Keyword research isn't about finding high-volume terms. Learn the Intent-First method to find high-value search terms that convert — not just rank.
Most guides treat keyword research as a data exercise. They walk you through filtering by volume, checking difficulty scores, and exporting a spreadsheet. That's not keyword research — that's keyword harvesting.
The difference matters enormously. Keyword harvesting produces long lists of terms that look promising in a tool but represent wildly different stages of the buyer journey, wildly different levels of commercial intent, and wildly different conversion probabilities. What most guides won't tell you is that the most dangerous outcome of keyword research isn't targeting the wrong terms — it's targeting the right terms for the wrong reasons.
A keyword with 8,000 monthly searches on 'what is project management' is not the same opportunity as a keyword with 300 monthly searches on 'best project management software for construction firms.' One fills your analytics dashboard. The other fills your pipeline. The obsession with volume also leads teams to overlook what we call 'Buyer Vocabulary' — the specific, often awkward, sometimes industry-specific language your actual customers use when they're close to a purchase decision.
That language rarely shows up as a high-volume keyword. But it converts at a rate that makes broad terms look embarrassingly inefficient.
The textbook definition of keyword research is straightforward: it's the process of identifying the words and phrases people type into search engines so you can create content that appears in those results. That definition is technically accurate and practically incomplete. A more useful definition is this: keyword research is competitive intelligence.
It's the process of mapping what your market wants, how urgently they want it, what stage of awareness they're in, and which search terms signal that they're close to taking action. When you approach keyword research this way, everything changes. You stop treating it as a content planning exercise and start treating it as a business intelligence function — one that tells you where buyer demand is concentrating, which problems your market is actively searching to solve, and where your competitors have blind spots you can exploit.
There are four dimensions to every keyword worth understanding before you decide whether to target it:
1. Volume — How many people search for this term each month. Important context, but not the primary decision factor.
2. Intent — What is the searcher trying to accomplish? Are they learning, comparing, or buying? This is the most important dimension and the most commonly ignored.
3. Specificity — How narrow is the term? Broad terms attract broad audiences. Specific terms attract specific buyers.
4. Commercial Gravity — Does this keyword sit close to a transaction? Are there paid ads running on it? Does the content that ranks for it include pricing, demos, or CTAs? Commercial gravity is the hidden signal most tools don't surface directly.
When you evaluate keywords across all four dimensions — not just volume — you start seeing the search landscape differently. You notice that the terms everyone is fighting over are often informationally rich but commercially thin. And you notice that the terms no one is targeting are sometimes sitting right at the point where intent converts to action.
Keyword research, done properly, is how you find those moments before your competitors do.
Before you open any keyword tool, write down the five questions your sales team hears most often on discovery calls. Those questions are keywords. They're often mid-to-bottom funnel, they're in buyer language, and they're almost never high competition.
Treating keyword difficulty as the second filter after volume. Difficulty tells you how hard it is to rank, not whether ranking is worth the effort. Apply an intent filter before a difficulty filter, every time.
This is the framework I wish existed when I started doing keyword research. After years of watching high-volume, low-conversion campaigns underperform, we developed a scoring method that goes beyond the standard volume-difficulty matrix. We call it the PESO Signal Framework — not to be confused with the marketing model of the same name. PESO here stands for: Paid presence, Exact-match transactional phrasing, SERP commercial features, and Outcome-specific language.
Each signal tells you something different about the commercial temperature of a keyword.
P — Paid Presence If advertisers are spending money on a keyword, there's commercial intent behind it. When you search a term and see paid ads above the organic results, that's a strong signal that businesses have tested this keyword and found it worth paying for. More ads mean stronger commercial intent. No ads often means informational intent — worth being aware of before you invest.
E — Exact-Match Transactional Phrasing Words like 'buy,' 'hire,' 'cost,' 'pricing,' 'vs,' 'best,' 'for [specific industry],' and 'near me' are linguistic markers of transactional intent. A keyword containing one or more of these signals is closer to a conversion than one without them. When evaluating a keyword list, isolate and prioritise any term containing transactional language — regardless of its volume.
S — SERP Commercial Features Open the search results for your target keyword. What do you see? If the page is dominated by product listings, comparison pages, review sites, or service providers, that's a commercially active SERP. If it's dominated by Wikipedia entries, forum threads, and definition articles, the intent is informational. The SERP doesn't lie. Look at what Google has decided searchers want when they use that term.
