Keyword research sits on the line between strategy and execution. Most people ask which keywords they should rank for. The better question is where your Ranking Opportunities actually sit, and what is stopping you owning them.
This page runs keyword research through Opportunity and Authority. The work is not building the biggest spreadsheet anybody has ever seen. The work is finding the place where search demand and your ability to be believed overlap.
I found that place by accident, in a printing business. We printed whatever came through the door, there were a lot of printers in Auckland, most of them with fancier machines than ours, so really we could only compete on service or location. The average job was something like a hundred and thirty dollars. Then a book came through, a perfect bound paperback with a glued spine and a cardboard cover, and we had to find a contract finisher to bind it because we had no idea how. That job was closer to nine hundred dollars. So I built a website called bookprint.co.nz, set up some Google Ads, and in its first year that site generated a hundred thousand dollars of printing work, all of it new customers. Getting new customers into the generic printing business had been genuinely hard. Getting them for books was much easier, and not one thing about our machines had changed.
Keyword research is that decision made on purpose instead of by accident. Not "printing". "Book printing".
The Keyword Illusion
Most keyword research goes the same way. Somebody opens a tool, types in a seed term, and stares at thousands of variations with a volume and a difficulty score beside each one. A spreadsheet gets built. Priorities get set by traffic potential. A content calendar gets planned around the high-volume, low-difficulty finds.
Then nothing happens.
Rankings do not move, traffic stays flat, and nothing converts, so the tool gets blamed, or the strategy does, or the market does. Almost nobody goes back and asks whether the keyword research was ever a strategy in the first place.
It was not. It is reconnaissance. Keywords are signals showing where people are searching, and knowing where people search only helps once you also know why they are searching there and whether you are genuinely positioned to answer them.
I am going to invent two keywords now, because inventing them openly is more honest than quoting numbers I have not measured. Say the first carries fifty thousand searches a month, and the top ten results are all household names sitting on domains that are older than your business. You are not taking that this year. Say the second carries five hundred searches a month, and the top ten are thin posts, guides that were current three years ago, and a competitor who has clearly stopped caring. That one is an Opportunity.
Keyword research does not create Ranking Opportunities, it finds the ones already sitting in the market, and your job is recognising them and then choosing which of them are worth a year of your life.
Topics Over Keywords
The shift from keywords to topics changes what you are actually building.
A keyword is a phrase. "Best CRM for startups". "How to optimise conversion rates". "What is semantic SEO". Keywords are discrete: they either get searched or they do not, and you either rank or you do not.
A topic is a concept, searched dozens of different ways by different people at different stages. Customer relationship management arrives as "CRM for small business", "how to choose a CRM", "CRM implementation", "CRM training", "open source CRM alternatives", and hundreds more. One topic holds all of it.
Build around keywords and you optimise one search at a time. One article for "best CRM for startups", another for "CRM for nonprofits", another for the implementation checklist. Plenty of publishing, nothing accumulating, every piece standing on its own.
Build around topics and you pick the handful of areas where your Authority genuinely lives. Five or ten of them, not fifty. Take a software company selling to finance teams, which I am inventing: financial reporting, cash flow forecasting, budgeting, accounts payable, data integration, audit compliance, the financial close. Those are not keywords. That is the ground the expertise stands on.
Inside each topic you build a pillar and its clusters. The pillar covers the ground at a high level and answers the whole question. The clusters take the specific angles, the use cases, the mistakes and the comparisons, and they link back to the pillar and sideways to each other.
That does two jobs at once. For readers, somebody lands on one thing, finds the connected thing, and works out over three clicks how deep you go, so trust compounds because the evidence compounds. For Google, sites get read across clusters rather than page by page, so interlinked coverage of one topic tends to pull in variations and questions you never wrote a page for.
Entity-Based Thinking
Google moved away from pure keyword matching years ago, and the shift towards entities is the most underrated change in this field. I am describing a pattern people in the industry observe, not a mechanism Google has documented.
An entity is a thing that exists whether or not anybody searches for it. Apple is an entity. The iPhone is an entity. Steve Jobs is an entity. Marketing attribution is an entity. The colour blue is an entity. Entities have properties, and they have relationships to each other.
