10 Claude Prompts for SEO Keyword Research Success

Keyword research is the decision that everything else in SEO depends on. Pick the wrong keywords and you spend months creating content that never ranks, serves no one, or ranks for queries that send zero qualified traffic. Pick the right ones and every piece of content you publish compounds — building topical authority, attracting the exact audience you want, and generating organic traffic that does not disappear when you stop paying for it. The right Claude prompts for SEO keyword research give you a structured system for finding those right keywords — grounded in search intent, competitive reality, and your specific domain situation.

Below are 10 prompt patterns for every dimension of keyword research — from initial discovery to intent mapping to competitor gap analysis to long-tail opportunities. Each includes a ready-to-use example, an explanation of why it works, and a tip for getting even more from it.

Traditional keyword research tools give you data: search volume, keyword difficulty, cost-per-click. What they do not give you is interpretation. They cannot tell you which keywords match your domain authority, which intent categories align with your content capabilities, or which competitor gaps represent realistic opportunities for your specific site. Claude bridges the gap between raw keyword data and actionable strategy — helping you interpret what the numbers mean for your situation and decide which keywords are actually worth pursuing.

The prompts below are designed to do exactly that. They work best when combined with data from keyword tools like Ahrefs, Semrush, or Google Search Console — you bring the numbers, Claude brings the strategic thinking that turns those numbers into a prioritised keyword plan you can actually execute.

Every keyword strategy starts with a seed — a broad topic that represents your core business or content area. This prompt systematically expands a seed keyword into every relevant sub-topic, modifier category, and audience angle that could generate rankable content.

"Expand the following seed keyword into a comprehensive map of related keyword opportunities. Seed keyword: [X]. My business or site: [describe what you do and who you serve]. For this seed, identify: (1) 5 broader parent topics this keyword sits within, (2) 10 more specific sub-topic keywords that drill deeper into the subject, (3) 8 modifier categories — e.g. 'best', 'how to', 'vs', 'for beginners', 'near me', 'cost', 'alternatives' — applied to this topic, (4) 5 audience-specific angles — how different audience segments search for this topic differently, (5) 3 adjacent topic areas I might be missing that are closely related to this seed but not immediately obvious. Organise the output by category."

Why it works: The modifier categories instruction is what produces the most actionable output. Searchers rarely type just the seed keyword — they add modifiers that reveal intent: ‘best’ signals comparison intent, ‘how to’ signals learning intent, ‘cost’ signals purchase intent. Mapping these modifiers systematically surfaces keyword opportunities most people discover only through months of trial and error. The adjacent-topics question consistently reveals the keywords a site should own but is completely ignoring.

Search intent — what the searcher is actually trying to do — determines what type of content should rank for a keyword. Getting intent wrong is the most common reason well-written content fails to rank. This prompt classifies a list of keywords by intent and explains the content implications of each classification.

"Classify the following keywords by search intent and explain the content implications of each. Keywords: [paste a list of 10-20 keywords]. For each keyword, identify: (1) the primary intent — informational (want to learn), navigational (want to find a specific site), commercial investigation (want to compare before buying), or transactional (want to complete a purchase or action), (2) the secondary intent if present — e.g. a keyword that is primarily informational but has a commercial sub-intent, (3) the content type most likely to rank — blog post, product page, comparison page, landing page, tool, FAQ, (4) the stage in the buyer journey this keyword represents, (5) whether I should target this keyword with new content or optimise an existing page. Flag any keywords where the intent is ambiguous and explain why."

Why it works: The secondary intent instruction captures the nuance most keyword research misses. A keyword like ‘best project management software’ is primarily commercial investigation — but it also has a strong informational sub-intent because searchers want to understand features before comparing. Content that addresses only the commercial intent without the educational sub-intent will rank below content that satisfies both. The new-content-vs-existing-page instruction turns the classification into an immediate editorial decision.

Your competitors are ranking for keywords you are not. Some of those keywords are irrelevant to your business; others represent your highest-value ranking opportunities. This prompt analyses competitor keyword coverage to identify the specific gaps worth prioritising.

"Help me identify keyword gaps by analysing my competitors' content coverage. My site: [describe your niche and current content focus]. My top competitors: [name 3-5 competitors]. What I know about my competitors' content: [describe the topics and content types they cover that you do not]. My current keyword focus areas: [list the topics you already cover]. Based on this, identify: (1) the topic categories my competitors cover that I have no content for, (2) within those gaps, the 5 highest-priority keywords to target first — based on likely search volume and the probability I could rank for them given my site, (3) the gaps that represent a strategic risk — competitor territory that is pulling my potential audience to their site, (4) one topic area where I could leapfrog competitors by going deeper than they have, (5) the gap that would produce the fastest ranking results if I created content for it now."

