Keyword Research and Topic Clusters: How to Plan Content That Ranks

Keyword research is often treated as a hunt for the biggest search volume. That approach produces traffic that does not buy and content that competes with itself. Good research is a planning exercise: it tells you what your customers are looking for, how they phrase it and which pages on your site should answer them. This guide shows a practical process, from the first list of ideas to a content map built around topic clusters.
Start from the business, not the tool
Before opening any software, list what you sell, the problems you solve and the questions customers ask before they buy. Talk to your sales and support people, and read emails and chat transcripts. These sources reveal the real language of your market, which is usually different from the industry jargon you use internally. Write the raw list down. It becomes your set of seed topics.
Expand the list with free and paid sources
You can do solid research with free sources, and paid tools add speed and scale.
- Google Search Console shows the queries that already bring impressions to your site. It is the most reliable data you have because it comes from your own results.
- Google autocomplete, related searches and People Also Ask reveal how people extend a query.
- Google Keyword Planner gives volume ranges and ideas, and works without a large ad budget.
- Competitor pages. Look at what the sites ranking for your target searches actually cover, and which headings and questions they use.
- Third-party SEO tools offer difficulty estimates, keyword clustering and competitor comparisons. Their numbers are estimates and differ from tool to tool, so use them for direction rather than precision.
Classify by search intent
Intent is more important than volume. Sort each keyword into one of four groups.
| Intent | What the searcher wants | Best page type |
|---|---|---|
| Informational | To learn or understand | Guide, article, FAQ |
| Commercial | To compare and evaluate | Comparison, case study, review |
| Transactional | To buy, book or enquire | Service, product or pricing page |
| Navigational | To reach a known site or brand | Home page, brand pages |
Check intent by reading the live results. If the top ten for a phrase are all tutorials, a service page is unlikely to rank for it, and the reverse is also true. Let the results tell you what format Google believes people want.
Judge keywords on more than volume
A useful way to prioritize is to score each candidate on four questions.
- Business value. If this visitor converted, how valuable would it be?
- Intent fit. Does the searcher match someone we can help?
- Competitiveness. Can our site realistically compete given its authority and the pages already ranking?
- Opportunity. Is there a gap, such as outdated results, thin pages or missing angles?
A search with modest volume and strong buying intent often beats a high-volume term that attracts casual browsers. Long-tail phrases, which are longer and more specific, tend to convert better and are easier to win, especially for newer sites.
Group keywords into topic clusters
Search engines understand that many different phrasings can mean the same thing, so you do not need a separate page for each variation. Group keywords that share an intent into one cluster, and give each cluster a single page. A topic cluster is then a set of related pages connected by internal links.
- The pillar page covers a core subject broadly and targets the main, higher-value term. For a business site this is often a service page.
- Supporting pages answer specific related questions in depth, such as pricing factors, comparisons, how-to guides and common problems.
- Internal links run both ways. Supporting pages link up to the pillar, the pillar links down to its supporting pages, and related supporting pages link to one another where it helps the reader.
For example, a cluster around local search might use a pillar on local SEO services supported by articles on ranking in Google Maps, reviews, citations and location pages. Each piece covers one angle and points the others toward the pillar.
Build the content map
Put everything in a simple spreadsheet with these columns: keyword group, primary keyword, intent, target URL, page type, priority and status. This map is your control panel. It shows which searches already have a page, which need one and which pages overlap. It also keeps your plan connected to your overall approach, as described in our guide to building an SEO strategy.
Prevent keyword cannibalization
Cannibalization happens when two or more of your pages target the same search and compete with each other. Rankings become unstable, and neither page performs as well as one stronger page would. To find it, search Google for site:yourdomain.com with a target phrase, or look in Search Console for queries that show several of your URLs. The fix is usually to merge the pages, redirect one to the other, or sharpen each page’s focus so they serve different intents.
