Most articles about Claude and SEO give you the same thing: a list of single-sentence prompts that produce the same generic output you would get from any AI tool. “Write an SEO article about [keyword].” That is not a prompt. That is a coin toss.
The quality of Claude’s output is directly proportional to the specificity of your prompt. The editing time disappears, content quality jumps, and the workflow goes from six hours per post to ninety minutes when you engineer your prompts properly.
This guide gives you prompts organized by the actual SEO job you are trying to do. Keyword research, content briefs, on-page optimization, technical audits, meta tags, internal linking, schema markup, and AI citation optimization. Every prompt is copy-paste ready. Replace the brackets with your own details and run it.
One thing before you start: replace every bracket. Every [like this] in the prompts below is a placeholder. Generic input produces generic output. The prompts work because they are structured. They fail when you leave the placeholders blank.
Why Claude Handles SEO Prompts Better Than Other AI Tools
Claude 4.6 has a massive context window and a unique affinity for structured inputs. This allows you to separate your core instructions from your background data, ensuring the model never loses the thread during complex technical audits. By formatting your inputs correctly, you turn a generative model into a strict, analytical SEO assistant.
Standard language models suffer from “lost in the middle” syndrome. If you upload a massive Screaming Frog export to them, they tend to analyze the first hundred rows, hallucinate the middle, and read the end. Claude minimizes this degradation. Its recall accuracy remains high even when processing dense, unformatted CSV data.
That structural advantage is why the prompts in this guide are formatted the way they are. Give Claude a role, a task, context, constraints, and a required output format. That combination produces output you can actually use.
How to Structure Any Claude SEO Prompt
The best Claude prompts include a specific role, clear task definition, format requirements like word count and structure, constraints about what NOT to do, and relevant context. Claude follows complex, multi-part instructions more reliably than most models.
Every prompt in this guide follows this structure:
Role → Tell Claude who it is. “Act as a senior SEO strategist with 10 years of experience” produces dramatically better output than just asking a question.
Task → State the specific deliverable clearly.
Context → Give Claude the background it needs. Your niche, audience, existing pages, competitor URLs.
Constraints → Tell Claude what to avoid. This is as important as telling it what to do.
Output format → Specify exactly how you want the response structured.
Keyword Research Prompts
Prompt 1: Keyword Cluster Builder
Use this when you have a raw list from a keyword tool and need to organize it into a content strategy.
You are an SEO strategist. I will paste a list of keywords from a keyword research tool.
TASK: Group them into topical clusters. For each cluster, suggest one
pillar page and three supporting posts.
CONTEXT:
- Niche: [your niche, e.g. "AI SEO tools for small businesses"]
- Primary audience: [e.g. "freelance SEO consultants"]
- Existing pages we already have: [list URLs or topics, or write "none yet"]
CONSTRAINTS:
- Cluster by search intent, not just word overlap
- Flag any keywords that cannibalize each other
- Mark each cluster as informational, commercial, or transactional
OUTPUT FORMAT:
Cluster name | Pillar page title | 3 supporting posts | Intent type
KEYWORDS:
[paste your keyword list here]Prompt 2: Search Intent Analysis
Before writing any page, you need to know what Google is actually rewarding for that query. This prompt handles that.
Act as an SEO analyst. I want to understand the search intent behind this keyword before I create content.
KEYWORD: [your target keyword]
TASK:
1. Classify the intent: informational, navigational, commercial, or
transactional
2. Describe what the searcher is actually trying to accomplish
3. Tell me what content format Google is most likely rewarding
(listicle, how-to guide, comparison page, tool page, etc.)
4. List 5 related questions this searcher probably also has
5. Identify one angle that is underserved based on what typically
ranks for this intent
Do not give me a generic answer. Be specific about what this particular
keyword signals.Prompt 3: Featured Snippet Targeting
This approach is how SEOs consistently rank in featured snippets. W-questions and step-by-step content work best. Claude finds the questions and writes the answers at the same time.
Act as an SEO specialist focused on featured snippet optimization.
TASK: Give me 15 keywords or questions in [industry/niche] that are
likely to trigger featured snippets or People Also Ask boxes.
Focus on W-questions (who, what, where, when, why, how) and
step-by-step queries.
For each keyword include:
- The search intent
- The type of featured snippet it likely triggers (paragraph, list,
table, steps)
- A 50-word answer optimized for that snippet format
INDUSTRY/NICHE: [your niche]
TARGET AUDIENCE: [describe your reader]Content Brief and Outline Prompts
Prompt 4: Full Content Brief
A detailed content brief is what separates a Claude article that ranks from one that just fills a page. Include the keyword, search intent, competitor URLs, and the structure you want before asking Claude to write anything.