A search for "iPhone battery life" does not behave like a hunt for pages carrying those three words in that order. It behaves like a hunt for pages that understand the phone, the battery, power draw, low power mode, and what two years of charging does to the thing.
That reframes the question you take into the tool. Instead of asking which keywords to rank for, ask which entities your Authority actually covers, and how you would prove you understand them.
Start by listing the core ones. A marketing analytics company might list data collection, attribution, the customer journey, conversion tracking, cross-channel reporting and user cohorts, with adjacent entities like the marketing funnel and customer acquisition cost sitting one step out.
Then map how they connect. How the customer journey relates to attribution. How conversion tracking feeds cross-channel reporting. Write the relationships rather than the definitions, so instead of explaining what attribution is, explain how it changes what you spend next month.
The pay-off is that content built this way tends to surface for variations nobody targeted, and it builds Authority faster, because conceptual coverage is much harder to fake than word coverage.
Search Intent: The Real Framework
Volume and difficulty matter. Intent matters more.
Intent is the reason behind the search, and it falls into four rough groups.
Informational: somebody wants to learn. "How does semantic SEO work". "What is the difference between authority and trust". They want the explanation, and a guide or a framework satisfies them.
Navigational: somebody wants to reach a specific place. Your brand name, or a resource they already know exists. They want directions.
Commercial: somebody is researching a purchase. "Best project management software for teams". "Asana vs Monday.com". They want a comparison, and they are not ready to be sold to.
Transactional: somebody has already decided. "Buy iPhone 15 Pro". "Sign up for HubSpot". They want the shortest path to done.
The expensive mistake is the mismatch. You write a thorough guide called "how to choose project management software" and target the keyword "project management software", which is a term people type when they want to sign up for something. Your guide does not let them do that. Somebody's product page does, and that is who ranks.
Checking is free and takes two minutes. Search the term, look at the top ten, and notice whether they are guides or product pages, reviews or tutorials, sales pages or explainers. That is the intent Google has already settled on, and arguing with it from page two is a slow way to spend a year.
Competitive Gap Analysis
The best Ranking Opportunities are not the highest-volume keywords, they are the places where the answers already published are weak.
A big term with a hundred credible competitors is a fight against people who have been building for a decade. A smaller term where the top ten are thin, dated and clearly unloved is a fight you can win this quarter. Both of those are invented examples, and the real version comes from opening your own market and looking.
Gap analysis means reading what your competitors published and noticing what they did badly. A few patterns keep turning up.
Outdated content. The ranking article is three years old, the field moved, the tools changed, nobody went back to it, and so the current version is yours to write.
Shallow coverage. Somebody published a beginner's guide where searchers wanted the intermediate answer, or the reverse. The intent layer got missed, and the gap is sitting there in plain sight.
Missing angles. Everybody wrote the same guide from the same seat. Nobody wrote it for remote teams, or for the business where one person does three jobs. Write that one.
Weak supporting content. A competitor holds a strong pillar with almost nothing underneath it, so build the cluster properly and you can outrank a better page with a better structure.
This changes what selection means. You are not chasing volume, you are hunting for demand with weak supply, which is the cheapest ground to take while your Authority is still growing.
Keyword Difficulty in Context
Every research tool scores keywords for difficulty, each under its own name and its own formula, and the scores are genuinely useful right up to the point where somebody treats them as gospel.
Treat a high score as a warning rather than a verdict. A brand-new site aiming at the hardest terms in its market spends a year learning something it could have learned in an afternoon. A site with real topical depth can take terms that look hard on the tool, because the score is reading the page while the Authority is sitting on the domain.
Three things the score cannot see.
Where you stand right now. New site: take the easy ground. A year of publishing behind you: aim higher. A name people already recognise in the field: aim wherever you like.
What the keyword is worth. A hard keyword that converts into paying customers beats an easy one that collects people who will never buy, and no difficulty score has any idea what your margin is. You do.
Who is actually sitting there. A middling score can hide ten weak pages that the formula rated moderately because nobody had linked to them yet. A gentle-looking score can hide ten industry leaders. Open the results and look at them.
