A Practical Keyword Research Guide Built on Real Search Demand

Keyword research often gets reduced to a spreadsheet of high-volume phrases that nobody on the team actually uses to find a business like yours. The result is content that chases traffic instead of customers. Good keyword research starts somewhere else: with the questions and phrases real people type when they have a problem you can solve.

This guide walks through a process you can repeat. It covers the intent behind searches, the tools worth using, how to organise what you find into a usable map, and how to decide what to work on first. The aim is a list grounded in genuine search demand, not vanity metrics.

Start with search intent, not volume

Every search carries an intent. Understanding it tells you what kind of page to build and whether a searcher is anywhere near making a decision. There are four common intent types.

• Informational: the person wants to learn something. Phrases like "how does SEO work" or "what is a meta description". They are not ready to buy, but they are forming opinions.
• Commercial: the person is comparing options before a purchase. Think "best CRM for small business" or "SEO agency vs in-house". These searches signal active consideration.
• Transactional: the person is ready to act. "Hire SEO consultant Sydney" or "buy standing desk online" fall here. Conversion potential is high.
• Navigational: the person is looking for a specific brand or page, such as "Control Tower contact" or "Xero login".

Matching intent to page type matters more than matching the exact keyword. An informational query deserves a guide or explainer. A transactional query deserves a service or product page. Put a sales page in front of someone who wanted to learn, and they bounce.

Build a seed list from what you already know

Before you open a single tool, write down the obvious. List your services, the problems you solve, the words customers use in sales calls and support tickets, and the way you would describe your offer to a stranger. This seed list is the foundation everything else expands from.

Then add adjacent language. Customers rarely use the same internal jargon you do, so capture the plain-English versions too. A firm might call it "managed search marketing" while customers search "help getting my website on Google".

Use the right tools in the right order

No single tool gives you the whole picture. Combine a few and cross-check what they tell you.

Google Search Console

If your site has any history, Search Console is the most honest source you have. It shows the actual queries that already bring impressions and clicks to your site, plus your average position for each. Look for queries where you rank on page two: these are real demand you are close to capturing. Search Console reflects your audience specifically, not a generic estimate.

Google Keyword Planner

Keyword Planner, part of Google Ads, gives volume ranges and related terms straight from Google. The volume figures are broad, especially without an active ad spend, but it is reliable for discovering related phrases and gauging relative popularity. Treat the numbers as directional rather than exact.

Third-party tools

Tools such as Ahrefs, Semrush, and similar platforms add keyword difficulty scores, competitor keyword gaps, and richer related-term suggestions. Their difficulty metrics are estimates based on the strength of pages currently ranking, so use them as a guide rather than gospel. Their real strength is showing you what competitors rank for that you do not.

Search engine results pages

The results page itself is free research. Look at what currently ranks, the "People also ask" boxes, related searches at the bottom, and autocomplete suggestions. These reveal how Google interprets a query and what subtopics it expects a strong page to cover.

Group keywords into topics and a map

A flat list of hundreds of keywords is hard to act on. Group them into topics, where one core idea gathers all the variations and related questions around it. People search the same need in dozens of ways, and a single well-built page can satisfy many of those variations at once.

For each topic, decide:

• The primary phrase that best represents the topic and its dominant intent.
• Supporting phrases and questions the page should address.
• The page type that fits the intent (guide, service page, comparison, FAQ).
• Whether it belongs to an existing page or needs a new one.

This becomes your keyword map: a document that ties topics to specific URLs. It prevents two pages competing for the same query and shows where content gaps sit. Keep it somewhere the whole team can see, because it doubles as a content plan.

Prioritise by value, not just difficulty

Once you have a mapped list, you need an order. Three factors drive priority.

1. Intent and commercial value: a lower-volume transactional phrase often beats a high-volume informational one. Twenty searches a month from people ready to hire can be worth more than two thousand casual browsers.
2. Difficulty versus your authority: be realistic about what you can rank for now. A new site competing for a fiercely contested head term will struggle for a long time. More specific, longer phrases usually have less competition and clearer intent.
3. Relevance to what you actually offer: ranking for something you cannot serve well brings traffic that never converts and can hurt engagement signals.

A simple way to sort is to plot each topic against effort and expected return. Quick wins, things close to ranking with clear commercial value, go first. High-effort, high-value topics become longer-term projects. Low-value items drop off the list.

Avoid vanity keywords

Vanity keywords are the broad, high-volume terms that look impressive in a report but rarely deliver business. "Marketing" or "shoes" might pull huge numbers, yet the intent is impossible to read and the competition is enormous. Chasing them ties up resources for traffic that does not convert.

Watch for these traps:

• Picking a keyword purely because the volume is large.
• Ignoring intent because a phrase sounds relevant.
• Targeting terms a far larger, more established competitor dominates.
• Building content around words your customers never actually use.

The better instinct is to ask whether a search reflects a person who could become a customer. If the answer is unclear, the keyword probably is not worth the effort yet.

Keep the research alive

Keyword research is not a one-off task. Search behaviour shifts, new questions emerge, and your own Search Console data keeps revealing fresh opportunities. Revisit your map every few months. Promote queries that are climbing, refresh pages slipping in position, and prune topics that never gained traction.

If you want a keyword strategy that maps cleanly to real demand and to pages that convert, Control Tower can help you build and prioritise one. The goal is always the same: research that turns into traffic that turns into customers, not just numbers on a page.