SEO Keywords: What They Are and Why Most Marketers Use Them Wrong

An SEO keyword is a word or phrase that people type into a search engine when looking for information, products, or services. When you optimise a page around a keyword, you are signalling to Google that your content is the most relevant answer to that query. The keyword is not a magic code you embed in a page, it is a proxy for what your audience actually wants to find.

That distinction matters more than most SEO guides admit. Keywords are demand signals, not just ranking targets. Understanding the difference changes how you build a content strategy and, more importantly, whether that strategy produces commercial results.

What Does an SEO Keyword Actually Mean?

Strip away the tooling and the jargon and an SEO keyword is simply a record of what someone typed into a search box. Google aggregates those queries, identifies patterns, and uses them to understand what kinds of content people want. When you do keyword research, you are reading that aggregate data and deciding which demand signals your site should respond to.

Early in my agency career I watched clients treat keywords like lottery tickets. Find a high-volume term, stuff it into a page, wait for traffic. When I was building out the SEO practice at iProspect, one of the first things we had to undo was that mindset. We were managing significant ad spend across dozens of sectors, and the pattern was consistent: the accounts that chased volume without understanding intent were the ones with traffic graphs that looked impressive and revenue graphs that did not move.

Keywords have meaning in two directions. They tell you what users want. They also tell you what Google expects a page to deliver. Both matter. A keyword is a contract between the searcher, the search engine, and your content. If your content does not honour that contract, rankings will be temporary even if you earn them.

The Four Types of Keywords and Why the Distinction Is Commercially Critical

Most SEO frameworks categorise keywords by intent, and that categorisation is genuinely useful, not just academic. The four types are informational, navigational, commercial investigation, and transactional. Each one represents a different moment in a buyer’s thinking.

Informational keywords are queries where someone wants to understand something. “What is content marketing” or “how does PPC work” are informational. The user is not ready to buy. They are building knowledge. Content targeting these keywords should educate, not sell. When I judged the Effie Awards, some of the most effective long-game campaigns I saw were built on informational content that introduced a brand at the awareness stage and then converted that audience over months, not days.

Navigational keywords are searches where the user already knows where they want to go. “Semrush login” or “Hotjar pricing” are navigational. These are largely irrelevant to your keyword strategy unless you are the brand being searched for, or you are trying to intercept a competitor’s branded traffic, which is a different conversation entirely.

Commercial investigation keywords sit in the middle of the buying cycle. “Best CRM for small business” or “Ahrefs vs Semrush” are commercial investigation queries. The user is comparing options. They are close to a decision but not there yet. This is where comparison content, detailed reviews, and category pages earn their keep. Semrush’s keyword selection guide covers this territory well if you want a practical walkthrough of identifying these mid-funnel terms.

Transactional keywords signal purchase intent. “Buy running shoes online” or “book a marketing consultant” are transactional. Users searching these terms want to act. Your content needs to make that action as frictionless as possible. These are the keywords that drive direct revenue, and they are often the most competitive because everyone knows they are valuable.

The mistake I see repeatedly, including in accounts I have inherited after agency transitions, is treating all keywords as if they sit in the same moment. A client once pushed us to optimise their product pages for informational queries because the volume looked attractive. The pages ranked. Bounce rates were high, conversion rates were low, and the client was confused. The content was fine. The keyword match was wrong.

If you want to build a keyword strategy that connects to business outcomes rather than just ranking metrics, the broader framework matters. This article sits within a complete SEO strategy that covers everything from positioning to technical foundations to competitive analysis. Understanding keywords in isolation is useful. Understanding how they fit into a full search strategy is what actually moves the needle.

Short-Tail vs Long-Tail: Where the Volume Myth Breaks Down

Short-tail keywords, sometimes called head terms, are broad, high-volume queries. “Marketing”, “SEO”, “shoes”. Long-tail keywords are more specific, lower-volume phrases. “B2B SaaS content marketing strategy” or “running shoes for flat feet under £80”.

The conventional wisdom is that head terms are the prize and long-tail terms are the consolation. That is backwards in most commercial contexts.

Head terms are competitive to the point of being inaccessible for most sites. The domains ranking for “marketing” or “SEO” have decades of authority and thousands of linking domains. Chasing those terms as a growing business is not ambition, it is misallocated effort. Long-tail terms, by contrast, are often where intent is clearest and competition is thinnest. A user searching “B2B SaaS content marketing strategy for Series A startups” knows exactly what they want. If your content answers that question precisely, you can rank and convert at a rate that a head-term ranking might never match.

When I was growing the agency from around 20 people to over 100, our own SEO strategy was almost entirely long-tail. We could not compete with the established players on broad terms. What we could do was produce specific, expert content that answered the exact questions our target clients were asking. That content built credibility and generated inbound enquiries from the kinds of clients we actually wanted to work with. Volume was modest. Quality was high. That trade-off was deliberate.

