Keyword research is the foundation of every effective SEO strategy. Get it right and you're building content around queries with real demand, realistic ranking potential, and a clear path to conversion. Get it wrong and you're producing content that ranks for nothing — or worse, ranks for traffic that never converts.
In 2026, keyword research has evolved. The explosion of AI search and the increasing sophistication of Google's semantic understanding has changed which keywords are worth targeting and how you should organise them. This guide covers the full process from first principles.
What Is Keyword Research?
Keyword research is the process of identifying the search queries your target audience uses, understanding the intent behind those queries, assessing their demand and competition, and mapping them to the right content.
The output of keyword research isn't a list of words — it's a strategic content plan built around the searches your audience is actually making.
Good keyword research answers three questions:
- What are people searching for (query discovery)?
- Why are they searching for it (intent analysis)?
- Can you realistically rank for it (competition assessment)?
Step 1: Start with Search Intent
Search intent — the underlying reason behind a query — should be the first thing you establish for any keyword, before you look at volume or competition.
Google classifies intent into four categories:
Informational — The user wants to learn something. "What is schema markup?", "how does PPC work?", "why is my website slow?" These queries drive content pages, guides, and educational resources.
Navigational — The user wants to find a specific site or page. "Dynamically digital marketing agency", "Google Search Console login." Optimising for navigational queries is primarily about maintaining brand presence.
Commercial investigation — The user is researching before a purchase decision. "Best SEO agency Liverpool", "Ahrefs vs Semrush comparison", "digital PR services review." These drive comparison content, service pages, and review management.
Transactional — The user is ready to act. "Hire SEO agency", "get SEO audit", "PPC management quote." These queries should land on conversion-optimised service or product pages.
Intent determines content type. A transactional query never converts on a blog post; an informational query will never meaningfully convert on a bare service page. Misaligning content with intent is one of the most common reasons pages don't rank.
How to identify intent
For any keyword, the quickest way to assess intent is to search for it and look at the results. Google has already done the intent analysis — the SERP layout tells you what content type ranks. If the top results are all blog posts and guides, that's an informational query. If they're ecommerce category pages, it's transactional. If they're comparison articles, it's commercial investigation.
Step 2: Seed Keyword Discovery
Seed keywords are the broad terms that define your topics. They're not what you'll target directly — they're the starting points for finding the specific queries worth optimising for.
Sources for seed keywords
Your own knowledge — What does your business do? What problems do you solve? What does your audience call those problems? Start with your core topics and work outward.
Your website and analytics — Google Search Console shows what queries your site already appears for. GA4 shows which pages get organic traffic. These are your existing keyword positions — the baseline you're building from.
Your competitors' content — What are your competitors ranking for that you're not? Screaming Frog or site: searches give you their content landscape; Ahrefs and Semrush can show their top-ranking keywords specifically.
Customer language — Customer support tickets, sales call recordings, online community discussions (Reddit, LinkedIn groups), and product reviews reveal the vocabulary your audience actually uses.
Step 3: Keyword Expansion
With a list of seed topics, expand to specific queries using research tools.
Paid tools
Ahrefs Keywords Explorer — The industry benchmark for keyword data. Enter a seed term and get related keyword suggestions, search volume, keyword difficulty (KD), and click potential. The "Questions" filter is particularly useful for informational and AEO research.
Semrush Keyword Magic Tool — Similar capability to Ahrefs with a slightly different data set. The "Intent" filter is useful for sorting keywords by their dominant intent signal.
Google Keyword Planner — Free (requires a Google Ads account). Provides search volume ranges and competition data. Less precise than Ahrefs/Semrush but reliable and free.
Free tools
Google Search Console — Shows actual query data for your site. Filter by impressions to find queries you're appearing for but not ranking well — these are high-priority optimisation targets.
Google Autocomplete — Type your seed term in Google and examine the autocomplete suggestions. These reflect real search behaviour. Also scroll to "Related searches" at the bottom of the SERP.
Answer The Public — Visualises the questions people ask around a topic. Excellent for informational keyword discovery and FAQ content planning.
Google Trends — Shows relative search interest over time and by region. Useful for understanding seasonality and whether a topic is growing or declining.
Step 4: Keyword Filtering and Prioritisation
Expanding from seed keywords typically generates hundreds or thousands of keyword ideas. You need to filter and prioritise.
Keyword difficulty vs domain authority
Keyword difficulty (KD) scores are relative. A KD of 40 might be unachievable for a site with DR 20 but very winnable for a DR 60 domain. Always assess difficulty against your site's current authority.
A practical approach: pull the top 10 ranking pages for your target keyword and compare their DR and backlink counts to yours in Ahrefs. If the top 5 pages all have DR 70+ and 500+ referring domains, and you have DR 35 with 50 referring domains, that keyword is not a realistic near-term target regardless of what the KD score says.