O — Outcome-Specific Language The most commercially valuable keywords describe a specific outcome the searcher wants to achieve. 'Keyword research tool' is generic. 'Keyword research tool for SaaS content teams' is outcome-specific. 'How to do keyword research' is informational. 'How to do keyword research to increase qualified traffic' is outcome-specific and indicates a more sophisticated, results-oriented searcher — often a buyer or decision-maker.
To apply the PESO Signal Framework, score each keyword in your list from 0 to 4 based on how many signals it carries. Keywords scoring 3 or 4 are your primary targets. Keywords scoring 1 or 2 are supporting content. Keywords scoring 0 are brand-awareness plays at best.
Run your top 20 target keywords through the PESO framework before you commit to a content calendar. You'll typically find that 30-40% of the keywords you thought were valuable drop to a score of 1 or below when you apply all four signals.
Assuming paid ads always equal high commercial intent. Some keywords attract advertisers with poor targeting. Always cross-reference paid presence with at least one other PESO signal before elevating a keyword's priority.
Here's a pattern we see repeatedly across industries: companies do keyword research by starting with their own product or service category and expanding outward. They search for 'project management software,' find related terms, filter by volume, and build their content around those results. The problem is that every competitor is doing exactly the same thing, starting from the same seed keyword, using the same tools, arriving at the same list.
The Buyer Vocabulary Gap method starts from the opposite end. Instead of beginning with what you sell, you begin with what your customers say — before they know your product exists.
Here's how to execute it:
Step 1: Mine First-Party Buyer Language Go to your sales call recordings, customer support tickets, onboarding surveys, and product reviews. Look for the exact phrases customers use to describe their problem before they found you. Not what they call your solution — what they called their pain. This language is typically simpler, more emotional, and far more specific than any category term. 'We kept missing deadlines because no one knew who owned what' is a buyer vocabulary phrase. 'Project management inefficiency' is a marketer's phrase. One converts. One doesn't.
Step 2: Check Those Phrases Against Keyword Tools Take the raw phrases from your first-party data and run them through a keyword research tool. Many won't have measurable search volume. Some will. The ones that show even modest volume (50-500 searches per month) with low competition are your gap keywords — terms that describe real buyer problems in real buyer language that no one is targeting.
Step 3: Build 'Problem-First' Content Create content that matches the vocabulary of the problem, not the vocabulary of the solution. A page titled 'Why Your Team Keeps Missing Deadlines (And How to Fix Accountability Gaps)' will often outperform 'Best Project Management Software' for conversion — even if the latter ranks for more volume — because it meets buyers at the exact moment they're articulating their pain.
Step 4: Map to the Buyer Journey Organise your gap keywords by buyer awareness stage. Early-stage buyers describe symptoms ('team keeps missing deadlines'). Mid-stage buyers describe solutions ('task assignment software'). Late-stage buyers compare options ('Asana vs Monday for remote teams'). A complete keyword strategy covers all three stages with vocabulary-matched content.
The Buyer Vocabulary Gap method consistently surfaces keywords with low competition, high specificity, and strong commercial intent — the exact combination that keyword tools won't hand you automatically.
Ask your customer success team for the top five phrases customers use when they first describe why they signed up. Those phrases are already proven buyer language — they came from people who converted. Build keyword strategy around them.
Editing buyer vocabulary to sound more professional or SEO-friendly. If customers say 'can't get my team to stay organised,' don't translate that to 'team organisation software challenges.' Use the raw language. It ranks for how buyers actually search.
Search intent is the single most important concept in keyword research, and it's the one most consistently misapplied. Intent describes what the searcher is trying to accomplish — and understanding it correctly means the difference between creating content that ranks and creates revenue, and content that ranks and creates nothing.
There are four recognised intent categories, but the nuance within each category is where the real strategy lives.
Informational Intent The searcher wants to learn something. Keywords like 'what is keyword research,' 'how does SEO work,' or 'types of backlinks' fall here. These are valuable for building topical authority and capturing early-stage awareness, but they convert at low rates unless there's a strong next-step architecture embedded in the content — think lead magnets, email captures, or diagnostic tools.
Navigational Intent The searcher is looking for a specific website or brand. Unless you're targeting branded terms, navigational keywords aren't your primary opportunity.
