The Best Claude Prompts for SEO (Copy-Paste Templates That Actually Work)

The best Claude prompts for SEO are specific, role-framed, and built around a concrete deliverable — not open-ended questions. Most prompt lists online give you thirty generic templates and call it a workflow. The difference between Claude producing a genuinely useful keyword cluster and producing a forgettable topic list comes down entirely to how precisely you frame the task. Below are copy-paste templates organized by real SEO workflow stage, with notes on what each prompt is actually doing — and where human judgment is non-negotiable before anything touches your site.

Why Prompt Structure Matters More Than the Prompt Itself

Claude follows layered instructions unusually well. That means the way you structure a prompt — the role, the context, the format constraint — shapes the result more than the topic you ask about. A vague prompt produces a vague answer. A prompt that specifies audience, intent, format, and limitations produces something usable on the first pass.

Every template below follows the same four-part structure:

• Role: Tell Claude what kind of expert it's acting as — "senior on-page SEO auditor," "keyword strategist," "technical SEO consultant."
• Context: Give it the specifics it can't guess — your niche, existing content, target audience, competitor data.
• Task: Be precise about the deliverable. "A list" is weak. "A prioritized list of 10 items with a one-line rationale for each" is far stronger.
• Format: Tell Claude exactly how to structure its output — table, numbered list, JSON-LD, Q&A pairs. Claude follows format instructions reliably.

When all four components are present, Claude's output becomes genuinely usable. When one is missing, you get something generic that needs significant rework. It's also worth being honest about what Claude cannot do: it has no access to live ranking data, cannot crawl your site, and will occasionally produce plausible-sounding statistics that are simply wrong. For diagnosis that requires crawl data, Search Console integration, and competitive intelligence layered together, a proper SEO audit and strategy process is still essential. These prompts speed up the thinking — they don't replace the analysis.

Before building your workflow around Claude, it's also worth reading our breakdown of what Claude does well in SEO content writing and where it genuinely falls short — it will save you from the most common over-reliance mistakes.

Best Claude Prompts for SEO Audits

Audits are where Claude earns its keep fastest. It can triage a list of technical issues, map on-page gaps, and prioritize fixes — all in a format you can hand to a developer or work through yourself. These prompts slot directly into the early diagnostic phase of any site review.

On-Page Audit Prompt

Act as a senior on-page SEO auditor. I'm auditing a [industry] website targeting [primary keyword]. Review the following page elements: [paste title tag, meta description, H1, first 150 words, internal links, image alt text]. For each element: rate it 1–10, explain what's weak, and give a specific rewrite or fix. Prioritize by likely ranking impact. Format as a numbered checklist, highest impact first.

Why it works: Prioritization by impact is the critical instruction. Without it, Claude lists everything at equal weight — which is useless when you have limited time.

Technical Issue Triage Prompt

Act as a technical SEO consultant. I have the following crawl issues from a recent audit: [list issues — e.g., 47 broken internal links, 12 pages with duplicate title tags, 3 redirect chains of 3+ hops, missing canonical tags on paginated pages]. Rank each issue by potential ranking impact. For each: explain what SEO damage it's causing now, what fixing it will achieve, and how long a fix typically takes. Present as a table with columns: Issue | SEO Impact | Fix | Time Estimate.

Competitor Gap Analysis Prompt

Act as an SEO strategist. My site covers [topic/niche] and my main competitor is [competitor description]. They outrank me for [keyword cluster]. Based on strong SEO practice, identify five likely content or structural advantages they may have over me. Then suggest five specific actions I could take to close that gap within 90 days. Be specific — no generic advice.

Best Claude Prompts for Keyword Research and Search Intent

Claude doesn't have live search volume data, so don't use it like a keyword tool. Use it for what it's genuinely good at: mapping search intent, structuring clusters, and expanding a seed list you already have. Always cross-reference output against Ahrefs, Semrush, or Google Keyword Planner before building content around it.

Keyword Cluster Builder Prompt

Act as a keyword strategist. My primary topic is [topic] for a [business type] targeting [audience]. Generate a topical cluster of 20 keywords: five head terms, eight long-tail variations, and seven question-based keywords matching informational intent. For each, note: (1) search intent type (informational, commercial, transactional, navigational), (2) funnel stage (awareness, consideration, decision), (3) recommended content format. Present as a grouped list by type.

Intent Classification Prompt

Act as a search intent analyst. Here is a list of keywords I'm targeting: [paste list]. Classify each by primary intent (informational, navigational, commercial, transactional). Then flag any where the likely SERP format — list, guide, product page, tool — conflicts with the type of content I'd need to create. Format as a table: Keyword | Intent | Likely SERP Format | Content Type Required.

