Keyword Research Essentials: Ranking in the AI Era

Keyword research has always been the starting point for effective SEO, but what it means to do it well has changed substantially. Search engines no longer reward pages that match exact phrases; they reward pages that answer intent clearly, cover a topic with genuine depth, and signal authority across a subject area.

This guide on keyword research essentials covers the full keyword research process for UK and Irish businesses, from defining objectives and mapping search intent to localising for regional audiences and optimising for AI-driven results. Whether you are new to SEO or reviewing a strategy that has stopped delivering, the framework below gives you a structured, practical approach.

You will find sections on the five core research steps, UK and Irish market nuances, AI search optimisation, and the tools worth your time at every budget level.

What Keyword Research Actually Means Now

The basics are still the basics: keyword research is the process of identifying the terms and phrases people use when searching for something online, then deciding which of those terms your content should target. What has changed is the layer of analysis that sits on top of that process.

From Strings to Entities

Search engines have shifted from matching character strings to understanding entities. An entity is a real-world thing: a person, a place, a concept, a product. When Google processes a query, it does not just look for the literal words; it maps the query to a network of related entities and surfaces the pages that best address the underlying topic.

For keyword research, this means thinking in clusters rather than individual phrases. A page targeting “local SEO tips” is not just competing for that phrase; it is competing across a semantic neighbourhood that includes terms like “Google Business Profile optimisation,” “local search ranking factors,” and “how to appear in the local map pack.” Your research should map that whole neighbourhood, not just the headline phrase.

Why Search Volume is a Secondary Metric

Search volume tells you how many people search for a term each month. It tells you nothing about whether those people will buy, enquire, or engage meaningfully with your business. A term generating 10,000 monthly searches in the UK might be driven by students researching an essay topic; a term generating 200 monthly searches might represent buyers ready to spend.

The better primary metric is business value: how closely does this query match the problem your product or service solves? A keyword with modest volume and strong commercial intent will almost always outperform a high-volume keyword with diffuse intent. Learn more about this principle through ProfileTree’s SEO services, where intent mapping sits at the core of every strategy.

The Zero-Click Reality

A growing proportion of Google searches end without a click. Featured snippets, AI Overviews, People Also Ask boxes, and knowledge panels answer the question directly on the results page. This does not make keyword research irrelevant; it makes it more important to target queries where the user genuinely needs to visit a page to get a complete answer. Informational queries with simple factual answers are increasingly zero-click territory. In-depth guides, comparisons, and process-oriented content retain click-through value because the results page cannot fully replicate them.

The Role of Topical Authority

Google’s evaluation of a page does not happen in isolation. The site it sits on, the internal links pointing to it, and the breadth of related content around it all feed into how the page is assessed. A strong keyword strategy accounts for this by building content clusters: a pillar page covering a broad topic supported by several focused pieces addressing specific subtopics. Each piece reinforces the authority of the others.

The Five-Step Keyword Research Framework

Effective keyword research follows a consistent sequence regardless of industry or budget. The five steps below map to the full process, from initial topic identification through to content assignment.

Step 1: Define Business Objectives and Seed Topics

Before opening any tool, write down what you want to achieve. Are you trying to generate enquiries for a specific service? Drive traffic to a product page? Build awareness in a new geographic market? The answer shapes every decision that follows.

From those objectives, identify your seed topics: the broad subject areas your business operates in. A web design agency might identify seeds like “website design,” “CMS platforms,” “website speed,” and “conversion rate optimisation.” These seeds become the starting points for tool-based research. Do not filter yet; the goal at this stage is breadth.

Step 2: Analyse Search Intent

For every keyword you are considering, identify the intent behind it. The four categories used across the industry are informational (the user wants to learn), navigational (the user wants to find a specific website or page), transactional (the user wants to complete an action or purchase), and commercial investigation (the user is comparing options before deciding).

The content format you need changes with the intent. Informational queries call for guides, tutorials, and explanatory articles. Commercial investigation queries call for comparisons, case studies, and transparent criteria. Transactional queries call for landing pages with clear calls to action. Matching format to intent is one of the most direct ways to improve both rankings and conversions. ProfileTree’s approach to content marketing is built around this principle.

Step 3: Competitor Gap Analysis

Take a shortlist of competitors and feed their domains into a keyword research tool. Look for terms they rank for that you do not. These gaps represent opportunities where there is proven search demand but no presence from your site. Prioritise gaps where the intent closely matches your commercial objectives and where the keyword difficulty is within range, given your current domain authority.

Pay attention to the pages they have built around those terms, not just the terms themselves. A competitor ranking for a keyword cluster with a single complete guide signals that a similar guide from you, with better depth or a different angle, could capture a share of that traffic. Review their use of meta keywords and on-page signals for additional context on how they are structuring their SEO.

Step 4: Evaluate Metrics

With a list of candidate keywords, score each one across three dimensions: search volume (how often it is searched in your target market), keyword difficulty (how hard it is to rank for given current top-ranking pages), and business value (how closely it matches your commercial objectives). No single metric tells the full story; the combination does.

