Keyword Research: Stop Guessing What Your Customers Search For

Keyword research is the process of identifying the words and phrases people type into search engines when looking for products, services, or information related to your business. Done well, it tells you where demand exists, how competitive a space is, and which terms are worth your time and budget. Done poorly, it produces a spreadsheet full of numbers that bears no relationship to how your customers actually think.

Most teams land somewhere in the middle. They run a tool, export a list, and start writing. The problem is that keyword research is not a data collection exercise. It is a customer understanding exercise that happens to involve data.

This article is part of the Complete SEO Strategy Hub, which covers everything from technical foundations to content planning and link acquisition. If keyword research is where you are starting, you are in the right place. But it connects to a much larger system.

What Does Keyword Research Actually Tell You?

The basic output of any keyword research process is a list of terms with associated data: monthly search volume, keyword difficulty, cost-per-click, and sometimes trend data. That is what the tools give you. What the tools cannot give you is judgement about which of those terms represent genuine commercial opportunity for your specific business.

I have sat in enough strategy sessions to know that keyword data gets misread constantly. A team sees a term with 40,000 monthly searches and assumes that is the prize. What they miss is that the top ten results are all major publications with domain authority scores north of 80, that the term is informational rather than commercial, and that ranking for it would bring in traffic that has zero intention of buying anything. The number looks exciting. The opportunity is not.

Keyword research tells you three things when you read it correctly. First, it tells you what language your customers use, which is not always the language your marketing team uses internally. Second, it tells you where demand exists and roughly how much of it there is. Third, it tells you how hard it will be to compete for a given term. Everything else is interpretation.

The tools themselves are worth understanding clearly. Semrush has a thorough breakdown of the main keyword research platforms and what each one does well. Most experienced SEOs use two or three tools in combination because no single source has complete data. Google’s own tools, particularly Search Console for existing sites, are underused relative to their value. They show you what real users are already searching to find your content, which is more reliable than any third-party estimate.

Why Search Intent Is the Variable That Changes Everything

Search intent is the reason someone typed a query. It sounds simple, and the concept is straightforward, but getting it wrong is the single most common reason keyword strategies fail to convert.

There are four broadly recognised intent categories: informational (the user wants to learn something), navigational (they are trying to find a specific website or brand), commercial investigation (they are comparing options before making a decision), and transactional (they are ready to buy or act now). Each category demands a different type of content, and ranking with the wrong content type for a given intent is a waste of everyone’s time.

When I was building out the SEO programme at iProspect, one of the early discipline shifts we made was forcing intent classification before any content brief was written. It sounds obvious in retrospect, but the team had been writing informational blog posts to target transactional queries and wondering why the conversion rate was flat. The traffic was fine. The intent match was not.

The fastest way to assess intent is to search the term yourself and look at what Google is already ranking. Google has done the intent classification for you. If the first page is all product pages and e-commerce listings, the intent is transactional. If it is all how-to guides and explainer articles, it is informational. Try to rank a product page against a page of guides and you will struggle regardless of how good your on-page SEO is. The format matters as much as the keyword.

This is also where the relationship between SEO and paid search becomes genuinely useful. PPC testing can sharpen keyword decisions for SEO because paid campaigns generate real conversion data quickly. If a term converts well in paid search, that is a strong signal it carries commercial intent worth targeting organically. If it burns budget without converting, that is useful information before you invest six months of content effort chasing the same term.

How to Build a Keyword List Without Drowning in Data

The practical process of keyword research has a few distinct stages. Most guides cover the mechanics well enough. What they underemphasise is the filtering and prioritisation phase, which is where the real decisions get made.

Start with seed keywords. These are the broad terms that describe your business, your products, or your services. They are not the terms you will necessarily target directly, but they are the starting point for expansion. If you run a B2B software company, your seeds might be the core problem your product solves, the job title of your primary buyer, and the category name your industry uses. Three to five seeds is usually enough to generate a working list of several hundred candidate terms.