Commercial Investigation Intent This is where things get interesting. Searchers using terms like 'best keyword research tools,' 'keyword research tool comparison,' or 'keyword research tool for small business' are actively evaluating options. They're not buying yet, but they're close. Commercial investigation keywords are high-value targets because they reach buyers during the decision-making process.
Transactional Intent The searcher is ready to act. 'Keyword research service,' 'hire SEO consultant,' 'buy keyword research tool' are transactional. These keywords have the highest conversion potential and, typically, the strongest paid competition.
What Most Guides Won't Tell You About Intent Intent is not fixed. The same keyword can shift intent based on context clues in the full query. 'Keyword research' alone is ambiguous. 'Keyword research for B2B SaaS' signals commercial investigation. 'Keyword research pricing' signals transactional. Training yourself to read these context signals within keyword phrases — rather than applying broad categorical labels — dramatically improves the quality of your targeting decisions.
Also worth noting: Google often knows intent better than keyword tools. The SERP result is Google's interpretation of intent. If you target a keyword expecting commercial investigation traffic but Google serves informational content, you're fighting the algorithm's intent classification — and you'll lose.
For every target keyword, open an incognito browser, search the term, and screenshot the first page. Note whether the results are definitions, comparison pages, product pages, or forum threads. That pattern tells you exactly what type of content Google believes searchers want.
Targeting transactional keywords with informational content, or informational keywords with sales pages. Mismatched intent and content type is one of the most common reasons well-researched content fails to rank despite strong optimisation.
The keyword research tool landscape is crowded, and the differences between major platforms are smaller than their marketing suggests. What matters more than which tool you use is what process you follow when you use it. Here's the process we've refined over years of working with founders and operators across competitive markets.
Phase 1: Seed Keyword Generation (Before You Open Any Tool) Write down every way a potential buyer might describe their problem — without using your product category. Then write down your product category and every logical synonym. Then list your competitors. Then list the questions your sales team hears. This pre-tool exercise typically surfaces 30-50 seed keywords that become your research starting points. Skipping this phase means your research is only as creative as the tool's suggestion algorithm.
Phase 2: Expansion and Data Collection Run your seed keywords through your preferred research tool. Collect: search volume, keyword difficulty, related terms, and SERP features. At this stage, cast a wide net. Export broadly and filter later. Look specifically for: question-based variations ('how to,' 'what is,' 'why does'), comparison variations ('X vs Y,' 'alternatives to'), and specificity variations (adding industry, role, company size, or use case to broad terms).
Phase 3: PESO Signal Scoring Apply the PESO Signal Framework to your expanded list. Score each keyword 0-4. This reduces your list significantly and focuses your effort on commercially valuable targets.
Phase 4: Buyer Vocabulary Gap Check Cross-reference your filtered list against your first-party buyer language data. Are the terms your customers actually use represented? If not, add them manually even if the tool shows low or no volume. Zero-volume keywords that match exact buyer language often drive qualified traffic because the content matches long-tail voice and conversational search patterns.
Phase 5: Cluster and Prioritise Group remaining keywords into topic clusters — collections of related terms that a single piece of content (or a small content hub) can address together. Prioritise clusters based on: commercial intent score, estimated difficulty relative to your current domain authority, and strategic fit with your business stage.
This five-phase process takes longer than the standard 'search, filter, export' approach. It produces dramatically different results.
Keep a 'living keyword document' rather than a one-time spreadsheet. Add new terms every quarter as market language evolves, competitor positioning shifts, and your product expands. Keyword research is a continuous intelligence function, not a project with a completion date.
Over-relying on a single keyword tool and treating its data as definitive. Every tool has gaps, biases, and update lags. Cross-reference two tools and always validate with manual SERP checks before committing to a keyword as a primary target.
Long-tail keywords are search phrases that are longer, more specific, and lower in individual search volume than head terms. They get less attention from SEO professionals because the numbers look smaller. That's exactly why they're the most undervalued asset in any keyword strategy.
Here's the dynamic that most guides describe without fully explaining: the majority of all searches ever performed are searches that have never been performed before, or searches performed only a handful of times per month. The search landscape is not a pyramid where a few big terms dominate everything. It's an enormous, expanding landscape where specificity is the rule, not the exception.