Cost-per-click data from Google Keyword Planner is a useful proxy for commercial intent, even if you are not running paid campaigns. Advertisers pay more for terms that convert; high CPC values are a reasonable signal that the query attracts buyers rather than browsers.

Step 5: Map Keywords to the Content Funnel

Assign each keyword to a stage in the buying journey: awareness, consideration, or decision. Awareness-stage keywords support top-of-funnel content (educational guides, explainers, trend pieces). Consideration-stage keywords support comparison and evaluation content. Decision-stage keywords support service pages, pricing pages, and case studies.

This mapping exercise often reveals imbalances. Many sites have reasonable coverage at the awareness stage and almost nothing at the consideration and decision stages, which is precisely where commercial value sits. Addressing that gap directly improves the return on your content investment. ProfileTree’s digital strategy team works through exactly this process with SME clients across Northern Ireland, Ireland, and the UK.

Localisation: The UK and Irish Market

The majority of keyword research content online is written from a US perspective, with US search volume data, US terminology, and US examples. For businesses operating in the UK and Ireland, that creates both a blind spot and an opportunity. The blind spot is that US-centric data can misrepresent real demand in your market. The opportunity is that localised content is significantly less competitive than its generic equivalent.

Terminology Differences That Affect Volume

Vocabulary differences between US English and UK English are more significant than they first appear. A term with substantial US search volume may have minimal demand in the UK simply because the preferred phrasing is different. Common examples include “attorney” versus “solicitor,” “apartment” versus “flat,” “sneakers” versus “trainers,” and “vacation” versus “holiday.” Running both variants through a keyword tool with UK region filters applied will often show stark differences.

For Irish businesses, additional regional phrasing applies. Terms like “estate agent” (UK and Ireland) versus “realtor” (US), or specific product and service terminology with Irish-market usage, are worth researching separately from their UK equivalents. The dialect and cultural specificity of Northern Irish search behaviour, for example, can differ meaningfully from London or Dublin patterns. Connolly Cove’s guide to Northern Ireland’s cities is one example of how local context shapes the framing of content for this audience.

Regional Intent Differences

Local intent varies by geography in ways that purely national-level keyword data obscures. A user in Belfast searching for “web design services” is expressing fundamentally different commercial intent from a user in London doing the same search; the competitive landscape, pricing expectations, and preferred suppliers differ substantially. Tools like Google Keyword Planner allow you to filter by country, and Ahrefs and SEMrush allow regional filtering, but none of them offers city-level granularity for organic search in most cases.

The workaround is combining national tool data with local intent signals: Google Search Console data filtered to your target geography, Google Trends with location settings applied, and direct observation of local SERPs. What ranks in Belfast for a given query may not be what ranks in Dublin or Edinburgh; understanding those local SERP landscapes directly improves your targeting decisions.

UK-Specific Research Tools and Data Sources

Google Keyword Planner is the most accessible starting point for UK volume data, though its estimates are broad. Ahrefs and SEMrush both offer UK-specific filtering and provide more granular data on local difficulty and SERP composition. For identifying regional trends, Google Trends set to the United Kingdom or Ireland is more useful than global data, particularly for seasonal patterns.

UK search behaviour has distinct seasonal drivers: post-Budget commercial activity, Q4 retail patterns, and financial year-end cycles all affect demand in ways that US data will not reflect. ProfileTree’s SEO services for UK and Irish businesses account for these regional factors as a standard part of the research process.

Building a UK and Irish Keyword Table

A practical way to surface localisation opportunities is to build a comparison table of your seed terms against their US and UK variants, with volume data for each in both markets. Terms where the UK volume is a small fraction of the US volume flag potential underperformance from US-centric research. Terms where the UK has comparable or disproportionate volume relative to its population size flag strong local intent worth targeting specifically.

Optimising for AI Overviews and Generative Search

Search behaviour has changed more in the past two years than in the preceding decade. AI-generated answers from Google’s AI Overviews, Perplexity, ChatGPT, and Gemini now appear above or alongside traditional organic results for a growing proportion of queries. For keyword research, this creates a new layer of targeting: not just ranking for a query, but being the source that AI systems cite when answering it.

How AI Systems Select Sources

AI systems in search draw from pages that are authoritative on a topic, structured in a way that makes information easy to extract, and up to date. Pages that open each section with a direct, concise answer to the implied question, cover multiple sub-questions within a topic, and use clear heading hierarchies are systematically overrepresented in AI citations. This is not fundamentally different from good SEO practice; it is an extension of it.

“The shift we are seeing is from keyword ranking as the primary metric to citation frequency in AI answers. For businesses targeting the UK SME market, being cited in a Perplexity or ChatGPT answer about digital marketing services is increasingly as valuable as holding a top-five position in organic results,” says Ciaran Connolly, founder of ProfileTree.

Finding Keywords That Trigger AI Snippets

Queries that tend to trigger AI Overviews are typically informational or mixed intent, phrased as questions or broad topic searches, and cover subjects where there is no single obvious best answer. Process queries (“how to do X”), comparison queries (“X versus Y”), and definition queries (“what is X”) are reliable AI Overview territory. Targeting these query types with structured, answer-first content gives you the best chance of appearing in the AI layer.