From there, use a keyword tool to expand each seed into related terms, questions, and variations. Crazy Egg has a solid walkthrough of the keyword research process that covers the expansion phase in detail. The output at this stage will be large. That is fine. You are building a raw material list, not a final strategy.

The filtering stage is where judgement comes in. Apply four filters in sequence. First, relevance: does this term relate to something your business actually does or sells? Second, intent: does the intent behind this term match the type of content you can credibly produce? Third, competition: is the difficulty level realistic given your current domain authority and content programme? Fourth, volume: is there enough search demand to justify the effort? Note that volume comes last. It is a tiebreaker, not a primary filter.

After filtering, group related terms by topic cluster rather than treating each keyword as an isolated target. A cluster might have one primary term and four to eight supporting terms that share similar intent. That cluster becomes the basis for a content piece or a page, with the primary term as the main target and the supporting terms addressed naturally within the content.

The Long-Tail Argument: Why Smaller Numbers Often Mean Better Business

Long-tail keywords are specific, lower-volume search phrases that typically indicate clearer intent than broader head terms. The phrase “accountant” has enormous search volume and almost no actionable intent signal. The phrase “accountant for freelance graphic designers in Manchester” has a fraction of the volume and tells you almost everything you need to know about the searcher and what they need.

The commercial logic for prioritising long-tail terms is straightforward, and I have seen it play out consistently across industries. When I was working with a professional services client early in my agency career, we shifted the entire content programme away from broad category terms toward specific, intent-rich phrases. Traffic dropped in the first quarter. Enquiries went up. Revenue from organic search increased significantly. The broad terms had been generating visitors who were nowhere near a buying decision. The long-tail terms were pulling in people who already knew what they wanted and were looking for someone to deliver it.

Long-tail keywords are also more achievable for most businesses. If your site is relatively new or your domain authority is modest, competing for a head term against established players is a slow and often fruitless exercise. Long-tail terms with lower difficulty scores let you build ranking momentum, generate early traffic and conversions, and establish topical authority that eventually supports your ability to rank for more competitive terms.

The proportion of long-tail to head terms in your strategy should reflect your current competitive position. A brand new site in a competitive vertical should be weighted heavily toward long-tail. An established site with strong authority can afford to pursue a broader mix. Neither extreme is right for every situation, but most businesses underinvest in long-tail relative to its actual return.

Keyword Research for Different Business Types: It Is Not One Size

The keyword research process shares the same underlying logic across business types, but the execution varies significantly depending on who you are and who you are selling to.

For local businesses, geography changes everything. A plumbing company in Birmingham is not competing for national search volume. It is competing for a specific set of location-modified terms across a defined service area. The keyword strategy needs to reflect that. If you are working through local SEO for a trades business, the local SEO approach for plumbers is a useful reference for how geography, service terms, and intent interact in a local search context. The same principles apply across trades and professional services operating in defined geographic markets.

Healthcare and regulated sectors add another layer of complexity. A chiropractic practice, for example, needs to balance informational terms that build trust and explain treatments with local terms that pull in patients actively looking for a provider. The SEO approach for chiropractors illustrates how keyword strategy in healthcare needs to account for patient intent at different stages of the decision process, from initial symptom search through to provider selection.

For B2B businesses, the keyword landscape reflects longer buying cycles and multiple stakeholders. The terms a CFO searches are different from the terms an IT manager searches, even if they are both evaluating the same product. A strong B2B keyword strategy maps terms to buyer personas and stages of the purchase experience rather than treating the audience as monolithic. If you are working with a specialist in this space, understanding what a B2B SEO consultant actually does, and how they approach keyword strategy differently from a generalist, is worth understanding before you engage external support.

E-commerce keyword research is different again. Product pages, category pages, and informational content each need their own keyword approach. Product pages target transactional terms. Category pages target broader commercial investigation terms. Blog content targets informational queries that build brand awareness and topical authority. Running these three content types with the same keyword logic produces mediocre results across all three.