You are a senior SEO content strategist. Create a detailed content brief for the following article.
PRIMARY KEYWORD: [keyword]
SECONDARY KEYWORDS: [list 3-5 related terms]
TARGET WORD COUNT: [1,500 / 2,500 / 3,500]
TARGET AUDIENCE: [describe reader, experience level, goals, pain points]
CONTENT GOAL: [rank for keyword / capture leads / build topical authority]
INCLUDE IN THE BRIEF:
- Recommended H1 title (include keyword near the front)
- Meta description under 155 characters with a clear hook
- 6-8 H2 sections with H3 subsections where needed
- For each section: what to cover, what question it answers,
any data points or examples to include
- FAQ section with 5 People Also Ask-style questions
- Internal link suggestions (list pages on our site that should
link to this)
- What NOT to include: filler, generic intros, restatements of
the title
Do not write the article. Write the brief only.Prompt 5: Competitor Gap Analysis
Use this before creating content on any competitive topic. Paste competitor content directly into the prompt.
Act as an SEO content analyst.
TASK: Analyze the competitor content I have pasted below and identify:
1. Core NLP entities, industry terms, and semantic keywords they
all share
2. Specific subtopics, questions, or data points they cover that
my draft completely misses
3. Content angles or perspectives that are absent from all of them
(this is your gap opportunity)
Then give me a prioritized list of additions I should make to outrank
these pages.
TARGET KEYWORD: [keyword]
MY DRAFT OR OUTLINE: [paste your draft or write "not started yet"]
COMPETITOR 1: [paste content or URL]
COMPETITOR 2: [paste content or URL]
COMPETITOR 3: [paste content or URL]Prompt 6: Topic Cluster Planning
One prompt. A full month of content planned.
Act as a content strategist with deep SEO knowledge.
I want to build topical authority in [niche] and rank for
[primary topic or keyword].
TASK: Build me a 90-day content calendar with:
- 1 pillar page (topic + recommended title + target keyword)
- 8-10 supporting cluster posts that link back to the pillar
- For each piece: target keyword, search intent, recommended
format, and which existing page it should internally link from
EXISTING SITE CONTENT: [list your main pages or write "starting fresh"]
AUDIENCE: [describe your reader]
GOAL: [organic traffic / lead generation / topical authority]
Prioritize topics by ranking difficulty. Start with lower-competition
supporting posts before the pillar.On-Page Optimization Prompts
Prompt 7: Page Scoring and Fix List
Paste any existing page into this prompt and get a specific action list back.
Act as an on-page SEO specialist.
TASK: Score this page out of 10 for SEO quality and give me a
specific list of fixes ordered by ranking impact.
FOR EACH FIX INCLUDE:
- What the problem is
- Why it matters for rankings
- Exactly what to change (be specific, not general)
EVALUATE:
- Title tag and meta description
- H1, H2, H3 structure
- Keyword placement and density
- Internal linking
- Content depth and gaps
- E-E-A-T signals (author bio, expertise signals, citations)
- Schema markup presence or absence
TARGET KEYWORD: [keyword]
PAGE CONTENT: [paste full article text or HTML]Prompt 8: Meta Tag Generator at Scale
This prompt cuts meta tag writing from 20 minutes to 2 minutes. Five options per page. Pick the best one and move on.
Act as a technical SEO and copywriting specialist.
I have a [type of page] targeting the keyword [your keyword].
Write:
- 5 meta title variations under 60 characters each with different
angles (curiosity, benefit, question, urgency, direct)
- 5 meta description variations under 155 characters each
- The recommended schema markup type for this page with a
brief explanation of why it helps SEO
For titles: put the keyword near the front. No clickbait.
No punctuation at the end.
For descriptions: end each one with a soft call to action.
PAGE TYPE: [blog post / product page / landing page / category page]
TARGET KEYWORD: [keyword]
PAGE TOPIC: [brief description of what the page covers]Prompt 9: CTR Recovery for Ranking Pages
Use this on any page sitting between positions 4 and 10 with a click-through rate under 3 percent.
You are a CRO-minded SEO writer.
TASK: Rewrite the title tag and meta description for a page
that ranks but does not get clicked.
CURRENT STATE:
- Page URL: [URL]
- Current title: [current title]
- Current meta: [current meta description]
- Target keyword: [keyword]
- Current CTR: [X]% at position [X]
- Audience: [who is searching this]
CONSTRAINTS:
- Title 60 characters or fewer, keyword near the front
- Meta 155 characters or fewer, ends with a soft call to action
- Do not imply content the page does not deliver
- Match intent: commercial or informational based on the keyword
OUTPUT:
- 5 title options ranked by predicted CTR
- 3 meta descriptions ranked by predicted CTR
- One-line reasoning for your top pickTechnical SEO Prompts
Prompt 10: Technical Audit Prioritization
Technical audits are overwhelming. You get twenty issues and no idea where to start. This prompt cuts through all of that. Paste the issues in, get a priority list out.