The long-tail also aggregates. A single long-tail keyword might bring 50 visits a month. Two hundred long-tail keywords bring 10,000 visits a month, often with better conversion rates than a single head term generating the same volume. The maths is straightforward once you stop treating search volume as the primary metric.

How Search Volume Data Should (and Should Not) Influence Decisions

Search volume is a useful directional indicator. It tells you roughly how often a query is searched over a given period. It does not tell you how much of that traffic you will capture, what those users will do when they land on your page, or whether the traffic is commercially valuable.

I have a consistent scepticism about how marketers treat data, and search volume is a good example of the pattern. The number exists, it is concrete, and so it gets treated as meaningful in isolation. A keyword with 40,000 monthly searches sounds better than one with 400. But if the 40,000-search keyword is dominated by a handful of authoritative domains and the 400-search keyword has clear commercial intent and weak competition, the 400-search keyword is the better investment.

Volume data also has accuracy limitations that tools do not always make obvious. Keyword research platforms pull from various data sources and apply modelling to fill gaps. The numbers are estimates, not census data. For high-volume terms the estimates are reasonably reliable. For niche or emerging queries, the numbers can be significantly off. I have seen keyword tools show zero volume for phrases that were clearly generating traffic for clients, simply because the query was too new or too specific for the tool’s data to capture it.

Use volume to prioritise and to sense-check. Do not use it as the sole basis for a content decision. A keyword with no volume but clear relevance to your audience is still worth a page if you have the domain authority to rank and the content to do the topic justice.

Keyword Difficulty and Why It Is Not a Reason to Avoid a Topic

Keyword difficulty (KD) scores are another metric that gets misread regularly. Most tools assign a difficulty score based on the domain authority or page authority of the pages currently ranking for a keyword. High difficulty means strong incumbents. Low difficulty means weaker competition.

The problem is that difficulty is a snapshot, not a verdict. A high-difficulty keyword is hard to rank for today. It may be more accessible in twelve months if your domain authority has grown, if you build the right links, or if the existing content ages and weakens. Ruling out a keyword because its difficulty score is high is a short-term decision that can cost you long-term positioning.

The more useful question is whether you can produce content that is meaningfully better than what currently ranks. I have seen pages from relatively young domains outrank established players on competitive terms because the content was more thorough, more accurate, and more genuinely useful. Google’s ranking systems are not purely about domain authority. Content quality, user engagement signals, and topical relevance all play a role. Difficulty scores do not capture those variables.

That said, difficulty is a reasonable input to resource allocation. If you have limited content budget, spending it on keywords where you have a realistic chance of ranking in the near term makes more sense than targeting terms where you will be invisible for years. It is a planning tool, not a binary gate.

Semantic Keywords and the Shift Away from Exact Match Thinking

Google’s ability to understand language has changed significantly over the past decade. The shift toward semantic search means that Google no longer requires an exact keyword match to understand that a page is relevant to a query. A page about “running shoe fit” can rank for “how should running shoes fit”, “running shoe sizing guide”, and “do running shoes run small” without using each of those phrases verbatim.

This has practical implications for how you write. Obsessing over keyword density, the old practice of hitting a target percentage of keyword repetitions per page, is not just outdated, it produces worse content. Pages written to hit a keyword frequency target read like they were written to hit a keyword frequency target. Users notice. Engagement suffers. Rankings follow.

The better approach is to write comprehensively about a topic. Cover the subject in enough depth that the semantically related terms appear naturally. Use variations, synonyms, and related concepts because they make the content clearer and more complete, not because you are trying to capture additional keyword variants. Moz’s thinking on AI and content for SEO touches on how semantic coverage is becoming increasingly important as search engines grow more sophisticated at understanding topical depth.

LSI keywords, or Latent Semantic Indexing keywords, get mentioned frequently in SEO content. The term is technically outdated as a description of how Google works, but the underlying idea is sound: pages that cover a topic thoroughly and use the full vocabulary of that topic tend to perform better than pages that repeat a single phrase. Think about what a subject matter expert would write, not what a keyword tool would generate.

Where Keywords Come From: Research Methods That Go Beyond the Tools

Keyword research tools are the obvious starting point. Semrush, Ahrefs, Google Keyword Planner, and similar platforms give you volume estimates, difficulty scores, and related query suggestions. They are useful. They are also incomplete.

The best keyword intelligence I have gathered over the years has come from sources that tools cannot access. Sales call recordings. Customer service transcripts. The exact language prospects use when they describe their problem. The questions that come up repeatedly in sales conversations are almost always better keyword candidates than anything a tool surfaces, because they reflect real demand in real language.