How to Read Keyword Difficulty Without Being Misled

Keyword difficulty scores are estimates. Every tool calculates them differently, and none of them are measuring exactly the same thing. What they are approximating is how hard it will be to rank on page one for a given term, based primarily on the authority of the sites currently ranking and the number of backlinks pointing to those pages.

The number is useful as a rough signal, not as a precise measurement. I have seen terms with high difficulty scores where the actual ranking pages were thin, poorly optimised, and holding their position more through domain authority than content quality. In those cases, a well-researched, well-structured piece from a site with reasonable authority can outperform the difficulty score’s implied ceiling. I have also seen terms with moderate difficulty scores where every ranking page was exceptional, the competition had clearly invested heavily in the topic, and even excellent content struggled to break through.

The more reliable approach is to look at the actual pages currently ranking rather than relying solely on the difficulty score. Open the top ten results and ask: how strong is this content? How many backlinks does each page have? Is the ranking domain a direct competitor or a tangential player? Are there gaps in what the current results cover? If the existing content is weak and the backlink profiles are modest, a high difficulty score matters less. If the existing content is genuinely excellent, a moderate difficulty score should give you pause.

Understanding how Google’s search engine evaluates and ranks content is foundational here. Keyword difficulty is a proxy for competitive intensity, but Google is in the end making a relevance judgement. The site that best answers the searcher’s intent will rank. That is a content quality and relevance question as much as it is an authority and backlinks question.

The Role of Competitor Keyword Analysis

One of the most efficient ways to accelerate keyword research is to analyse what your competitors are already ranking for. This is not about copying their strategy. It is about understanding the competitive landscape and identifying gaps, overlaps, and opportunities you might have missed.

Most keyword tools allow you to enter a competitor’s domain and see which terms they rank for, their estimated traffic, and the pages generating that traffic. This is genuinely useful data. It tells you which terms the market has already validated as worth pursuing, and it shows you where your competitors are investing their content effort.

The more valuable analysis is gap identification. Which terms are your competitors ranking for that you are not? Of those, which ones are relevant to your business and achievable given your current position? That intersection is your opportunity list. It is more targeted than starting from scratch, and it is grounded in real competitive evidence rather than theoretical demand.

There is a version of competitor analysis that goes wrong, and I have watched it happen more times than I would like. A team identifies a competitor ranking for a high-volume term, assumes they should target the same term, and builds a content plan around it without asking whether the competitor’s position is actually generating commercial return. Ranking for a term and converting from that ranking are two different things. The competitor may be ranking for a term that drives traffic but no revenue. Chasing their traffic numbers without understanding their conversion picture is a misdirection of effort.

Connecting Keyword Research to Content Planning

Keyword research without a content plan is just a spreadsheet. The point of the exercise is to produce content that ranks, attracts the right visitors, and converts them into customers or enquiries. That requires a clear line from keyword to content brief to published piece.

The connection point is the content brief. A good brief translates keyword data into content direction: the primary term to target, the supporting terms to address, the intent the piece needs to satisfy, the format that best fits that intent, the word count range appropriate for the topic, and the specific questions the content needs to answer. Without that translation step, writers produce content that may be good but is not strategically aligned to keyword targets.

Topic clusters are the structural approach that makes this work at scale. Rather than treating each keyword as a standalone target, you organise related keywords into clusters, each anchored by a pillar page covering the broad topic and supported by more specific pieces addressing related subtopics. The pillar page targets the head term. The supporting pieces target long-tail variations. Internal links connect them, signalling to Google that your site has genuine depth on the topic.

Moz’s product mindset approach to SEO strategy is worth reading alongside any content planning exercise. The core argument, that SEO should be treated with the same rigour and user focus as product development, applies directly to how keyword research feeds into content decisions. You are not producing content for its own sake. You are solving a user problem at a specific stage of their experience, and the keyword is the signal that tells you what that problem is.

