keyword research, you're guessing. You guess which words your customers use, which topics matter to them, and in what order they decide. Sometimes you guess right, but usually you end up writing about things nobody searches for, or using jargon your audience never types in. SEO starts not with writing text, but with understanding what people search for and why.
From topic to page logic
Keyword research doesn't just tell you which words exist, it also shows how your topics relate to each other. That determines your page architecture: which topic deserves a main page, which questions are subtopics, and which belong on a separate page. Think of it as the floor plan of a building. Before you put up walls, you decide which rooms there will be and how you move between them.
If you skip that logic, you often end up with multiple pages on almost the same topic. That's called keyword cannibalisation: your own pages compete with each other for the same search query. Google then doesn't know which page is best, and splits the attention. Instead of one strong page, you end up with three weak ones. Good keyword research prevents this because you see in advance which terms belong together and which are genuinely different.
The difference between writing what you want and creating what people search for
Many companies write about what they find interesting themselves: their process, their history, their own terminology. That reads fine internally, but rarely matches demand in the market. Keyword research forces you to switch perspective. Not: what do we call this? But: how does our customer search for this?
That difference is bigger than it looks. A manufacturer might use a technical product name, while the customer types in an everyday word. Publish around the technical term, and you miss everyone who uses the common one. Keyword research makes that gap visible and lets you choose the language of your audience.
What you research in keyword research
Keyword research is about more than the word itself. You look at three things at once: what someone means, how big the demand is, and how specific the search query is. Only when you look at all three together can you make good choices.
Search intent
Search intent is the purpose behind a search query. Someone who wants to understand something searches differently than someone who's ready to buy. Roughly speaking, there are four types of intent: looking for information, finding a specific website, comparing options, and making a purchase or taking action. If you don't know the intent, you might write an explainer article while the searcher wants a product, or the other way around.
For the Dutch market this is extra important, because one word can have multiple meanings. The same word can be a purchase query one time and a purely informational question about something completely different the next. You can't guess intent based on the word alone. You infer intent from the search results Google shows. You can read more about this in our article on search intent.
Search volume and feasibility
Search volume shows how often people search for a term on average. It's tempting to aim only for the biggest numbers, but high volume almost always means heavy competition. Feasibility is about whether you can realistically rank for that term, given who's already there.
The trick is balance. A term with less volume that you can actually win often delivers more than a high-volume term you'll never rank for on page one. Volume tells you how big the pond is. Feasibility tells you whether you have a rod that catches anything in it.
Long tail keywords
Long tail keywords are longer, more specific search queries. Instead of a broad single-term word, it's a concrete question or combination. Individually they have less volume, but added together they often make up the largest share of search demand. And they're more specific, so the intent is clearer and the competition is usually lower.
For most sites, the real gain lies in that long tail. Someone who searches very specifically knows better what they want, and is often closer to a decision. If you answer those concrete questions well, you attract visitors who fit your topic exactly.
How to do keyword research step by step
Keyword research follows a fixed sequence of steps. You start broad, expand, test against reality, organise, and only then link everything to pages. Skip a step, and it catches up with you later.
Step 1: start with seed keywords
Seed keywords are your starting words: the handful of core terms that describe your topic. Write down how you name your product or service, but also ask colleagues and customers how they refer to it. This is your rough starting point, still without tools.
Step 2: expand with keyword research
Enter your seed keywords into a keyword tool and collect all related terms, questions, and variants. That way your list grows from a handful of words to hundreds of search queries. In this step you're collecting, not filtering. Don't throw anything away yet.
Step 3: check the SERP and search intent
Search for your most important terms yourself and look at the results page. What kind of pages rank at the top: explainer articles, product pages, comparisons? That tells you what Google sees as the right answer, and therefore what the intent is. This step stops you from creating content that doesn't match what the searcher expects.
Step 4: cluster and prioritise
Group terms that share the same underlying question. A cluster is a set of search queries that belong on one page, because they want to know the same thing. Then you prioritise based on intent, volume, and feasibility. That's how you decide where to start.
Step 5: link keywords to pages
Assign each cluster to one page. Every important search query belongs to exactly one page, so your pages don't compete with each other. This is where keyword research and page architecture come together. Here you turn the floor plan into actual blueprints.
- Gather your seed keywords from your own knowledge and conversations.
- Expand with a keyword tool until you have a broad list.
- Check the SERP for each term to determine the search intent.
- Cluster by topic and prioritise on value and feasibility.
- Link each cluster to one page and lock in your structure.
Example of keyword research
An example makes it concrete. Say you have a broad topic and don't know how to break it down. Keyword research helps you pull the topic apart into clear content opportunities.
