Blog › SEO & GEO Glossary
Keyword Research
The process of finding the exact terms your buyers type into search engines — and in 2026, the questions they ask AI platforms. Keyword research is still the starting point for all content strategy, but what counts as a “keyword” has expanded.
Keyword research is the process of identifying the specific words, phrases, and questions that people use when searching for information, products, or services related to your business. It determines which terms to target in content, which search intent to serve, and where organic traffic opportunities exist. Effective keyword research is the foundation of both SEO content strategy and GEO prompt optimisation.
What is search intent and why does it matter?
Every keyword reflects a user’s underlying intent — what they are actually trying to do. Understanding intent is more important than matching exact keyword strings.
Informational intent
The user wants to learn something. Queries like “what is keyword research” or “how does schema markup work.” These queries feed blog posts, glossary pages, and educational content. In 2026, these are also the queries most likely to be answered by AI Overviews before the user clicks anything.
Navigational intent
The user wants to find a specific website or brand. Queries like “Ahrefs login” or “NBound SEO glossary.” Navigational queries have low conversion value for non-brand sites but are critical for protecting your own brand terms.
Commercial investigation intent
The user is researching options before buying. Queries like “best GEO software 2026” or “Peec AI vs Profound.” These are the highest-value content opportunities for B2B companies — and the queries where AI-generated answers now play a significant role in shaping buyer opinions before any click happens.
Transactional intent
The user is ready to act. Queries like “buy SEO audit” or “managed SEO agency Surrey.” These drive direct conversions and should map to landing pages or service pages, not blog posts.
💡 Keyword research in 2026 means prompt research too. 68% of Google searches now end without a click, and a growing share of B2B research happens in ChatGPT, Perplexity, and Google AI Mode. The questions buyers ask AI platforms are longer and more conversational than traditional search queries. Effective keyword research now includes mapping the prompts your buyers use in AI tools — not just the short-tail terms they type into Google. This is the foundation of GEO (Generative Engine Optimisation).
How to do keyword research: a practical process
Start with seed keywords. Seed keywords are the broad topic areas your business covers. For an SEO agency: “SEO services”, “link building”, “technical SEO”. Expand seeds into longer queries using tools like Ahrefs Keywords Explorer, Google Search Console, or the “People also ask” and autocomplete features in Google.
Analyse search volume and keyword difficulty together. High-volume keywords are competitive. A keyword with 200 monthly searches and low competition often drives more traffic faster than a keyword with 20,000 searches dominated by well-funded competitors. The metric to focus on is realistic ranking potential, not volume in isolation.
Identify AI Overview-intercepted queries. In Google Search Console, look for queries with high impressions but very low click-through rates. These are queries where a Google AI Overview is answering the question before anyone clicks your result. These queries need a GEO strategy — your goal is to be cited inside the AI Overview, not just to rank below it.
Cluster keywords by topic and intent. A keyword cluster is a group of related terms that share the same search intent and can be served by a single page. “Keyword research tool”, “best keyword research tools”, and “keyword research software” all serve the same intent and belong on the same page. Clustering prevents cannibalization and builds topical depth.
Map keywords to content types. Informational keywords → blog posts and glossary pages. Commercial investigation keywords → comparison pages and reviews. Transactional keywords → service pages and landing pages. Mismatching content type to intent is one of the most common keyword strategy failures.
⚠️ Watch out: Monthly search volume figures are estimates, not precise counts. Different tools produce different volume numbers for the same keyword because they use different data sources and sampling methods. Use volume as a relative comparison metric between keywords, not as an absolute traffic prediction. A keyword with 500 searches per month will not drive 500 visits per month even at position one — actual CTR depends on SERP features, competition, and whether an AI Overview intercepts the query first.
Frequently asked questions about keyword research
What is keyword research in SEO?
Keyword research in SEO is the process of identifying the search terms people use when looking for information, products, or services related to your business. It involves analysing search volume (how often a term is searched), keyword difficulty (how competitive it is to rank for), and search intent (what the user actually wants). The output of keyword research is a prioritised list of terms to target in content, mapped to appropriate page types and organised into topic clusters.
What tools are used for keyword research?
The most widely used keyword research tools include Ahrefs Keywords Explorer, Google Search Console (for data on terms you already rank for), Google Keyword Planner, Semrush, and Moz Keyword Explorer. For prompt research — finding what buyers ask AI platforms — tools like Profound’s Conversation Explorer (400M+ real user prompts) and Peec AI’s prompt suggestion feature surface AI-native query patterns that do not appear in traditional keyword tools.
What is the difference between short-tail and long-tail keywords?
Short-tail keywords are broad, typically one or two words with high search volume and high competition (“SEO”, “keyword research”). Long-tail keywords are longer, more specific phrases with lower search volume but higher conversion intent and less competition (“how to do keyword research for a B2B SaaS website”). Most content strategies prioritise long-tail keywords because they are more achievable, closer to buying intent, and better matched to the conversational query patterns used in AI searches.
How often should keyword research be updated?
Keyword research should be revisited at least every six months, and immediately after any significant market change, product launch, or Google algorithm update. Search behaviour evolves, new competitors enter, and AI Overview coverage shifts which queries drive clicks and which are intercepted. Google Search Console provides ongoing signal about which queries are losing CTR to AI Overviews — reviewing this monthly keeps the research current without requiring a full re-run every time.
Does keyword research apply to AI search and GEO?
Yes, but the unit shifts from keywords to prompts. In AI search, users ask questions in natural language rather than typing abbreviated keyword strings. The equivalent of keyword research for GEO is prompt research — identifying the conversational questions your target buyers ask ChatGPT, Perplexity, or Google AI Mode about your category. These prompts typically start with “what is the best”, “which tool should I use for”, “how do I choose between”, and “what do companies use for”. Mapping these prompts and optimising content to answer them is the GEO equivalent of traditional keyword-to-content mapping.
Related glossary terms: On-Page SEO · Technical SEO · Featured Snippets · Schema Markup · GEO Guide
Services: Managed SEO — keyword research and topic cluster strategy included in every retainer · Managed GEO Services — prompt research and AI visibility tracking