The cadence of content production matters too. Keyword research generates a prioritised list of opportunities, but executing against that list requires a realistic production schedule. I have seen teams produce 40-page keyword strategies that generate one piece of content per quarter because no one connected the research to a workable production plan. The research is only as valuable as the content it eventually produces.

Keyword Research and Paid Search: Using Both Sides of the Data

If your business runs paid search alongside SEO, you have access to conversion data that most purely organic keyword strategies lack. Paid search tells you which terms actually convert, not just which ones attract clicks. That is a different and more commercially useful signal than organic traffic volume alone.

The practical approach is to use paid search as a testing ground for organic keyword priorities. Before committing to a six-month content programme targeting a set of terms, run those terms in paid search for four to six weeks. Look at the conversion rate, the cost per conversion, and the quality of the traffic. If the paid data shows strong conversion performance, those terms deserve priority in your organic strategy. If the paid data shows high click volume but weak conversion, be cautious about investing heavily in organic content for the same terms.

Semrush’s guide to PPC keyword research covers the mechanics of paid keyword strategy in detail. Even if your primary focus is organic, understanding how paid keyword research works helps you read the data more intelligently when you use it to inform SEO decisions.

The reverse flow is equally valuable. Terms that are performing well organically, generating traffic and conversions without paid spend, are worth protecting in paid search when organic rankings are volatile or during periods when you need to defend visibility against aggressive competitors. Knowing which terms are commercially valuable from your organic data makes your paid strategy more targeted and more efficient.

How Often Should You Revisit Your Keyword Strategy?

Keyword research is not a one-time exercise. Search behaviour changes, markets evolve, new competitors emerge, and Google’s understanding of topics and intent shifts over time. A keyword strategy built two years ago and never reviewed is almost certainly leaving opportunities on the table and potentially targeting terms that no longer reflect how your customers search.

The practical cadence for most businesses is a light quarterly review and a more substantive annual refresh. The quarterly review looks at performance data from Search Console and your analytics platform: which terms are generating impressions, which ones are converting, which ones have dropped in ranking, and whether any new terms are appearing in your data that were not in your original list. The annual refresh goes back to first principles: are the seed keywords still right, have competitor positions shifted significantly, and are there new topic areas your business should be pursuing?

New product launches, market expansions, and significant changes in your competitive set all warrant an out-of-cycle keyword review. If you launch into a new vertical or a new geography, the keyword landscape for that area needs to be mapped from scratch rather than assumed to mirror your existing market.

There is also a monitoring layer that sits between reviews. Tracking your ranking positions for priority terms on an ongoing basis lets you catch drops early, before they translate into significant traffic loss. Most keyword tools offer rank tracking as a standard feature. The discipline is in actually acting on the data rather than watching rankings decline without response.

The Backlink Dimension: Why Keywords and Link Acquisition Are Connected

Keyword research and link building are often treated as separate workstreams, and in execution they often are. But they are strategically connected in ways that matter for how you approach both.

The pages that rank for competitive terms almost always have strong backlink profiles. That is not a coincidence. It is how Google’s ranking algorithm works. Backlinks from authoritative external sites signal that your content is credible and worth surfacing. For competitive keywords, content quality alone is rarely sufficient. You need links pointing to the pages you want to rank.

This means your keyword prioritisation should factor in your link acquisition capacity. If you have no active link building programme, targeting highly competitive terms is an exercise in patience at best and futility at worst. The more realistic approach is to target terms where the competition is achievable with your current authority, build that authority progressively through content and links, and then expand into more competitive territory as your position strengthens.

When it comes to anchor text in backlinks, there is a nuance worth understanding. Using the same keyword anchor text across all your backlinks is a pattern that can look unnatural and trigger algorithmic scrutiny. A natural backlink profile has variety: branded anchors, partial match anchors, naked URLs, and generic phrases alongside exact-match keyword anchors. This is relevant to how you approach outreach and link acquisition as part of your broader SEO programme.