Volume is relative
Don't chase volume — chase relevance. A keyword with 50 monthly searches from your exact target customer is worth more than a keyword with 5,000 searches from an audience that will never convert. For B2B businesses especially, low-volume highly-specific keywords often deliver the highest conversion rates.
Prioritisation framework
Score each keyword candidate on:
- Relevance — How closely does this query map to your product/service?
- Intent match — Can you create content that genuinely serves this intent?
- Attainability — Does your current authority make ranking realistic within 6–12 months?
- Business value — If you rank and convert on this query, what is it worth?
Keywords that score high on all four dimensions are your priority targets.
Step 5: Long-Tail Keyword Strategy
Long-tail keywords — specific, lower-volume queries — are where most businesses win organic visibility, particularly early in a programme.
According to Ahrefs, 91.8% of all keywords get 10 or fewer searches per month. This long tail of search represents enormous collective volume and typically lower competition than head terms.
Why long-tail works:
- Lower keyword difficulty for newer or lower-authority sites
- Higher conversion rates (more specific intent = more qualified traffic)
- Easier to match content precisely to what the user wants
- Compound value: ranking for hundreds of long-tail terms generates significant aggregate traffic
Building a long-tail strategy:
- Cluster related long-tail queries under topic areas rather than creating one page per keyword
- A single comprehensive page targeting a topic can rank for dozens of related long-tail variants simultaneously
- Use FAQ sections to explicitly address the specific questions that generate long-tail query traffic
Step 6: Keyword Research in the AI Era
AI search has introduced new considerations for keyword research.
Natural language queries
AI search users are more likely to type natural language questions than keyword fragments. "What are the best digital marketing tactics for a small business in 2026?" rather than "small business digital marketing 2026."
This doesn't mean abandoning traditional keyword research — Google still processes billions of exact-match and near-match queries. But it does mean your content should also cover the natural language variants of your target queries, and your heading structure should include question-form headings alongside keyword-targeted ones.
Zero-click keywords
Some informational queries increasingly receive direct answers in Google AI Overviews without generating any click to source pages. This changes the calculus for purely informational keywords — the traffic may be lower than the search volume implies.
The response isn't to avoid informational content (being cited in the AI Overview still generates brand visibility and some traffic) but to ensure your informational content links strategically to commercial pages.
Topic authority over keyword targeting
AI systems — and increasingly Google's ranking algorithm — reward demonstrated topical authority over keyword-by-keyword optimisation. A site that comprehensively covers a topic area from multiple angles is more likely to rank consistently across the topic than one that targets keywords in isolation.
This reinforces the pillar-and-cluster content model: a comprehensive pillar page covering a broad topic, supported by cluster articles covering specific sub-topics in depth.
Step 7: Keyword Mapping
With your prioritised keyword list, map each keyword (or keyword cluster) to a specific page on your site — either existing pages you'll optimise, or new content you'll create.
The mapping should be:
- One primary keyword per page — the main term you're optimising for
- Supporting keywords — related terms with similar intent that the same page will naturally capture
- No keyword cannibalisation — two pages on your site shouldn't target the same primary keyword, as they'll compete against each other
Document your keyword map in a spreadsheet with columns for: URL, primary keyword, intent, monthly search volume, current ranking position, and target position.
FAQs
How often should I do keyword research? A comprehensive keyword research exercise should be done at the start of any SEO programme, then refreshed annually. Quarterly reviews for your top-priority pages are good practice — competitive landscapes shift, new opportunities emerge, and intent patterns evolve.
Do keyword tools show accurate search volumes? All keyword tools estimate search volume from different data sources. Ahrefs and Semrush are the most reliable for paid tools, but volumes are still estimates. Use them for relative comparison (this keyword is more searched than that one) rather than absolute targets.
Should I target branded keywords? Yes. Branded search terms — people searching for your company name — should be monitored and owned. They're typically very easy to rank for (you're the most relevant result) and often represent high-conversion traffic from people already familiar with your brand.
How do I find keywords my competitors rank for? In Ahrefs or Semrush, use the competitor analysis features: enter a competitor's domain and filter for keywords where their site ranks in the top 10. This surfaces their organic strategy and reveals gaps where you could compete.
What's the difference between a keyword and a topic? A keyword is a specific search query. A topic is the broader subject area that encompasses many related keywords. Modern SEO is increasingly about building topical authority — dominating a subject area across many related keywords — rather than targeting individual keywords in isolation.
Building a keyword strategy that drives real organic results starts with the right research. Talk to our SEO team about developing a keyword architecture built around your specific business goals.