For transactional and commercial investigation queries, AI Overviews appear less frequently, but when they do, they tend to cite trusted sources with strong domain authority and verifiable credentials. Building that authority through consistent, high-quality content across a topic cluster is the long-term approach. ProfileTree’s work on AI transformation for businesses covers how this shift affects marketing strategy more broadly.

Natural Language Queries and Voice Search

AI-assisted search has accelerated the shift toward conversational, natural-language queries. Where a 2015 user might have searched “keyword research tools,” a 2026 user is more likely to ask “what are the best free keyword research tools for a small business in the UK.

” The implications for your keyword targeting are meaningful: FAQ-style content, question-based subheadings, and direct, plain-English answers are not just good editorial practice; they are structural optimisations for AI-driven search. ProfileTree’s AI training programmes for SMEs cover this shift in accessible, practical terms.

Structured Content for AI Extraction

Pages that AI systems extract from share a common structural pattern: they answer the query in the first one or two sentences of each section, use descriptive subheadings that could stand alone as questions or statements, and contain at least one well-structured data table or comparison element. Writing your content with extraction in mind does not mean writing for robots; it means being clear, direct, and well-organised, which benefits human readers equally.

Tools, Tracking, and Common Mistakes

The tools available for keyword research span free options that are more than adequate for most SME use cases through to enterprise platforms costing several hundred pounds per month. What matters is not which platform you use but whether you are using it consistently and interpreting the outputs correctly.

Tools at Every Budget Level

At the free tier, Google Keyword Planner provides volume ranges and related keyword suggestions. Google Search Console shows which queries your existing pages already rank for, making it one of the most valuable and underused research tools available. Google Trends provides seasonality and regional data. AnswerThePublic and AlsoAsked visualise the question-based queries surrounding a topic, which directly feeds FAQ and subheading planning.

At the paid tier, Ahrefs and SEMrush are the dominant platforms, offering keyword difficulty scores, SERP history, competitor gap analysis, and traffic estimates. SEMrush’s Keyword Magic Tool and Ahrefs’ Keywords Explorer are both capable of generating large, well-organised keyword lists from seed terms. For businesses running Google Ads alongside organic search, the paid data and organic data often inform each other usefully.

For businesses just beginning, a combination of Google Search Console, Google Keyword Planner, and AnswerThePublic covers the core research process at no cost. The gap between free and paid tools is real, but not disqualifying at the SME level. ProfileTree’s digital training programmes teach exactly this workflow to business owners and marketing managers who want to manage their own SEO.

Grouping Keywords and Avoiding Cannibalisation

Once you have a keyword list, organise it into thematic clusters before assigning content. Each cluster should represent a distinct topic that warrants its own page. Two pages competing for the same keyword cluster dilute each other’s ranking potential rather than strengthening the site’s overall position.

Cannibalisation is a common problem on larger sites where content has been produced over several years without a structured mapping exercise. If you have two or more pages targeting near-identical queries, the options are to merge them into a single, authoritative piece or to differentiate them clearly enough that search engines treat them as addressing different intents. Review your existing content in Google Search Console, filtered by query, to identify where this is happening. Evolving keyword insertion is one paid-search technique that sometimes masks cannibalisation issues on organic pages; a useful overview of how it works is available in ProfileTree’s guide to evolving keyword insertion.

Tracking Performance and Updating Strategy

Keyword research is not a one-off exercise. Search behaviour shifts, competitors publish new content, and algorithm updates change what the SERP landscape looks like for a given query. A minimum review cycle of every six to twelve months is sensible for most SMEs; sites in fast-moving sectors may need quarterly reviews.

Use Google Search Console as your primary performance tracker. Monitor impressions, clicks, and average position for your target queries. A page with high impressions and a low click-through rate is likely ranking in a position where it is being seen but not compelling enough to click; the fix is usually a title tag or meta description revision. A page with declining impressions suggests a ranking drop that needs investigation. For a deeper look at what drives organic performance, ProfileTree’s guide to organic traffic drops covers the common causes and remedies.

Pitfalls Worth Knowing

Chasing volume without intent alignment is the most common mistake. The second is neglecting branded search: queries that include your company name, product names, or combinations thereof. Branded queries are often high-intent and easy to rank for; failing to optimise for them means competitor content or third-party reviews occupy the space.

The third is ignoring content freshness: pages that haven’t been updated in two or more years may hold rankings in stable niches but are increasingly vulnerable as competitors publish more current material. Treat content maintenance as an ongoing commitment, not a post-publication afterthought.

Conclusion

Keyword research done well connects your business to the questions your audience is actually asking, at the stage in their journey where your answer is most useful. For UK and Irish businesses, that means layering regional intent and local terminology over the standard process, and accounting for the growing proportion of searches that now surface AI-generated answers rather than a list of blue links.

If you are ready to build a keyword strategy that works for your market, talk to the ProfileTree team.

FAQs