If you are working with external support on link acquisition, understanding what SEO outreach services actually involve, how they work, and what good looks like is important before you engage. Link building done poorly can harm rankings rather than improve them. The connection to keyword strategy is that the pages you are building links to should be the pages targeting your most competitive priority terms.

Choosing Tools Without Getting Distracted by Them

There are more keyword research tools available today than any team needs, and the marketing around them encourages a degree of tool obsession that is not particularly productive. The tools are a means to an end. The end is a prioritised, intent-mapped keyword list that informs a content and optimisation strategy. Any tool that produces that output is good enough.

The major paid platforms, Semrush, Ahrefs, and Moz, all do the core job well. They differ in their data sources, their additional features, and their interface, but for most keyword research tasks the outputs are broadly comparable. If you are choosing between them, try the free trials and pick the one your team will actually use consistently. The best tool is the one that gets used.

For businesses with limited budgets, free keyword research tools can cover a significant portion of the research process. Google Search Console, Google Keyword Planner, and Google Trends are all free and all genuinely useful. They have limitations relative to paid platforms, particularly around competitor analysis and comprehensive volume data, but they are more than adequate for building an initial keyword strategy for a small or medium-sized business.

The discipline I try to bring to tool selection, whether for my own work or when advising clients, is to define the question before choosing the tool. What do I need to know? Which tool answers that question most efficiently? That framing keeps tool selection practical rather than aspirational. You do not need a platform that does 47 things if you are only going to use three of them.

What Good Keyword Research Looks Like in Practice

Good keyword research produces a prioritised, actionable list of terms grouped by topic cluster, tagged by intent, assessed for realistic competitive achievability, and connected to a content plan. It is specific enough to brief a writer and strategic enough to guide a six-month content programme.

Bad keyword research produces a large spreadsheet sorted by volume, with no intent classification, no competitive assessment, no grouping logic, and no connection to what the business actually sells or who it sells to. I have seen this delivered by agencies charging significant monthly retainers. The spreadsheet looks impressive. The business value is close to zero.

The difference between the two is not the tool used or the number of terms researched. It is the quality of the thinking applied to the data. Keyword research is an analytical exercise, and like all analytical exercises, the value is in the interpretation and the decisions it drives, not in the volume of data collected.

If you are evaluating external support for your SEO programme, including keyword research as part of a broader mandate, the comparison of leading SEO agencies is a useful reference for understanding what to expect from different types of providers and how to assess the quality of their strategic approach before you commit.

One thing I always check when reviewing keyword strategies produced by others is whether the person who built it has actually read the search results for the priority terms. It takes thirty minutes to open the top ten results for your twenty highest-priority keywords and assess what is actually ranking. That manual review almost always reveals something the tool data missed: a format preference, a content gap, an intent nuance, or a competitive dynamic that changes the priority order. The tools are the starting point. The human judgement is what makes the strategy work.

Keyword research is not the most glamorous part of SEO, and it rarely gets the attention that content production or technical audits do. But it is the foundation. Get it wrong and you spend months producing content that ranks for terms no one searches, attracts visitors who were never going to buy, and generates traffic numbers that look fine in a report but deliver nothing commercially. Get it right and every subsequent investment in content, links, and optimisation compounds on a solid base.

The full SEO Strategy Hub at The Marketing Juice covers how keyword research connects to the broader strategic picture, from technical SEO and content architecture through to measuring organic performance and understanding where SEO fits within a full acquisition mix. If you are building or rebuilding an SEO programme, the hub gives you the complete framework rather than isolated tactics.

About the Author

Keith Lacy is a marketing strategist and former agency CEO with 20+ years of experience across agency leadership, performance marketing, and commercial strategy. He writes The Marketing Juice to cut through the noise and share what actually works.

Frequently Asked Questions