One of the most effective things I did at iProspect was get the SEO team into client sales conversations. Not to sell, but to listen. The vocabulary gap between how clients described their problems and how our content described solutions was significant. Closing that gap produced rankings that actually mattered commercially because the content matched what buyers were searching for, not what we assumed they were searching for.

Other sources worth mining: Reddit threads in relevant subreddits, Amazon reviews for products in your category, Quora questions, and the “People Also Ask” and “Related Searches” sections in Google’s own results. These surfaces show you real language from real users. They are not mediated by a tool’s data model. Search Engine Journal has covered how Google’s crawling and indexing works, which gives useful context for understanding how keyword signals flow through the system.

User feedback tools can also surface language patterns. Hotjar’s UX analytics and similar platforms let you see how users interact with your site, which pages they leave quickly, and where they spend time. That behaviour data does not tell you keywords directly, but it tells you whether your current content is answering the queries that brought people to your site, which is the other half of the keyword equation.

Keyword Mapping: Connecting Keywords to Pages Without Creating Chaos

Keyword mapping is the process of assigning specific keywords to specific pages on your site. Done well, it prevents keyword cannibalism (where multiple pages compete for the same query), ensures every page has a clear primary purpose, and makes your site architecture more logical for both users and search engines.

The basic rule is one primary keyword per page, with supporting secondary keywords that are semantically related and share the same intent. A page targeting “content marketing strategy” might also cover “content strategy for B2B”, “how to build a content plan”, and “content marketing framework” as secondary terms, because a user searching any of those phrases wants essentially the same thing.

Where mapping breaks down is when sites grow without a plan. I have audited sites with thirty pages all partially optimised for variations of the same keyword, none of them ranking well because Google cannot determine which one is the authoritative answer. Consolidating those pages into one comprehensive resource, and redirecting the others, typically produces better results than trying to rank all thirty separately.

Keyword mapping also informs internal linking. When you know which page owns which keyword, you can link to it deliberately from related content, reinforcing its relevance signal to Google. That internal link structure is one of the most underused levers in SEO, partly because it requires you to have a clear map in the first place.

The Relationship Between Keywords and Content Quality

Keywords tell you what to write about. They do not tell you how to write it well. This distinction gets lost in a lot of SEO guidance that treats keyword optimisation as the primary content task.

Google has become progressively better at assessing whether content genuinely serves the user’s query or merely appears to. Thin content that hits keyword targets but provides little actual value has been penalised repeatedly through algorithm updates. The pattern is consistent: short-term keyword manipulation produces short-term rankings. Content that genuinely answers questions and demonstrates expertise holds rankings over time.

When I was building content programmes for enterprise clients, the brief was never “write 1,200 words targeting this keyword”. The brief was “answer this question better than anyone else on the internet”. That framing produces different content. It requires actual knowledge of the subject, not just an ability to arrange keywords into sentences. The keyword is the entry point. The content quality is what determines whether the ranking lasts.

This is particularly relevant for competitive queries where the existing content is already strong. If the pages ranking for your target keyword are genuinely good, you do not beat them by producing more of the same. You beat them by finding the angle they missed, the question they did not answer, or the format that serves the user better. That requires thinking, not just keyword research.

Keywords are one input into a search strategy, not the whole of it. If you are working through how all the pieces connect, from keyword research to on-page optimisation to link building to tracking, the complete SEO strategy hub on this site covers the full picture in a way that keeps commercial outcomes at the centre.

Common Keyword Mistakes That Cost Rankings and Revenue

Targeting keywords with no commercial relevance to your business is more common than it should be. High volume is seductive. A keyword that sends 20,000 visits a month to a page that has nothing to do with what you sell is not an asset, it is a distraction that inflates traffic numbers while contributing nothing to revenue. I have seen this in performance reviews where clients were proud of traffic growth that was entirely disconnected from any commercial outcome. The traffic was real. The value was not.

Ignoring local intent is another frequent error for businesses with a geographic component. A keyword like “accountant” behaves very differently from “accountant in Manchester”. Google understands local intent and serves local results for queries where geography is implied even if not stated. Moz’s local SEO guidance covers how to approach this for businesses where location is a material factor in the buying decision.

Failing to update keyword strategy as the market changes is a structural problem in many content programmes. Keywords that were high-value three years ago may be saturated, irrelevant, or superseded by new terminology. A keyword audit is not a one-time exercise. Markets evolve, language changes, and new query patterns emerge. The businesses that treat keyword strategy as a living document rather than a completed project tend to maintain rankings more consistently.

Finally, treating every keyword as a separate content opportunity rather than grouping related keywords into coherent topics leads to fragmented, thin content. A site with four hundred pages each targeting a slightly different keyword variant, none of them comprehensive, will almost always underperform a site with one hundred well-developed pages that each cover a topic thoroughly. Depth beats breadth in most content strategies.

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 works.

Frequently Asked Questions