Why it works: The leapfrog-by-going-deeper instruction is the most strategically valuable output. Keyword gap analysis typically identifies topics to add; this instruction identifies the specific topic where creating more comprehensive content than anyone else currently has will be rewarded with rankings even without matching a competitor's domain authority. The fastest-ranking-results instruction prioritises the gap with the most immediate payoff rather than the most important long-term opportunity.

Long-tail keywords — specific, lower-volume queries with three or more words — are where most sites can realistically rank, especially in the early stages of domain authority building. This prompt generates high-value long-tail variations from a broader target keyword.

"Generate a comprehensive set of long-tail keyword variations for the following target keyword: [broad keyword]. My site context: [describe your niche, audience, and domain authority level — new / mid-level / established]. Generate long-tail variations across these categories: (1) problem-based — keywords that describe a specific problem my audience is trying to solve, (2) solution-specific — keywords that name a specific method, tool, or approach, (3) comparison and alternative — '[X] vs [Y]' and '[X] alternative' patterns, (4) audience-qualified — keywords that include a specific audience descriptor like 'for freelancers', 'for enterprise', 'for beginners', (5) location or context qualified — if relevant to my business. For each variation, note whether the intent is informational, commercial, or transactional, and flag the 5 with the highest likely conversion value for my business."

Why it works: The audience-qualified category is what most long-tail generators miss. ‘Project management software’ is highly competitive; ‘project management software for construction companies’ is specific enough that a domain with moderate authority can rank for it and attract a highly targeted audience. Qualification by audience type, use case, or context is one of the most reliable strategies for finding long-tail keywords with both ranking potential and commercial value.

Having a list of keywords is not a strategy. Knowing which to pursue first — based on the ratio of ranking potential to traffic opportunity for your specific domain — is what makes keyword research executable. This prompt prioritises a keyword list based on the specific characteristics of your site.

"Help me prioritise the following list of keywords for execution. My site situation: Domain age: [X years]. Estimated domain authority: [low / medium / high, or specific DA if known]. Current content volume: [approximate number of published pages]. Monthly organic traffic: [approximate]. Keywords to prioritise: [paste your keyword list with any data you have — volume, difficulty, current ranking position]. For each keyword, assess: (1) whether it is realistic for my domain to rank on page one given my authority level, (2) the likely traffic value if I do rank on page one, (3) whether there is an existing page I could optimise vs. needing to create new content. Then group the keywords into: Quick wins (rank within 3 months), Medium-term targets (3-9 months), Long-term targets (9+ months). Recommend the first 5 I should act on this month."

Why it works: The three-tier grouping by time horizon is the most practically useful output for planning content production. Most keyword research ends with a flat list sorted by volume or difficulty. A list sorted by realistic ranking timeline for your specific domain tells you what to build now, what to build next quarter, and what to build when your authority has grown. The first-five-to-act-on instruction collapses the prioritisation into a concrete next step.

Question-based keywords — how, what, why, when, which — are among the highest-value keyword types because they signal specific informational intent and often populate Google’s People Also Ask boxes, featured snippets, and voice search results. This prompt mines question keywords systematically.

"Generate a comprehensive list of question-based keywords around the topic: [topic]. My audience: [describe — who they are, what level of knowledge they have, what problems they are trying to solve]. Generate questions in each category: (1) definition and explanation questions — 'what is X', 'what does X mean', 'how does X work', (2) how-to and process questions — 'how to X', 'how do I X', 'steps to X', (3) comparison and evaluation questions — 'is X better than Y', 'X vs Y', 'which X is best for [use case]', (4) troubleshooting questions — 'why is my X not working', 'X not [doing something]', 'fix X [problem]', (5) cost and value questions — 'how much does X cost', 'is X worth it', 'X pricing'. For each question, identify which are most likely to appear in Google's People Also Ask and which are most likely to generate a featured snippet if answered concisely."

Why it works: The featured-snippet identification is the strategically important output. Featured snippets — the boxed answer that appears above organic results — generate disproportionate click-through rates and can be earned even by sites with moderate authority if the answer is formatted correctly. Identifying which question keywords are most likely to trigger a featured snippet tells you where to invest in concise, direct answer formatting rather than comprehensive article treatment.

Keyword cannibalisation — where multiple pages on your site compete for the same keyword — is one of the most common and most damaging SEO problems. It confuses search engines about which page to rank, splits authority between pages, and often results in neither page ranking as well as a single consolidated page would. This prompt diagnoses cannibalisation in your existing content.

"Help me identify and resolve keyword cannibalisation in my content. I will describe my pages and their target keywords: [list your pages with their URLs and target keywords — or describe the topic clusters you have]. Based on this: (1) identify any groups of pages that are targeting the same or overlapping keywords, (2) for each cannibalisation group, recommend whether to consolidate into one page, differentiate by intent (keep both but make intent clearly distinct), or redirect the weaker page to the stronger, (3) identify which page in each group should be the primary ranking target and why, (4) flag any topic areas where I am likely creating future cannibalisation by continuing to publish on the same theme without a clear keyword differentiation plan, (5) suggest a keyword mapping system I could use to prevent cannibalisation in future content planning."