Why Long-Tail Keywords Convert Better A searcher typing 'keyword research' is at the beginning of an exploration. A searcher typing 'keyword research process for B2B SaaS content teams' has already done their exploration. They know what they need. They're searching for a specific answer to a specific problem. Content that answers that specific search converts at a fundamentally higher rate because it meets a precisely defined need.
The Compound Growth Effect A single long-tail keyword page might drive 30-80 visits per month. That seems small until you have 40 of them. A network of specific, intent-matched long-tail content compounds into a significant, diversified traffic asset — one that's also far more resilient to algorithm updates than broad-term rankings, because the value comes from relevance depth rather than keyword competition.
How to Find Long-Tail Opportunities Systematically Use the 'specificity ladder' approach: take any head term and add one specificity dimension at a time. Start with the broad term. Add an industry. Add a company size. Add a role. Add an outcome. Each step down the specificity ladder is a potential long-tail target.
Example: 'keyword research' → 'keyword research for SaaS' → 'keyword research for early-stage SaaS' → 'keyword research process for early-stage B2B SaaS with limited content resources'
Each variation targets a smaller, more specific audience — and that audience is more likely to be your actual buyer.
What Most Guides Won't Tell You Long-tail keywords also protect you from the volatility of broad-term rankings. If a single high-volume keyword drops from position 2 to position 8, your traffic can fall dramatically. A diversified long-tail portfolio means no single ranking drives your organic revenue. That resilience is worth more than most SEOs acknowledge.
When building long-tail content, prioritise specificity dimensions that your buyers self-identify with: their industry, their company stage, their role, or their specific use case. Searchers who see their exact context reflected in a headline are far more likely to engage with the content.
Dismissing long-tail keywords because a tool shows zero or very low search volume. Low volume in a tool often means the term is searched in varied phrasings that add up to meaningful aggregate demand, or that it's a voice/conversational search pattern that tools undercount.
Modern SEO is increasingly a competition for topical authority — Google's assessment of whether your site has genuine expertise in a subject area, not just individual pages optimised for individual terms. Keyword research is the mechanism through which you build topical authority systematically rather than accidentally.
The connection works like this: when you map all the questions, comparisons, definitions, tutorials, and decision-support content related to your core topic area — and then create high-quality content that addresses that full map — Google begins to recognise your site as the authoritative source for that subject. Individual pages may rank for individual keywords, but the cumulative signal of comprehensive topic coverage is what drives compounding authority over time.
The Topic Cluster Model Effective keyword research identifies a 'pillar' topic (a broad, high-level concept) and a surrounding set of 'cluster' topics (specific subtopics, questions, and variations). The pillar page addresses the topic broadly and links to cluster pages. Cluster pages address specific aspects in depth and link back to the pillar. This internal linking architecture signals to search engines that your site has both breadth and depth on a topic.
Keyword research is what populates this model with real search demand data — ensuring that the cluster topics you create content for are topics people are actually searching, not topics you assume are relevant.
Keyword Research as an Authority Audit Tool You can also use keyword research defensively: identify all the questions and subtopics within your core subject area that you haven't yet addressed. Any gaps in your coverage are gaps in your topical authority signal. A competitor who has answered a question you haven't is one point of authority ahead of you on that topic.
Regular keyword research audits — comparing your existing content against the full keyword landscape of your topic — reveal these gaps systematically. Filling them is how you move from a site with good individual rankings to a site Google treats as the definitive resource in your space.
What Most Guides Won't Tell You Topical authority is not just about volume of content. It's about coherence. A site with 200 loosely related posts has less authority signal than a site with 50 tightly organised posts that comprehensively cover a defined topic territory. Keyword research helps you define the boundaries of your topic territory and stay within them — which is why it's a strategic function, not just a content planning task.
When building your topic cluster map, include keywords that are slightly adjacent to your core topic but strongly relevant to your buyers' world. Covering these adjacent questions extends your authority perimeter and captures buyers at earlier stages of their awareness journey.
Creating cluster content that's too thin or too similar to the pillar page. Each cluster page should answer a specific question with enough depth to stand alone as the best resource on that subtopic. Generic 200-word cluster posts do not build authority signals.
This is the question most guides skip entirely — and it's the one that separates teams who improve their keyword research from teams who repeat the same mistakes indefinitely. Measuring the success of keyword research is not the same as measuring rankings. Rankings are an intermediate output. The outcomes you should actually be tracking are further downstream.