SERP Feature Targeting Prompt

Act as an SEO content strategist. I want to target featured snippets and People Also Ask boxes for the topic [topic]. Identify 10 question formats that commonly trigger these SERP features. For each question, write a 40–50 word direct answer formatted for snippet extraction. Present as question-answer pairs.

Best Claude Prompts for Content Briefs and On-Page Writing

This is where Claude is most commonly misused. People use it to write complete articles when they should be using it to build rigorous briefs. A well-structured brief — produced in minutes — will improve the quality of whatever gets written afterward, whether by a human or AI. Claude's drafts consistently flatten nuance, avoid strong opinions, and occasionally hallucinate statistics, so treat all content output as a first draft requiring expert editorial review before publication.

Full Content Brief Prompt

Act as a senior SEO content strategist. Create a detailed content brief for an article targeting the keyword [target keyword]. Target audience: [describe — experience level, job title, pain point]. Content goal: [rank for X / support Y funnel stage / answer Z question]. Include: recommended title (3 variants), meta description draft, H1, H2 and H3 structure with intent notes per section, word count recommendation, key entities to mention, internal linking suggestions, and one differentiated angle that sets this piece apart from generic coverage. Do not include filler sections — every H2 must earn its place.

Meta Description Batch Prompt

Act as an SEO copywriter. Write three meta description options for each of the following pages: [paste page titles or topics + target keywords]. Each meta description must be under 155 characters, include the primary keyword naturally, and end with a clear reason to click. Present as a table: Page | Option 1 | Option 2 | Option 3.

Note that Google's own documentation on title links confirms it may rewrite your title if it determines yours doesn't match page content — so everything you give Claude should reflect what's actually on the page, not just the keyword you're targeting.

Content Refresh Prompt

Act as a content SEO specialist. I have an existing article targeting [keyword] that is underperforming. Here is the current content: [paste article or key sections]. Identify: (1) sections that are thin or don't fully satisfy search intent, (2) topics competitors likely cover that this article misses, (3) outdated claims or data needing updates, (4) structural improvements to aid scannability and featured snippet eligibility. Provide specific rewrite suggestions, not general notes.

Best Claude Prompts for Schema and Structured Data

Structured data is one of the most underused applications of Claude for SEO. Claude is genuinely strong at generating schema markup because it's a well-defined, rule-governed task with little ambiguity — and it's faster and more accurate than most online generators. Always validate output at schema.org's official validator before deployment, and run it through Google's Rich Results Test to catch logical errors that look valid but aren't.

FAQ Schema Prompt

Act as a technical SEO specialist. Generate valid JSON-LD FAQ schema markup for the following questions and answers, following schema.org/FAQPage specification exactly. Questions and answers: [list Q&As]. Output as clean, copy-paste-ready JSON-LD.

Article and Local Business Schema Prompt

Act as a technical SEO developer. Generate JSON-LD schema for the following page: [describe the page — article, product page, local business, how-to guide]. Include these details: [paste relevant fields — title, author, published date, address, steps, etc.]. Follow schema.org standards. After the markup, list any additional schema types I should consider adding for this page type.

Best Claude Prompts for Internal Linking Strategy

Internal linking is one of the highest-leverage on-page improvements available — particularly for sites with existing content that isn't passing authority effectively. It's also consistently skipped in most AI SEO prompt guides.

Internal Link Opportunity Prompt

Act as an on-page SEO specialist. Here is a list of articles on my site with their target keywords: [paste page titles, URLs, and keywords]. Identify the five most logical internal linking opportunities between these pages. For each, specify: source page, target page, recommended anchor text, and where in the source page the link should appear (intro, body, conclusion). Prioritize links that channel authority toward my highest-value commercial pages. Present as a table.

Anchor Text Variation Prompt

Act as an SEO copywriter. I need to link to my page targeting [target keyword] from multiple locations across my site. Generate 10 varied anchor text options including: exact match, partial match, semantic variants, and natural phrase variations. Flag which anchors are safest for link profile diversity and which carry the highest keyword signal.

Best Claude Prompts for AI Search Optimisation — The Category Most Lists Skip

Nearly every existing Claude SEO prompt list is optimized entirely for traditional Google rankings. None of them address the growing importance of being cited in AI-generated answers — which increasingly operate by different content signals than standard page-one rankings.