Act as an SEO strategist focused on technical performance.
I am running a technical audit and found these issues:
[list issues — 404 errors, redirect chains, slow page speed,
missing canonical tags, duplicate content, crawl errors,
missing schema, etc.]
TASK:
1. Rank these by SEO impact from highest to lowest
2. Tell me what fixing each one will achieve in terms of rankings
3. Flag anything causing immediate ranking loss right now
4. Separate into: fix this week / fix this month / fix later
SITE TYPE: [blog / ecommerce / SaaS / local business]
DOMAIN AGE: [rough estimate]
MONTHLY ORGANIC TRAFFIC: [rough estimate or "unknown"]Prompt 11: Schema Markup Generator
Act as a technical SEO specialist.
TASK: Generate valid JSON-LD schema markup for the following page.
PAGE TYPE: [FAQ page / article / product / local business /
how-to guide / review]
PAGE TOPIC: [describe the page]
For FAQ schema, here are my Q&A pairs:
Q: [question 1]
A: [answer 1]
Q: [question 2]
A: [answer 2]
[continue for all Q&As]
Output only the JSON-LD code block. No explanation before or after.
No "here is your schema" prefix. Just the clean, paste-ready code.Prompt 12: Internal Linking Audit
For best results, paste the full text of each donor page alongside the target page. The model identifies anchor text that reads naturally and flags any placement that would disrupt reading flow.
You are a technical SEO doing an internal linking audit.
TASK: Suggest internal link opportunities to strengthen
this target page.
TARGET PAGE:
- URL: [URL]
- Topic: [topic]
- Why it matters: [e.g. "ranks at position 12 for a high-value keyword"]
DONOR PAGES (pages already ranking well on related topics):
[list URLs and topics for each]
CONSTRAINTS:
- Suggest anchor text that uses the target keyword or a close variant
- Anchors must read naturally, not forced
- Flag any donor page where adding the link would disrupt reading flow
OUTPUT FORMAT:
| Donor URL | Sentence to insert or modify | Suggested anchor text |Content Writing Prompts
Prompt 13: Long-Form Article
Specifying what to avoid is as important as specifying what you want. Telling Claude not to use “In today’s digital world” or “It’s important to note” is not optional. It changes the output significantly.
You are an expert content writer specializing in [niche].Write a [word count]-word article titled: "[title]"
TARGET KEYWORD: [keyword] — include naturally in the first 100 words
and 2-3 headings
AUDIENCE: [describe reader — experience level, goals, pain points]
TONE: [e.g. practical and direct / conversational / authoritative]
STRUCTURE: Introduction → [number] H2 sections → FAQ section
DO NOT USE:
- "In today's digital world"
- "It's important to note"
- "In conclusion"
- Em dashes
- The words: delve, tapestry, underscore, showcase, pivotal, foster,
nuanced, garner
INCLUDE:
- One specific example per section
- Actionable advice the reader can use today
- At least one data point or stat
Write like you are explaining this to a colleague, not presenting
at a conference.Prompt 14: Article Introduction Rewrite
Use this on any piece where the opening is weak or sounds like AI wrote it.
Act as a copywriter who specializes in opening paragraphs for SEO content.
This introduction is too generic. Rewrite it.
RULES:
- Open with a specific fact, claim, or observation
- Do not start with "In today's..." or "When it comes to..."
- Do not warm up. Get to the point in the first sentence.
- Keep it under 100 words
- Make the reader want to keep reading
TARGET KEYWORD: [keyword]
TARGET AUDIENCE: [who is reading this]
CURRENT INTRODUCTION:
[paste your current intro]GEO and AEO Prompts
These are the prompts most SEOs are not using yet. GEO is Generative Engine Optimization, structuring your content to be cited by AI systems like ChatGPT, Perplexity, and Google AI Overviews. AEO is Answer Engine Optimization, getting your content surfaced as the direct answer.
AI search tools now answer questions directly without sending users to websites. Analysis across millions of AI-generated responses shows that brands cited in these answers are capturing a disproportionate share of discovery and purchase intent before a user ever visits a website.
Prompt 15: AI Citation Optimizer
Act as a GEO specialist focused on AI search optimization.
TASK: Review this content and tell me exactly what to change
to make it more likely to be cited by ChatGPT, Perplexity,
Claude, and Google AI Overviews.
EVALUATE:
- Direct answer formatting (does it answer the question in the
first 50 words?)