Why it works: The future-cannibalisation-prevention instruction is the most valuable long-term output. Cannibalisation audits typically address existing problems; this instruction identifies the systemic content planning habit that keeps creating new cannibalisation. A keyword mapping system — where every piece of content is assigned a unique target keyword before creation — is the fix that prevents the problem recurring rather than just clearing the backlog.

Search volume is not static. Some keywords peak at predictable times of year; others are rising or falling over multi-year trends. This prompt maps the seasonal and trend patterns in your keyword universe to help you time content publication for maximum impact.

"Help me identify seasonal patterns and emerging trends in my keyword universe. My niche: [describe your industry or content topic]. My primary keyword themes: [list the main topic areas you cover]. Based on typical patterns in this niche: (1) identify which of my keyword themes are likely to have significant seasonal variation and describe the typical peak periods, (2) identify keywords in my niche that are likely growing year-over-year — driven by technology change, cultural shifts, or industry evolution, (3) identify any keywords that are likely declining and should be deprioritised, (4) recommend specific content publication timing — how many months before a seasonal peak should I publish to give content time to rank, (5) suggest one emerging topic in my niche that has not yet become a highly competitive keyword but is likely to within the next 12 months."

Why it works: The publication timing instruction is the most practically valuable output. SEO content needs time to rank — typically 3-6 months for an established domain. Publishing a seasonal piece in the week before the peak produces content that ranks just as the peak ends. Understanding that you need to publish 3-4 months before a seasonal peak is the insight that separates timely content from perpetually late content. The emerging-topic instruction is the highest-upside output — content that ranks before a keyword becomes competitive earns authority that is nearly impossible to displace later.

For businesses with a local or geographic dimension — service businesses, multi-location retailers, location-specific content sites — geographic keyword research requires a different approach than general SEO keyword research. This prompt builds a geographic keyword strategy.

"Build a geographic keyword strategy for my business. My business: [describe — service type, locations served, whether you have a physical location or serve clients remotely]. My target locations: [list the cities, regions, or countries you want to rank in]. My core services or content topics: [list]. Build a strategy covering: (1) the primary local keyword patterns I should target — '[service] in [city]', '[service] near me', '[city] [service]' — and which pattern typically has higher volume in my industry, (2) the difference in approach between targeting a large metro vs. a smaller city or region, (3) how to create content that ranks for location keywords without creating near-identical pages for every location (duplicate content risk), (4) local intent modifiers specific to my service type — e.g. 'emergency', 'same day', 'licensed', 'affordable', (5) the one location I should target first and why, given my current domain strength."

Why it works: The duplicate-content risk instruction is the most important constraint in local SEO keyword strategy. The obvious approach — creating identical service pages for every location you serve, just with the city name swapped — is penalised by Google as thin duplicate content. Understanding how to create genuinely differentiated location pages while targeting geographic keywords is what separates a local SEO strategy that works from one that gets the site flagged.

Keyword research produces a list. A content strategy turns that list into a publishing plan that builds topical authority systematically. This prompt bridges the gap between a completed keyword research exercise and an executable content production plan.

"Help me turn my keyword research into a content strategy. My prioritised keywords: [paste your keyword list with intent classification and priority tier]. My content production capacity: [X pieces per month]. My domain authority and site age: [describe]. Build a content strategy that covers: (1) how to group these keywords into topic clusters — identify the pillar page for each cluster and the supporting pages that link to it, (2) the publication sequence that builds topical authority most efficiently — which cluster to start with and why, (3) how to handle keywords across different intent types in the same cluster — e.g. combining informational and commercial investigation pages in a coherent architecture, (4) a 90-day publishing plan that matches my production capacity to the priority keywords, (5) the one keyword that I should treat as my flagship piece — the content that, if it ranks, would do the most for my overall SEO trajectory."

Why it works: The flagship-piece instruction is the most strategically clarifying output. Most content plans treat all pieces as roughly equal in importance. Identifying the one piece of content that would most transform your SEO trajectory — the topic cluster anchor, the high-authority magnet, the piece most likely to earn backlinks — focuses creative and production energy on the content that will do the most work. The 90-day plan constraint matches ambition to reality by working within actual production capacity rather than generating a plan that looks great on paper and fails in execution.