When I built bookprint.co.nz I never looked at a difficulty score. I picked the ground where a small printer had an outsized position, which is the same decision a score is meant to help you make, only slower and with more arithmetic.
Use the scores to filter out the obviously impossible, then stop using them. Do not let a number talk you out of a target that matters, and do not let a low number convince you that an easy-looking keyword is easy.
Building the Keyword Roadmap
Research only matters when it turns into a sequence, and that sequence is the roadmap.
Start with the core topics, the handful of areas where your Authority genuinely lives. A financial software company might take cash flow management, forecasting, expense automation and reporting. An agency might take content strategy, paid media, analytics and positioning. Both of those are made up, and yours will not be.
From each topic, name the pillar. "Guide to cash flow forecasting." "Content strategy framework." Those are the hubs everything else hangs off.
Then the clusters. Narrower pieces on specific aspects, use cases and problems. "Forecasting cash flow for seasonal businesses." "Content strategy for B2B software." Each one links back to the pillar and sideways to its siblings.
Then the quick wins. Lower-difficulty, lower-volume terms you can publish against fast, which buy you presence and start the Authority clock running.
Then the long-term targets. The high-volume, hard terms that matter commercially and will take sustained work to take, and writing them down as three-year goals stops you quietly pretending they are this quarter's goals.
That is the content strategy. What to write, in what order, how to structure it, where to link it. Not a pile of keywords, but a plan for building topical authority in a sequence that compounds.
From Opportunity to Authority
Finding a Ranking Opportunity is the smaller half of this job. Converting it into Authority is the rest, and the rest takes considerably longer than anybody plans for.
Keywords show you where people are searching. They cannot tell you whether you deserve to be found there. Ranking needs two things no keyword tool has any view on: content that is genuinely better, and evidence that you have actually done the thing you are writing about.
Better means more complete, more current, more useful, and written from a position your competitors are not standing in. Better research. Better examples. The parts everybody else left out because they did not know them.
Evidence means showing the work. Internal links that make the relationships visible. Citations. Real examples with real detail in them. Clusters instead of one-off posts. Links earned from people who had an actual reason to link.
The way I became a book specialist was not by inventing a new kind of book. I did not differentiate a book. Saying no to everything that was not a book is what made me the specialist, and that is the Authority half of this equation doing its work.
Keywords are the entry point and nothing more. The real work starts the moment you know which Opportunities are worth having.
Learning From Expert Research
For deeper work in opportunity and Authority-based keyword research, three practitioners have published frameworks worth the time.
Koray Tuğberk GÜBÜR works on semantic SEO and topical authority, and his frameworks for entity-based content and topical clusters run well past traditional keyword research. His work is a bridge between finding keywords and building Authority.
Mark Williams-Cook works on user questions and intent at scale. His work with Also Asked data and question clustering shows the whole landscape of how people ask about a topic, which is a larger picture than any keyword list gives you.
Kyle Roof runs tests on what moves rankings. His testing on topical authority, content depth and link relevance keeps challenging assumptions, and it is a useful antidote to advice nobody has ever checked against anything.
Read their work, then apply it, and keyword research stops being about volume and difficulty scores and starts being about which opportunities you can genuinely take.
Connecting to the Framework
Keyword research feeds the wider Opportunity and Authority framework. Keywords show where the demand is, Opportunity means ranking is possible if the content is better, and Authority is what keeps you there once you arrive.
It also feeds the content strategy, which decides how topical authority gets built around the opportunities you chose.
Run the existing site through the diagnostic lens before any of that. What ranks now, where you are winning, and which opportunities you have been walking past. That history is the input to the roadmap.
And the technical SEO foundation has to hold, because good content aimed at good keywords still loses to a slow site, a crawl problem, or an internal linking structure that leaves half your pages orphaned.
Marketing Curious: Working the Noise carries the long version. This page is a rendering. The seed is the source. The book is the story of building it.
Part of the Marketing Universe. Explore Traffic Plus Offer : The Trust Algorithm : 4-Quadrant AI.