Optimize content for AI search citations and visibility

This skill optimizes content to be discoverable, extractable, and citable by AI systems - Google AI Overviews, ChatGPT, Perplexity, Claude, Gemini, and Copilot - so a brand gets cited as a source in AI-generated answers, not just ranked in traditional search. Before optimizing, it checks for an existing product-marketing context file and gathers current AI visibility, content and domain profile, goals, and competitive landscape.

Use it when optimizing content to be cited by LLMs and AI search systems, when the user asks about AI SEO, AEO, GEO, LLM visibility, or AI citations, or when traditional SEO alone doesn't cover AI-specific discoverability. It frames the key difference plainly: traditional SEO gets a page ranked, AI SEO gets it cited - a well-structured page can be cited even ranking on page 2 or 3, since AI systems select sources on content quality, structure, and relevance, not just rank. Cited stats: AI Overviews appear in roughly 45% of Google searches and reduce clicks to websites by up to 58%; brands are 6.5x more likely to be cited via third-party sources than their own domain; optimized content gets cited 3x more often than non-optimized; statistics and citations boost visibility 40%+ across queries.

Each AI platform selects sources differently: Google AI Overviews summarizes top-ranking pages with a strong correlation to traditional rankings; ChatGPT (with search) draws from a wider range than just top-ranked pages; Perplexity always cites sources with links and favors authoritative, recent, well-structured content; Gemini pulls from Google's index and Knowledge Graph; Copilot draws on the Bing index and authoritative sources; Claude uses Brave Search (when enabled) plus its training data. An AI visibility audit runs first: test 10-20 key queries ("What is X?", "Best X for Y", "X vs competitor", how-to, pricing phrasings) across Google AI Overviews, ChatGPT, and Perplexity to see who gets cited; analyze citation patterns against competitors on content structure, authority signals, freshness, schema markup, and third-party presence; run a content-extractability checklist (clear first-paragraph definition, self-contained answer blocks, sourced statistics, comparison tables, FAQ section, schema markup, expert attribution, recency within 6 months, query-matched headings); and verify AI bot access in robots.txt, since blocking a platform's crawler - GPTBot/ChatGPT-User (OpenAI), PerplexityBot, ClaudeBot/anthropic-ai (Anthropic), Google-Extended (Google), Bingbot (Microsoft Copilot) - blocks that platform's citations, while training-only crawlers like CCBot can be blocked separately. Optimization then follows three pillars. Structure: lead every section with a direct answer, keep key answer passages to 40-60 words for snippet extraction, use query-phrased H2/H3 headings, prefer tables and numbered lists over prose, and build definition, step-by-step, comparison, pros/cons, FAQ, and statistic content blocks. Authority: the skill cites Princeton's GEO research (KDD 2024, studied on Perplexity.ai), which ranked optimization methods by visibility boost - citing sources +40%, adding statistics +37%, adding quotations +30%, authoritative tone +25%, improved clarity +20%, technical terms +18%, unique vocabulary +15%, fluency optimization +15-30%, while keyword stuffing actively hurts visibility by -10%; fluency plus statistics is the best combination, and low-ranking sites gain even more, up to 115% with citations. Authority also means original, dated statistics over aggregated ones, named expert attribution, visible "last updated" freshness signals refreshed quarterly for competitive topics, and E-E-A-T alignment (first-hand experience, specific detail, transparent sourcing). Presence: third-party sources often outweigh a brand's own site - Wikipedia accounts for about 7.8% of ChatGPT citations, Reddit about 1.8%, alongside industry publications, review sites, YouTube, and Quora - so keeping a Wikipedia page accurate, participating in Reddit, and maintaining review-platform profiles all matter. Schema markup (Article/BlogPosting, HowTo, FAQPage, Product, ItemList, Review/AggregateRating, Organization) is linked to 30-40% higher AI visibility. By content type, citation share favors comparison articles (33%), followed by definitive guides (15%), original research and data (12%), best-of/listicles and product pages (10% each), opinion/analysis (10%), and how-to guides (8%); generic unstructured blog posts, thin product pages with marketing fluff, gated content, PDF-only content, and anything undated without author attribution underperform for AI citation. Monitoring tracks AI Overview presence, brand citation rate, share of AI voice against competitors, citation sentiment (how AI describes the brand), and source attribution (which pages get cited), run monthly at minimum, either via tools (Otterly AI, Peec AI, ZipTie, LLMrefs) or a manual DIY check across the top 20 queries logged month over month. Guidance is also given per content type: SaaS product pages aim to get cited in "what is [category]" and "best [category]" queries via a clear first-paragraph description, feature-comparison tables, specific metrics instead of vague claims, visible pricing, and an FAQ section; blog content aims for authoritative-source citation via one clear target query per post, original data or expert quotes, a visible last-updated date, and an author bio; comparison and alternative pages aim to get cited in "X vs Y" and "best X alternatives" queries via structured, fair, and balanced comparison tables with updated data (AI penalizes obviously biased comparisons); documentation and help content aims for "how to X with your product" citations via numbered step-by-step format, code examples, HowTo schema, and clear prerequisites. Named mistakes to avoid beyond the ones above: writing for AI instead of humans (content that reads as gamed won't convert even if cited), treating AI SEO as wholly separate from traditional SEO rather than built on top of it, and simply forgetting to monitor visibility at all.

For implementation it points to semrush and ahrefs for AI Overview tracking and content-gap analysis, gsc (Search Console) for query performance, and ga4 for referral traffic from AI sources, plus dedicated AI-visibility monitoring tools (Otterly AI, Peec AI, ZipTie, LLMrefs). Related skills named: seo-audit for traditional technical/on-page audits, schema-markup for implementing structured data, content-strategy for planning what to create, competitor-alternatives for citation-worthy comparison pages, programmatic-seo for building pages at scale, and copywriting for content that reads naturally to humans while staying AI-extractable.

Content, marketing, and SEO teams who want their pages cited inside AI-generated answers, not just ranked - and who are willing to audit current AI visibility, restructure content for extractability, build genuine authority signals, and show up in the third-party sources AI systems actually pull from, rather than treating AI SEO as a bolt-on to traditional SEO.