The most effective keyword research workflow runs these prompts in sequence. Start with the Seed Keyword Expander to map the full landscape of your topic area. Run the Search Intent Classifier on the resulting list. Use the Competitor Keyword Gap Finder to identify what you are missing. Generate long-tail variations for your highest-priority gaps. Run the Keyword Difficulty vs. Opportunity Prioritiser to build your execution sequence. Then bridge to content strategy with the final prompt. The entire sequence can be completed in a single focused session and produces a keyword strategy that most sites never develop.

Save the prompts that match your keyword research workflow as reusable templates in Chat Smith so you can deploy the Seed Keyword Expander at the start of any new content initiative, the Intent Classifier before briefing a writer, and the Prioritiser whenever your keyword list grows faster than your content production capacity — all in one click without rebuilding the prompt each time.

Using these prompts steers you away from the most consistent keyword research failures. Targeting head keywords with high search volume but competition you cannot match produces content that ranks on page five and generates zero traffic. Ignoring search intent produces content that is well-written but misaligned with what searchers actually want to find — high bounce rates and poor dwell time signal to Google that the content is not serving the query. Creating content across too many topics without depth in any of them produces no topical authority and no rankings. Failing to map keyword cannibalisation produces a growing site where individual pages undermine each other.

Each prompt in this guide addresses one of these failure modes. The Difficulty vs. Opportunity Prioritiser addresses unrealistic keyword targeting. The Intent Classifier addresses misaligned content. The Topic Cluster bridge addresses shallow topical coverage. The Cannibalisation Auditor addresses self-defeating page architecture. The pattern is the same: keyword research that goes beyond data into strategic interpretation produces rankings; data collection without interpretation produces a spreadsheet nobody acts on.

Keyword research is not a task you do once before launching a site. It is an ongoing intelligence practice — monitoring what is ranking, what is changing, what competitors are doing, and what new search behaviours your audience is developing. Done systematically, it transforms a content programme from output-based (we publish X pieces per month) to outcome-based (we rank for the keywords our buyers are searching). These 10 Claude prompts for SEO keyword research give you the system to conduct that practice at depth — whether you are starting from scratch or refining a programme that has been running for years. Start with the Seed Keyword Expander. Let the strategy build from there.

A systematic keyword research practice involves seed expansion, intent classification, competitor analysis, long-tail generation, prioritisation, cannibalisation audits, and content strategy bridging — all of which benefit from having the right prompt instantly available at each stage. Keeping all of those prompts organised and ready is exactly where Chat Smith comes in. Chat Smith is an all-in-one AI platform that lets you save every keyword research prompt as a reusable template, organise them by research stage, and launch any prompt in one click across Claude, GPT, Gemini, and other leading models.

Instead of rebuilding your intent classification prompt every time you expand into a new topic area, or hunting for your competitor gap analysis before a quarterly SEO review, Chat Smith gives you a clean, searchable library of your best-performing prompts. You can run the same seed expansion prompt across multiple AI models to compare the keyword landscapes they identify, share your keyword research prompt library with a content team or SEO agency, and build a keyword intelligence practice that compounds in quality with every research session you run.

1. Do I need a paid keyword tool to use these prompts?

No — but the prompts that involve prioritisation and competitive analysis become significantly more useful when you can provide Claude with real search volume and difficulty data. Free tools like Google Search Console, Google Keyword Planner, and Ahrefs Webmaster Tools provide enough data to make the Prioritiser and Gap Finder prompts highly actionable. Paid tools like Ahrefs or Semrush provide richer data that makes the entire workflow more precise. The prompts work at any budget level; the data quality determines how specific the strategic outputs can be.

2. How many keywords should I target at once?

Each page should target one primary keyword and 3-5 closely related secondary keywords that share the same intent. Trying to target 10 unrelated keywords on a single page produces a page that is unfocused and ranks for none of them. Across your site, the number of keywords in active targeting should roughly match your content production capacity: if you publish 4 pieces per month, your active keyword list for the next quarter should be around 12 priorities. The Keyword Difficulty vs. Opportunity Prioritiser prompt is designed to help you maintain a manageable, executable keyword focus.

3. How often should I update my keyword research?

A full keyword research refresh every 6-12 months is sufficient for most sites, with ongoing monitoring of your ranking positions, emerging competitor content, and Google Search Console query data monthly. Markets that change quickly — technology, finance, health — may warrant more frequent review. The Seasonal and Trending Keyword Planner prompt should be run at least quarterly to ensure your content calendar is aligned with the search behaviour patterns in your niche.

4. What is the difference between keyword research for a new site vs. an established site?

A new site has low domain authority and must target low-competition long-tail keywords to earn its first rankings and build the authority base required for more competitive terms. An established site with meaningful authority can target head terms with higher competition, but should still use long-tail and question keywords to build topical depth. The Keyword Difficulty vs. Opportunity Prioritiser prompt is explicitly calibrated to your domain situation — it produces different recommendations for a new site vs. an established one using the same keyword list, which is why it asks for your domain age and authority before prioritising.