The Three-Layer Measurement Model
Layer 1: Ranking Indicators Track position changes for your target keywords over time. These are your leading indicators — they show whether your content is competing effectively in search results. What you're looking for isn't just page-one rankings but ranking trajectory: is position improving, stabilising, or declining? Trajectory matters more than point-in-time position.
Layer 2: Traffic Quality Metrics Rankings without traffic mean little. Monitor organic sessions to your keyword-targeted pages, but more importantly, monitor engagement signals: time on page, pages per session, scroll depth, and return visits. High engagement signals indicate that the keyword you targeted was accurately matched to the content you created and to the searcher's intent. Low engagement despite strong rankings often means an intent mismatch — revisit the PESO Signal Framework score for that keyword.
Layer 3: Pipeline and Conversion Metrics This is where keyword research success or failure ultimately registers. Which keyword-targeted pages are producing email subscribers, demo requests, contact form submissions, or purchases? Annotate your analytics with the keyword intent classification you assigned during research, then compare conversion rates across intent categories. This analysis, done quarterly, will show you clearly whether your keyword selection is generating revenue or just traffic.
The Keyword Research Retrospective Every quarter, run a simple retrospective: look at your highest-traffic organic pages and check whether they match the keywords you targeted. Look at your highest-converting organic pages and check whether they came from your PESO-scored priority keywords. The gap between what you planned and what performed is where your keyword research system improves.
Over time, this retrospective practice builds an institutional knowledge base of which keyword types, intent signals, and specificity levels produce real commercial outcomes in your specific market — knowledge that no tool can give you and that compounds into a significant competitive advantage.
Set up a simple tracking document that logs: target keyword, publish date, current ranking, organic sessions, and conversion events for each piece of content. Review it monthly. After six months, you'll have clear evidence of which keyword types perform best in your specific market — data that's more valuable than any industry benchmark.
Measuring keyword research success at 30 or 60 days. SEO operates on a longer timeline. Most keyword-targeted content takes four to six months to reach stable rankings in competitive markets. Evaluating performance too early leads to abandoning strategies that would have compounded effectively with patience.
Gather first-party buyer language. Pull sales call notes, support tickets, onboarding surveys, and customer reviews. Extract raw phrases customers use to describe their problems before finding your solution.
Expected Outcome
A list of 20-40 buyer vocabulary phrases that form the foundation of your Buyer Vocabulary Gap research.
Generate seed keywords before opening any tool. List your product categories, synonyms, competitor names, and the questions your sales team hears most frequently. Combine with your buyer vocabulary phrases.
Expected Outcome
A diverse seed keyword list of 50-80 terms ready for tool-based expansion.
Run expansion research in your keyword tool. Collect volume, difficulty, and related terms. Export broadly — cast a wide net at this stage. Focus particularly on question variations, comparison variations, and specificity variations.
Expected Outcome
An expanded keyword list of 200-400 terms with raw data attached.
Apply the PESO Signal Framework. Score every keyword 0-4 across Paid presence, Exact-match transactional phrasing, SERP commercial features, and Outcome-specific language. Prioritise keywords scoring 3-4.
Expected Outcome
A filtered priority keyword list of 40-80 high-commercial-value targets.
Validate with manual SERP checks. For each priority keyword, open the search results in an incognito browser. Confirm intent alignment. Note dominant content types, SERP features, and competitive landscape.
Expected Outcome
Intent-verified keyword list with notes on what content type Google expects for each term.
Build your topic cluster map. Group priority keywords into pillar topics and supporting cluster topics. Identify your top three pillar topics and five to eight cluster keywords per pillar.
Expected Outcome
A structured topic cluster architecture ready to inform your content calendar.
Prioritise and sequence your content plan. Order your keyword targets by: commercial intent score, domain authority fit, and strategic business priority. Assign target keywords to content briefs.
Expected Outcome
A sequenced 90-day content plan with keyword targets, intent classification, and success metrics defined for each piece.
Set up your measurement framework. Create a keyword tracking document covering target keyword, publish date, ranking, organic sessions, and conversion events. Schedule a quarterly keyword retrospective.
Expected Outcome
A live measurement system that will compound your keyword research intelligence over time.