- Factual density (specific stats, named sources, verifiable claims)
- Heading structure (does each H2 match a real search question?)
- E-E-A-T signals (author expertise, citations, original data)
- Content freshness signals (dated references, old stats)
For each weakness: tell me exactly what to rewrite and how.
TARGET KEYWORD: [keyword]
CONTENT: [paste your article]Prompt 16: 90-Day Entity Authority Plan
Create a 90-day entity authority building plan covering on-site signals, content strategy, off-site signals, and entity consistency, to ensure your name and credentials appear consistently across all platforms AI systems crawl.
Act as an AI search visibility strategist.
TASK: Build me a 90-day entity authority plan so that AI search
engines like Claude, Perplexity, and ChatGPT recognize my site
as an authoritative source for [topic].
COVER:
1. On-site signals — schema markup, author byline structure,
content types to implement immediately
2. Content strategy — what types of content earn the most
AI citations
3. Off-site signals — what third-party mentions most influence
AI citation patterns
4. Entity consistency — how to ensure my brand appears
consistently across all platforms AI systems crawl
5. Month-by-month milestones to track progress
SITE/BRAND: [your site name and topic]
CURRENT STATUS: [brief description of where you are now]
Be specific. This is my roadmap, not a general overview.Competitor Analysis Prompts
Prompt 17: Backlink Gap Analysis
Act as a link building strategist.
TASK: I am building backlinks for a [industry] website that
publishes [type of content]. Give me 15 specific types of websites,
publications, directories, communities, and platforms most likely
to link to this content.
For each source include:
- Why they would link
- How to approach them
- What type of content to pitch
- Estimated link quality (high / medium / low)
MY SITE TOPIC: [topic]
MY BEST PERFORMING CONTENT: [list 2-3 URLs or titles]
COMPETITORS WHO OUTRANK ME: [list competitor domains]Prompt 18: SERP Analysis Before Writing
Use this before creating content to understand what is currently ranking and how to build something better. Paste the top five search results directly into the prompt alongside your draft.
Act as an SEO content analyst.
I will share the top 5 search results for [keyword].
Analyze them and identify:
1. Common topics and subtopics covered across all five
2. Content length and structure patterns
3. Types of evidence used (stats, examples, case studies, quotes)
4. Gaps or opportunities none of them address
5. One angle I could take that would make my version
worth clicking over these
KEYWORD: [keyword]
RESULT 1: [paste content or URL]
RESULT 2: [paste content or URL]
RESULT 3: [paste content or URL]
RESULT 4: [paste content or URL]
RESULT 5: [paste content or URL]The Workflow That Ties All of This Together
The real power is not any single prompt. It is running Prompt 2 (search intent analysis) into Prompt 4 (content brief) into Prompt 13 (long-form article) into Prompt 14 (intro rewrite) as a single connected workflow. You can go from keyword to publish-ready draft in under two hours.
The sequence that works:
Step 1: Run Prompt 2 to understand intent before writing anything.
Step 2: Run Prompt 5 to analyze what competitors are doing.
Step 3: Run Prompt 4 to build a proper brief based on what you learned.
Step 4: Run Prompt 13 to write the article using that brief as context.
Step 5: Run Prompt 15 to optimize for AI citations before publishing.
Step 6: Run Prompt 12 to identify internal linking opportunities.
Where ClickRank fits: none of these prompts tell you whether the keyword is worth targeting in the first place, what your current rankings look like, or how competitors are outpacing you in search. That data layer is what ClickRank provides. The prompts handle execution. ClickRank handles the decisions that make execution worth doing.
Are these prompts different from prompts for ChatGPT?
Yes. Claude responds especially well to longer, more detailed prompts, role-playing as specific experts, being told explicitly what not to do, and multi-step instructions given all at once rather than iteratively. Shorter single-sentence prompts that work in other tools underperform with Claude.
Can Claude do keyword research from scratch?
No. Claude cannot pull live search volumes or current SERP data. Use these prompts after you have exported data from a keyword tool. Claude's job is to analyze and organize that data, not generate it.
How do I get consistent output from these prompts?
Save your best-performing versions in a Claude Project. For agency work or multi-client SEO, create a Project per client, upload their brand guidelines, sitemap, and keyword targets once, and every prompt in that Project inherits the context automatically.
Should I verify Claude's output before publishing?
Always. Claude can confidently state things that are technically wrong, especially in data analysis. Treat every output like a first draft from a skilled but new team member. The structure will be strong. The facts need checking.
What Claude plan do I need for these prompts?
Most prompts in this guide work on the free tier for occasional use. For weekly SEO workflows involving large documents, the Pro plan at $20 per month is the right level. It handles 10 to 15 client articles per month alongside technical tasks without hitting limits.
