ai-seo

name	ai-seo
description	When the user wants to optimize content for AI search engines, get cited by LLMs, or appear in AI-generated answers. Also use when the user mentions 'AI SEO,' 'AEO,' 'GEO,' 'LLMO,' 'answer engine optimization,' or 'AI citations.'
allowed-tools	Read, Write, WebSearch, WebFetch, AskUserQuestion
model	sonnet

AI SEO — Answer Engine & LLM Optimization

Expert knowledge for getting your brand, product, and content cited by AI systems including ChatGPT, Perplexity, Claude, Gemini, and AI-powered search.

What AI SEO Is (and Isn't)

AI SEO (also called AEO — Answer Engine Optimization, GEO — Generative Engine Optimization, or LLMO — Large Language Model Optimization) is the practice of making your content the authoritative source LLMs cite when answering questions relevant to your domain.

Traditional SEO → rank in blue-link results. AI SEO → be the source AI systems quote, summarize, or recommend.

These goals overlap significantly but differ in emphasis:

• AI SEO prioritizes answer-shaped content, entity clarity, and corroborating citations
• Traditional SEO prioritizes keyword density, backlink profile, and click-through signals

How LLMs Source Information

LLMs are trained on large web corpora (Common Crawl, Wikipedia, news, forums, Reddit, docs). They also increasingly use Retrieval-Augmented Generation (RAG) to pull live web content. Understanding both paths is key:

Training data path: Your content must exist, be crawlable, and be repeatedly cited across authoritative sources before the LLM training cutoff.

RAG/live retrieval path: Your content must be indexed, fast-loading, and structured for snippet extraction. Perplexity and Bing Copilot lean heavily on this.

Core Strategy: The 7 Pillars of AI SEO

1. Entity Disambiguation

LLMs build knowledge graphs. Your brand/product must be a clearly defined entity.

Requirements:

• Consistent NAP (Name, Address, Phone) across all web properties
• Wikipedia page or Wikidata entry (if scale justifies it)
• Google Knowledge Panel claimed and verified
• Crunchbase, LinkedIn Company, G2 profiles all matching
• About page that clearly states: who you are, what you do, who you serve, where you're located

Entity signals the AI reads:

• Brand name mentioned alongside consistent descriptors across many sources
• Your domain cited by other authoritative domains
• Your brand mentioned in the same context as established competitors

2. Answer-Shaped Content

LLMs prefer content that directly answers questions. Write content as if you're answering a specific question — because you are.

Format rules:

• Lead with the direct answer in 1-2 sentences
• Follow with the explanation, evidence, and context
• Use clear heading questions ("What is X?" "How does Y work?")
• Include definition blocks for key terms
• Use numbered steps for processes
• Use comparison tables for alternatives

The "Answer Block" formula:

[Question as H2]
[Direct 1-sentence answer]
[2-3 sentence elaboration]
[Evidence or example]
[Link to deeper content]

3. Prompt-Matched Content

LLMs get asked questions in natural language. Your content must match the phrasing of those questions — not just the keywords.

Research approach:

• Use "People also ask" in Google to find natural-language questions
• Use Reddit, Quora, and forums to find how your audience phrases problems
• Use Answer The Public for question mapping
• Monitor what questions Perplexity and ChatGPT are answering in your niche

Content coverage checklist:

• "What is [your category]?"
• "How does [your product/approach] work?"
• "Best [your category] for [specific use case]"
• "[Your product] vs [competitor]"
• "How to [achieve outcome your product enables]"

4. Corroborating Citations

LLMs trust information that appears across multiple authoritative sources. A claim you make alone is weak. A claim corroborated by industry reports, news, and other sites is strong.

Build your citation network:

• Publish original research and data (studies, surveys, benchmark reports)
• Get your data cited in industry publications
• Contribute guest articles to authoritative domains in your space
• Earn mentions in analyst reports (Gartner, Forrester, G2 reviews)
• Build case studies that other sites reference

The corroboration flywheel:

1. Publish original data
2. Pitch it to industry journalists
3. They cite it → you earn backlinks
4. LLMs see your data cited in multiple sources → trust increases

5. Structured Data (Schema Markup)

Schema markup helps AI systems understand your content's structure and context.