Write briefs that make good content easy
A short brief for each page turns research into consistent output. Include the primary keyword and the intent, the questions the page must answer, the headings to cover, the internal links to include, examples or data to add, and the action the reader should take next. Then write for people first. Include the keyword where it is natural, such as the title, the H1, an early paragraph and a few headings, and let the rest of the language vary as it would in a real explanation.
Structure the page so it can also serve answer-style searches. Our guides to answer engine optimization and generative engine optimization show how to write direct answers and strengthen your brand signals within the same piece of content.
Review and refresh over time
Research is never finished. Search behavior shifts, competitors publish and your own pages age. Every quarter, check which pages are gaining or losing impressions, update those that have slipped, merge thin pages and add new content where the map shows a gap. Updating an existing page that already has some authority is often faster and more effective than publishing a new one.
Mistakes that waste time
- Targeting only high-volume terms and ignoring conversion potential.
- Creating one page per keyword variation instead of one page per intent.
- Trusting difficulty scores as facts instead of estimates.
- Skipping a look at the live results before choosing a page format.
- Publishing without internal links, leaving new content isolated.
A worked example: clustering for a web design studio
Suppose a small studio that builds WordPress sites wants to plan its content. Its raw list of ideas might include phrases such as WordPress web design, how much does a WordPress website cost, custom WordPress theme, WordPress website for a dentist, Elementor versus Gutenberg and WordPress maintenance plan.
Grouping by intent and topic could produce clusters like these.
| Cluster | Intent | Page |
|---|---|---|
| WordPress web design services | Transactional | Pillar service page |
| Cost and pricing of a WordPress site | Commercial | Supporting guide on cost factors, linked to the pricing page |
| Custom theme versus builder | Commercial | Comparison article |
| Industry sites such as dentists or solicitors | Transactional with a niche | One page per industry, each with relevant examples |
| Maintenance and support | Transactional | Service page for maintenance plans |
Each cluster gets one page, and the pages link to each other. The cost guide links to the service page, the comparison article links to both and the industry pages link back to the pillar. The studio now has a content plan driven by customer language rather than by a tool’s volume column.
A simple tool workflow
- Export your queries from Search Console and sort by impressions to find pages close to page one.
- Add related searches and People Also Ask questions for each core service.
- Paste the list into a spreadsheet, tag each row with intent and group similar rows.
- Use a keyword tool to add rough volume and difficulty for the groups, treating the figures as directional.
- Check the live results for the top groups to confirm the right page type.
- Assign each group a URL and a priority, then start with the highest-value gaps.
When your keyword map is ready, you may want outside help. See the types of agencies offering SEO, AEO and GEO services and AEO vs GEO optimization services compared.
Keyword research for shops works differently. See our WooCommerce SEO guide.
Frequently asked questions
How many keywords should one page target?
One main topic with its close variations. A page can rank for many related phrases if it covers the topic well, so you do not need to list dozens of exact keywords.
Is keyword density still important?
No. Use the main phrase naturally in key places such as the title, heading and opening, then write for the reader. Repeating a phrase to hit a percentage does not help and makes the text worse.
How do I find keywords with low competition?
Look for specific, long-tail phrases where the current results are weak, outdated or off-topic. Combine that with your own experience, because you can often answer niche questions better than generic pages.
How often should I redo keyword research?
Review quarterly using fresh Search Console data and do a deeper pass once or twice a year, or whenever you launch a new service. Our SEO strategy guide shows where this fits in a wider plan.
Related: how long SEO takes, AI SEO services.
A simple starting routine
- List your services, customer problems and common questions.
- Pull existing queries from Search Console and expand them with autocomplete and related searches.
- Classify every keyword by intent and group the similar ones.
- Assign each group to one page, new or existing.
- Score and prioritize by value, fit, competitiveness and opportunity.
- Write briefs, publish, link the pages together and review quarterly.
If you would like research and a content map prepared for your business, see our SEO services or contact us. We will show you where the most valuable opportunities are.