How Do You Do Keyword Research for SEO? A Beginner's Guide

To do keyword research, list the topics your customers care about, find the words they use to search for them, and group those terms by what the searcher actually wants. Then choose the terms you can realistically serve with a useful page. Search volume is only one input. Intent, relevance and competition usually matter more.

What is keyword research, and why does intent matter more than volume?

Keyword research is the process of finding the phrases people type or say into search engines, so you can create content that answers them. Done well, it tells you what to write, what to put on each page and which topics deserve priority.

Beginners often sort a list by volume and chase the biggest numbers. The problem is that a high-volume phrase can bring visitors who never wanted what you sell. A smaller phrase that clearly signals a need can bring fewer visitors who are far more likely to enquire.

Take the phrase "accountant". It could come from someone hiring, someone studying or someone looking for a job. "Accountant for small limited company" is narrower, but it tells you what the person needs.

What are the main types of search intent?

Most searches fall into one of four groups:

• Informational: the person wants to learn something, such as "how does a content audit work".
• Commercial: the person is comparing options, such as "best project management software for small teams".
• Transactional: the person is ready to act, such as "book a website audit" or "buy running shoes".
• Navigational: the person wants a specific site or brand, such as a company name.

Each type needs a different page. A guide suits informational searches. A comparison or service page suits commercial ones. A clear offer with an easy way to get in touch suits transactional ones. Matching the page to the intent is often what separates content that ranks and converts from content that does neither.

How do you do keyword research step by step?

1. Start with your customers, not a tool

Write down the problems you solve, the questions customers ask before buying and the words they use. Sales emails, enquiry forms and common phone questions are good sources. Customers rarely use your industry jargon.

2. Build a seed list

Pick five to ten broad topics. For each, add variations: problems, features, locations, audiences and questions. These seeds feed whichever research tool you use.

3. Expand with tools

Keyword tools suggest related terms and show rough demand. Free sources are useful for a first pass. Google Trends shows how interest in a topic changes over time and compares phrases. Autocomplete suggestions and the "People also ask" boxes on a results page show real wording.

4. Check the results page for each phrase

Search the phrase yourself. Look at what ranks. If the top results are all product pages, a blog post is unlikely to compete. If they are all guides, a sales page will probably struggle. The results page is the clearest evidence of what Google thinks the searcher wants.

5. Group keywords by topic and intent

Put similar phrases together and assign each group to one page. Avoid creating several pages that target the same intent, as they can compete with each other.

6. Prioritise

Score each group on three things: how relevant it is to what you offer, how likely the searcher is to become a customer, and how realistic it is to rank. A newer site usually does better starting with specific, lower-competition phrases and building from there.

How can AI speed up keyword research?

AI tools are good at the slow, repetitive parts of the job. Used sensibly, they can:

• Generate large lists of related phrases and questions from a single seed topic.
• Cluster thousands of keywords into topic groups in minutes.
• Classify phrases by likely intent.
• Summarise what the top-ranking pages have in common, such as format, length and subtopics covered.
• Spot gaps where competitors cover a subject and you do not.

This turns days of spreadsheet work into a first draft you can review the same afternoon. It also helps when you are researching an unfamiliar sector and do not yet know the language.

Where does human judgement still matter?

AI produces plausible lists, but it does not know your business. Several decisions still need a person:

• Relevance: a phrase may be popular and correctly classified, yet be a poor fit for your offer, location or capacity.
• Accuracy: AI can invent data or present guesses as facts, so check volumes and claims against a reliable source. Our article on why AI makes things up explains why.
• Commercial sense: only you know which services are most profitable, which customers you want and what you can deliver.
• Brand voice and angle: two businesses can target the same phrase and need very different pages.
• Reading the results page: judging whether you can genuinely beat what is already ranking takes context.

A useful rule is to let AI widen the list and let people narrow it.

What should you do after choosing your keywords?

Map each keyword group to a page, write for the intent behind it, and use the main phrase naturally in the title, headings and opening paragraph. Do not repeat it unnaturally. Then track progress. Google Search Console shows which queries already bring people to your site, which is often the best source of new ideas. Revisit your list regularly, as search habits change.

It is also worth writing in the conversational way people now ask questions. Our piece on appearing in Google's AI Overviews covers how clear, direct answers help.

Frequently asked questions

How many keywords should I target per page?

Target one main topic per page, with a handful of closely related phrases that share the same intent. Trying to cover several different intents on one page usually weakens all of them.

Are long-tail keywords better for beginners?

Often, yes. Longer, more specific phrases tend to face less competition and show clearer intent, so they are a sensible starting point for a newer site.

Do I need paid tools to do keyword research?

Not to begin with. Autocomplete, related searches, Google Trends and Search Console can give you a workable list. Paid tools add convenience and detail as your needs grow.

How often should I repeat keyword research?

Review it at least a few times a year, and whenever you launch a new service or notice a change in the questions customers ask.

Can AI do keyword research on its own?

AI can do a lot of the groundwork, but its output should be checked. Someone who understands your business needs to decide which keywords are worth pursuing.

Next steps

Virtuance Digital Marketing offers GEO/SEO, which includes AI-enhanced keyword research and real-time content and technical audits. If you would like to talk about your own keyword strategy, email contact@virtuance.co.uk or call +44 7852 620882.