Priority schemas for AI SEO:

• Organization — entity definition
• FAQPage — explicitly marks Q&A content for AI extraction
• HowTo — marks step-by-step processes
• Article / BlogPosting — marks editorial content with author and date
• Product — if you have product pages
• Review / AggregateRating — social proof signals

6. llms.txt

Emerging standard (analogous to robots.txt) that tells AI crawlers what content is most important to index. Create /llms.txt at your domain root with:

• Brief description of your site and its purpose
• Links to your most important pages for AI context
• Links to documentation, FAQs, or knowledge base

Format:

# [Brand Name]
> [One-sentence description of what you do and who you serve]

## Important Pages
- [Page Title]: [URL] — [One-sentence description]
- ...

## Documentation
- [Doc section]: [URL]

7. AI Directory and Platform Presence

AI-powered discovery tools maintain their own indexes. Be present in:

• Perplexity: Ensure your site is crawlable; Perplexity cites heavily from indexed sources
• ChatGPT with Browse: Bing-indexed content surfaces in browsing mode
• Product Hunt, G2, Capterra: LLMs frequently cite these for product comparisons
• GitHub: If you have open-source components, README + docs are indexed
• YouTube: Transcripts are indexed; create educational content that answers target questions

Content Optimization Checklist

For each page targeting AI citation:

Structure:

• H1 is a clear, question-shaped title or a definitive statement
• First paragraph answers the core question directly
• Subheadings are phrased as questions or clear topic signals
• Answer blocks used for each key question
• Comparison table present where alternatives exist

Entity signals:

• Brand name used consistently with same descriptor phrase
• Author byline present with credentials
• Publication date and last-updated date present
• Internal links to related authoritative pages
• External links to corroborating sources (not competitors)

Technical:

• FAQPage schema implemented
• Page loads in under 2 seconds (Core Web Vitals)
• Content accessible without JavaScript
• llms.txt references this page

Measuring AI SEO Success

Traditional rank tracking doesn't capture AI citation. Use:

Direct testing:

• Prompt ChatGPT, Perplexity, Gemini, Claude with your target questions
• Track whether your brand/domain/content appears in answers
• Screenshot and date-stamp for trend tracking

Indirect signals:

• Branded search volume growth (people looking for you by name after discovering via AI)
• Traffic from Perplexity (appears in referral analytics)
• Increase in zero-click branded traffic

Tools:

• Perplexity Pages analytics (if you publish there)
• BrandMentions / Mention for tracking AI output mentions
• Manual weekly prompting and logging

Common Rationalizations

Rationalization	Reality
"We already do SEO, so AI SEO is covered"	Traditional SEO optimizes for crawlers. AI SEO requires answer-shaped content and entity disambiguation — different skill set.
"LLMs don't crawl the web, so this doesn't matter"	Modern AI tools (Perplexity, Copilot, ChatGPT Browse) use RAG with live retrieval. Training data is only part of the picture.
"We'll wait until AI SEO standards stabilize"	Early movers in AI citation earn training data advantages that persist. Waiting means you start behind.
"Our content is good, it'll naturally get cited"	Structured answer blocks and schema markup significantly increase citation probability. Quality alone is not enough.
"We can't measure it, so it's not worth the effort"	Direct prompting tests give measurable results. Branded search growth is a reliable proxy signal.

Verification

• Entity disambiguation complete (Knowledge Panel, Crunchbase, LinkedIn all consistent)
• llms.txt created and published at domain root
• Priority pages rewritten with answer-block format
• FAQPage schema implemented on key pages
• Original research or data published and pitched for citations
• AI citation baseline established (manual prompting test with screenshots)
• Weekly prompting test scheduled to track citation changes