Keyword research isn’t a list of words — it’s a map of how your customers think
Keyword research is often talked about as a technical exercise: open Google Keyword Planner, look up search volumes, pick the keywords with high volume and low competition, deliver a report, commission the content, and hope for the best.
That approach rarely produces meaningful results, because it skips the question that actually matters — what is the potential customer really trying to solve at the moment they type that keyword into Google?
At its best, keyword research is a systematic way to understand how your customers think. It reveals the problems they’re trying to solve, how they talk about those problems in their own words, where they are on the buying journey, and how ready they are to make a decision. In this article I’ll cover keyword research from exactly that angle: as a strategic tool that informs business decisions.
Search intent matters more than search volume
Traditional keyword research focuses on two numbers: how many times a keyword is searched per month and how tough the competition is. Those numbers are useful, but they only tell part of the story.
The more important question is search intent — what the searcher actually wants to find. Google groups search intent into four main types, and each one calls for a different kind of content.
In an informational search, the user is looking for information. They want to understand something or learn something new. Examples include “what is search engine optimization” or “how to start an online store”. These searches typically sit at the early end of the buying journey, and they’re how you build awareness and trust.
In a navigational search, the user is looking for a specific website or service. They already know what they want and just need the right address. These searches are hard to capture unless the brand being searched is yours.
In a commercial search, the user is comparing options. They’ve identified the need and they’re looking for the best solution. “Best ecommerce platform for SMEs” or “accounting service price” are typical commercial searches. These are valuable, because the searcher is close to a decision.
In a transactional search, the user is ready to act. “Buy a website for my business” or “order an SEO audit” are searches where the conversion probability is highest. The volume on these is often smaller, but each visitor is significantly more valuable.
In practice this means a keyword with 50 monthly searches and commercial intent can be more valuable to your business than a keyword with 5,000 searches but purely informational intent. Search volume doesn’t tell you the value — intent does.
How keyword research actually gets done
Keyword research moves forward in stages, and each stage builds on the previous one. The process starts with understanding the business and ends in a concrete action plan.
Starting point: the business goal
Before you open a single tool, you need to understand which business goal the keyword research is serving. Do you want more leads for a specific service? Do you want to grow online-store sales in a certain product category? Do you want to build expert authority in a particular topic area? The goal shapes what kinds of keywords you look for and how you prioritise them.
Collecting seed keywords
The process starts with seed keywords — the broader terms that describe your services and your customers’ problems. You collect these from several sources: the company’s own expertise, the language customers actually use (sales conversations, customer feedback, support requests), competitors’ websites, and Google’s autocomplete.
Google autocomplete is a surprisingly effective way to find real search terms. As you start typing a keyword into Google, it suggests the most common searches. Those suggestions are based on real search data, so they tell you directly what people are actually searching for. The “People also ask” box on the results page is similarly useful — it surfaces related questions your content can answer.
Gathering and analysing data
Once the seed keywords are in place, they go into tools that return a wider keyword set with volumes and competition data. Google Keyword Planner is a free starting point, but professional keyword research typically uses Ahrefs, Semrush, or a similar tool that offers more accurate search volumes, keyword difficulty estimates, and competitor visibility analysis.
Google Search Console is especially valuable, because it shows you which keywords already bring traffic to your site. You’ll often find keywords where your site is showing up in the results but ranking at, say, position 8 or 15. These are low-hanging opportunities: a small content improvement can lift the ranking significantly and bring in noticeably more traffic.
Grouping and prioritising keywords
A bare list of keywords isn’t useful on its own. The keywords need to be grouped into topic clusters and prioritised by business value.
Topic clusters mean gathering keywords that belong to the same subject area. For example, “search engine optimization price”, “SEO service for SMEs”, and “benefits of SEO” all belong to the same cluster, even though they’re different keywords. One comprehensive piece of content can answer several keywords in the same cluster.
Prioritisation looks at three factors for each cluster: business value (how close to a decision the searcher is), search volume (how many searches there are in total), and competition (how hard it is to reach the top of the results). The best situation is finding clusters where all three line up — high value, reasonable volume, and manageable competition.
Long-tail keywords: small volumes, big results
One of the most important insights in keyword research is the value of so-called long-tail keywords. These are longer, more specific search phrases whose individual volume is small but which together make up a significant share of all searches.
“Websites” is a short, generic keyword with high volume but tough competition and unclear intent. “Websites for a small business price” is a long-tail keyword with much smaller volume but clear commercial intent and less competition. The searcher knows what they want and they’re ready to make a decision.
In practice, this means a company that produces content for tens or hundreds of long-tail keywords can capture more valuable traffic in total than a company trying to rank for one or two generic keywords. At the same time the competition is easier to win, because fewer competitors are optimising for those specific phrases.
From keyword research to an action plan
The real value of keyword research only shows up when it’s translated into a concrete action plan. A report without execution is just wasted resources.
The action plan answers three practical questions: which content gets created (new pages and articles), which existing content gets improved, and on what timeline the work happens.
For new content, keyword research tells you which topics to cover, which search intent to target, and which angle is worth emphasising. If a competitor analysis shows that the top results for a topic are surface-level overviews, an in-depth expert article can stand out clearly.
Improving existing content is often a faster path to results than creating brand-new pieces. The keywords in Google Search Console where your site already ranks at position 5 or below are concrete improvement targets. Expanding the content, sharpening the headline, or adding a missing section can lift the ranking by a step or two — which can multiply the traffic to that page.
Timing matters, because SEO is cumulative work. One article a month adds up to twelve new pieces of content a year, all of them bringing in organic traffic. Two years in, you have twenty-four optimised pieces, and each one strengthens the site’s overall authority.
How AI is changing how keyword research itself gets done
AI tools have meaningfully changed how keyword research is done in practice. Semantic analysis identifies terms and concepts related to a topic that pure search-volume lookups don’t reveal. Instead of producing a list of individual words, keyword research now produces a map of interconnected topics that you can use to build content covering a subject comprehensively.
At the same time, the kinds of keywords worth investing in have shifted. Because AI search engines answer more and more basic informational questions directly, traditional “what is X” keywords generate fewer clicks than they used to. The emphasis of keyword research moves toward keywords where the searcher genuinely wants to visit a site — comparison searches, price searches, location-specific searches, and other searches where an AI summary alone isn’t enough of an answer.
How to recognise quality keyword research
Not all keyword research is equal. Quality keyword research differs from a surface-level report in several ways.
Quality keyword research starts from the business goals, not from search volumes. It doesn’t list every possible keyword — it prioritises them by business value. It includes a search-intent analysis for every cluster, not just volume numbers. It tells you which content is worth creating and which is worth improving, not just which keywords exist. It treats the competitive situation realistically and identifies opportunities where a smaller player has a real chance to win.
A surface-level keyword research report is typically a list of keywords with volumes and no context for how to use them. It focuses on high-volume keywords without considering the competition, and skips search intent entirely. That kind of report looks comprehensive but doesn’t lead to concrete action.
Keyword research is an investment that informs all your content production
Keyword research isn’t a one-off project that gets done once and then filed away. It’s a living strategic tool that informs all of your digital content production.
When keyword research is done carefully, every new piece of content has a clear purpose: it answers an identified search need, it serves a specific audience at a specific stage of the buying journey, and it builds the site’s overall authority on a plan. Without keyword research, content production is guesswork. With it, content production becomes systematic work toward measurable results.