From broad topic to concrete content opportunities
Take a shop that sells garden furniture. The broad topic is clear, but you can't build a strategy on that alone. As soon as you expand with keyword research, you see that people search in very different ways. Some want to know which material holds up best against weather. Others search for maintenance. Others still want to buy a specific type of furniture right away.
Those aren't synonyms for the same thing. They're different intents, and therefore different pages. The informational questions become explainer articles. The purchase-oriented terms become category or product pages. What started as one vague topic becomes an organised set of pages that each answer their own search query. That's exactly the point of keyword research: from broad to usable.
Common mistakes when determining keywords
Most mistakes in keyword research aren't caused by too little data, but by misreading that data. Two pitfalls come up most often.
Why volume without context is misleading
A high search volume looks attractive, but says little on its own. If the competition is too tough, you won't rank, and you'll see none of that big number in visitors. And if the intent doesn't match what you offer, you attract traffic that will never become a customer anyway. Volume is a means, not a goal.
There's another catch: a word can have multiple meanings. The volume then adds up all those meanings, while only part of it relates to your topic. Look only at the number, and you overestimate your chances. So always check the SERP: who ranks there and about what tells you whether that volume is actually reachable and relevant for you.
Why one keyword isn't a strategy
Optimising your page for one single term is thinking in words, not topics. Google has long understood that a topic consists of many related questions. A page built around just one term misses the context that provides a complete answer.
Clustering by topic solves this. Instead of five thin pages around five separate words, you create one strong page that covers the whole topic. That's better for the visitor, and it stops your own pages from fighting over the same term.
“There is not one truth. There is a ranking order.”
Giacomo Perticara, founder and SEO strategist at GRP Digital
Tools for keyword research
Tools make keyword research faster and more concrete. They provide the data you don't have in your head: which terms exist, how often they're searched, and how tough the competition is. But a tool is an aid, not a strategy.
What a good keyword tool shows you
A useful keyword tool gives you more than just a word. You want to see search volume, an indication of competition, related terms and questions, and ideally a sense of intent. Trends over time help too: a term that's growing is worth something different than one that's declining. That combination is what separates a list of words from an informed choice.
Why tools alone aren't enough
A tool gives you numbers, but no judgment. It doesn't assess whether a term fits your audience, whether you can credibly rank for it, or how terms relate to each other. That clustering, that prioritising, and that relevance check are on you. A tool is like a tape measure: it gives you the measurements, but it doesn't build your house.
What's more, the most important check can't be filled in by software: can your brand even be relevant for this search intent? A high-volume term where the searcher wants something completely different isn't an opportunity, it's a distraction. That judgment stays human work.
Keyword research for SEO, content, and AI visibility
Keyword research isn't just for classic SEO. The same insights guide your content and increasingly determine whether you show up in AI answers. The underlying question stays the same: what does your audience want to know, and how do you answer that best?
What AI answers change
More and more people get their answer directly from an AI assistant or in a summary above the search results. They don't always click through anymore. That doesn't change what people want to know, but it does change how they find it. Your content now also has to be the source material an AI pulls its answer from.
SEO is not dead. But you’ll have to deploy it differently.
Giacomo Perticara, founder and SEO strategist at GRP Digital
That's exactly why keyword research becomes more important, not less. Map out your audience's real questions clearly and answer them completely, and both Google and AI models will see you as a trustworthy source. Thinking in topics and intents, not separate words, is exactly what works in this new playing field.
Frequently asked questions about keyword research
How long does keyword research take?
That depends on the scope of your topic. You often have a first usable picture within a day, but a complete study for an entire site takes more time. You then use the outcome for months.
Do you need a paid tool?
For a first exploration, free sources can get you far. For reliable search volume and competition data, a paid tool does help. Either way, the interpretation stays your job.
What's the difference between a keyword and search intent?
A keyword is what someone types in. Search intent is what they're trying to achieve with it. The same word can have different intents, so you always look at them together.
How often should you repeat keyword research?
Search behaviour changes, so repeat it periodically, for example a few times a year or when you launch new products. That way you stay in step with what your audience is searching for at that moment.
Can one page rank for multiple keywords?
Yes, and that's exactly the point. Terms with the same intent belong on one page, so that page gives a complete answer and doesn't compete with your other pages.
Conclusion: start with logic, not separate keywords
Keyword research isn't a word hunt, it's building logic. You start with your audience's question, organise those questions by intent and topic, and only then link everything to pages. Do it well, and you build a structure where every page has a clear job and doesn't compete with the rest.
So never look only at volume, and never at separate words. Think in topics, check the intent in the SERP, and ask yourself for every term whether your brand can really be relevant to it. That's the foundation both classic SEO and AI visibility rest on. Start with the logic, and the keywords fall into place on their own.
