A kitchen appliance business might put “induction hob” on a keyword list. But a person using those words could want to buy one, arrange installation, or understand why an existing hob is not heating. One broad phrase does not tell you which page to make.
Keyword research is the process of finding the words people use in search, understanding what they need, and choosing which needs your website can answer. The useful output is not a long spreadsheet. It is a short list of page decisions you can explain.
What Is Keyword Research?
A search query is what someone actually types. A keyword is a term you choose to research or target. The two can be identical, but one useful page may answer several differently worded queries.
Research therefore has three jobs: discover the language people use, work out the task behind each search, and decide whether an existing or new page should answer it. It does not mean repeating an exact phrase until a page ranks.
Why Does Keyword Research Matter?
Without research, it is easy to write for a term your customers do not use, or attract people whose needs your business cannot meet. A broad appliance term may bring many different searches. A specific installation question may reach fewer people but be a better fit for an installation service.
Keyword research helps you choose topics, name pages in language readers recognize, and spot gaps in your site. It is one part of SEO, not a promise of traffic: a useful page must still be accessible, credible, and suited to the search.
Where Can You Find Keyword Ideas?
Start with Customer Language
Choose one product or service. Write down questions from enquiries, support messages, site search, or conversations with customers. Keep their wording before replacing it with industry terms. “Hob not heating” and “induction cooker repair” may describe related needs, but the phrasing tells you how people frame the problem.
If you have no customer history yet, list the questions someone must answer before buying or booking. Treat these as hypotheses to check, not proof that people search for them.
Check Searches Your Site Already Appears For
If your site has Google Search Console data, open the Search results Performance report and inspect queries and pages together. A query that brings impressions to an existing page may show language you overlooked. Open the page before deciding whether to improve it or create another one.
Search Console does not show every query: some are omitted for privacy, and the table keeps only a portion of the data. Use it as evidence about your own site, not a complete list of market demand.
Expand the List Without Treating Every Number as Fact
Search suggestions and related searches can reveal additional wording. Keyword Planner can filter keyword ideas by language and estimate average monthly searches for a selected location. Google says that search count includes the term regardless of language. Google Trends helps compare relative interest over time; its 0–100 scale is not monthly search volume.
These tools answer different questions. An autocomplete suggestion is an idea, not a popularity ranking. Keyword Planner is built for advertising; its “Competition” column measures advertiser competition, not how hard it is to rank organically. Record the location, settings, date, and tool beside any number you keep. See how to read search volume and keyword difficulty before using those estimates to choose a page.
Competitor pages can also reveal candidate topics. Use competitor keyword gap analysis to decide which of those topics actually fit your site.
How Do You Choose Keywords Worth Targeting?
For each small group of related terms, ask four questions:
- Is it relevant to what you offer? A kitchen appliance seller should not chase every cooking query just because it has a large reported volume.
- What is the searcher trying to do? Check the results in your target market. A product listing, troubleshooting guide, and installation service answer different tasks.
- Can you provide a better answer? Check whether you can show the models, conditions, evidence, or service details the task requires. If you cannot serve the need, leave the term off your plan.
- Is the opportunity worth doing now? Use volume and difficulty estimates as context, then weigh business fit, effort, and the quality of pages already available. A low estimate does not make a highly relevant need worthless.
For a multilingual Malaysian site, check English, Malay, and Chinese terms separately. People may use different product names, and the results can differ by language and location. Translating an English keyword list word for word is not research. For narrower questions, see how to assess long-tail keywords. Before committing to a page, inspect the search results for that topic.
Turn Related Searches into a Page Plan
Imagine a Kuala Lumpur business that sells and installs induction hobs, but does not repair them. We start with six candidate phrases and check the kinds of pages surfaced by a public English web search on 2 October 2026. This is a page-planning exercise, not a client case study, a Google ranking report, or search-volume data.
Swipe across the table to see each decision.
| Candidate search | What the results suggest | Decision |
|---|---|---|
| buy induction hob Malaysia | People are comparing products. | Keep: product category page. |
| induction hob price Malaysia | Product listings show prices and models. | Group with the buying search; show prices on the category and product pages. |
| induction hob installation Kuala Lumpur | People are looking for local installation. | Keep: installation service page. |
| install induction cooker KL | The task is still finding installation help. | Group with the installation search if a closer results check confirms the same need. |
| induction hob not heating | Manufacturer help pages address faults. | Leave out: this business does not offer repairs or troubleshooting. |
| induction hob recipes | Cooking guides and recipes serve another task. | Defer: it does not help this first product-and-installation plan. |
The first plan needs two pages: a product category covering models, prices, and dimensions; and an installation service page covering the area, requirements, scope, and booking. Improve an existing page if one already answers the task. There is no reason to create a separate price page, a second page just for “KL,” or a repair page for a service the business does not provide. See how to map each search group to an owner page for the next step.
A Simple First Pass
Start with one offer and collect ten real or plausible questions. Check your own Search Console data if available, then inspect search results for the strongest candidates. Group phrases that need the same answer. For each group, write down an existing URL to improve, a new page to create, or a reason to postpone it.
Choose one page you can make genuinely useful now. If several candidates remain, use a keyword-prioritization queue to decide which to tackle first, then review its on-page SEO. A small, justified map is more valuable than hundreds of unassigned keywords.
Frequently Asked Questions
Can You Do Keyword Research for Free?
Yes. Customer questions, your own Search Console data, search results, and free access to tools such as Google Trends can support a first plan. Paid tools can broaden the data, but they do not replace checking relevance and intent.
Should You Always Choose the Highest-Volume Keyword?
No. A large estimate can hide several unrelated needs. Choose terms you can answer well and that connect to your audience; use volume as one signal rather than the decision itself.
Does Every Keyword Need Its Own Page?
No. Several phrasings can share a page when they ask for the same answer. Create separate pages when the task, required information, or next action is meaningfully different.
Which Keyword Should a New Website Target First?
Start with a specific search closely tied to a product, service, or topic you can cover thoroughly. Check the results and build the right page before expanding into broader terms.