As Search Engine Journal has documented, AI Overviews are reshaping how Google surfaces answers, and the content that earns citations follows different rules than the content that simply ranks well. We've covered the mechanics of this in depth in our analysis of whether AI Overviews are reducing organic clicks, and in our guide to how AI search engines decide what to cite — both of which are useful context for the prompts below.

AI Citation Readiness Prompt

Act as a generative engine optimization (GEO) specialist. Review the following content targeting [keyword]: [paste content]. Identify five specific ways this content could be rewritten to increase the likelihood of being cited in an AI-generated answer. Focus on: factual density, definition clarity, structured formatting, source authority signals, and entity disambiguation. Present as a list of specific, actionable edits with before/after examples where possible.

Entity Coverage Audit Prompt

Act as an entity SEO specialist. For the following content targeting [topic]: [paste content]. Identify: (1) key entities — people, organisations, concepts, locations — currently mentioned, (2) important related entities that are missing and should be included to strengthen topical authority, (3) whether entity relationships are made explicit or left implicit. Suggest additions that would help AI systems better understand the topic context and entity associations.

According to Moz's analysis of link quality signals, editorial relevance is the primary factor that makes external authority meaningful — and the same logic applies to AI citation: content that clearly belongs to a topic, is well-attributed, and uses explicit entity relationships is far more likely to be surfaced. If you're not yet thinking about how your content performs in AI-generated responses alongside traditional rankings, the distinction between SEO, GEO, and AEO is worth understanding before you build your next content plan.

SEO Task	Claude's Usefulness	Output Format to Request	Human Review Required?
On-page audit checklist	High	Numbered list, prioritized	Yes — validate against crawl data
Technical issue triage	High	Table: Issue | Impact | Fix	Yes — confirm against crawl output
Keyword cluster generation	Medium	Grouped list by intent	Yes — cross-reference volume data
Content brief creation	High	Structured brief with H2s	Yes — editorial review needed
Meta descriptions (batch)	High	Table: Page | 3 options	Yes — edit for brand voice
Schema markup	High	Clean JSON-LD	Yes — validate at schema.org
Internal linking strategy	Medium	Table with anchor + placement	Yes — verify anchor diversity
AI citation optimisation	Medium	Actionable edit list	Yes — requires SEO judgment

Prompt Hygiene: Getting Better Results From Every Template

Even the best templates degrade without good habits. A few principles that make a consistent difference:

• Always paste real material. Prompts with "here is the actual content: [paste]" outperform abstract requests every time. Claude's reasoning improves dramatically when it has specific text to evaluate rather than a general topic to guess about.
• Iterate, don't regenerate. If output is 70% right, follow up with targeted corrections rather than rerunning the whole prompt. "The competitor analysis section is too general — go deeper on [specific angle]" produces better results than starting over.
• Say "no generic advice" explicitly. Claude responds well to that constraint at the end of a prompt. It forces specificity rather than templated recommendations.
• Treat statistics with suspicion. Claude will generate plausible-sounding data. Verify any figures against primary sources like Google Search Central or published research before publishing.
• Save your strongest prompts. Build a personal prompt library. The prompts that produce strong, usable output for your specific niche are worth more than any generic list — including this one.

FAQ

Can Claude replace dedicated SEO tools like Ahrefs or Semrush?

No. Claude has no access to live keyword data, search volume, ranking positions, or backlink indexes. It works best alongside those tools — helping you interpret and act on the data they surface rather than replacing the data itself. Use it for ideation, clustering, brief creation, and structured output; use dedicated tools for measurement and validation.

How do I stop Claude from giving generic SEO advice?

Add specificity to every prompt: name your industry, paste real page content, give it actual data to respond to, and add "do not give generic advice — every recommendation must be specific to the content provided" at the end. Claude responds well to that constraint. The more raw material you give it, the more targeted its output.

Are Claude-generated meta descriptions good enough to publish directly?

Often close, but rarely publish-ready without a light edit. Claude tends to produce meta descriptions that are grammatically correct and include the target keyword, but they can be generic in tone and occasionally exceed 155 characters. Treat them as strong first drafts and edit for brand voice, specificity, and character count before publishing.

Should I use Claude to write complete SEO articles and publish them directly?

Not without significant editing. Claude's drafts typically lack first-hand experience, real examples, original data, and the kind of specific insight that earns links and citations in AI-generated answers. Treat its output as a structured first draft that needs expert review and editorial judgment before it's production-ready. The strongest AI-assisted SEO workflows use Claude as a research accelerator and structure generator — not an autonomous content producer.