Will AI Cite Your Content?

Foglift evaluates 8 weighted dimensions of how AI engines extract and cite content. Structured Data Richness (20%) is the heaviest weight because JSON-LD is the primary signal AI engines parse from a page. Heading Clarity (15%), FAQ Quality (15%), and Content Depth (15%) follow because they shape what AI engines can actually quote. Entity Identity (10%), Citation Formatting (10%), and AI Crawler Access (10%) round out the structural layer. Topical Authority (5%) is the smallest weight because cross-page authority signals tend to be measured better by GEO. Each dimension scores 0-100 and the overall score is a weighted average.

Foglift uses standard letter brackets: A is 90+, B is 80-89, C is 70-79, D is 50-69, and F is below 50. The thresholds match the public Foglift score-utils library, so the dashboard, API, and CLI always agree on the grade. A B-grade page is structurally cite-able by ChatGPT and Perplexity for most topics; an A-grade page is the upper band where AI engines reliably extract structured answers without rewriting your content.

The FAQ Quality dimension awards points across four signals: FAQPage JSON-LD schema is worth 40 points; three or more Question entities inside that schema add 20; a visible H2/H3 heading containing 'FAQ', 'frequently asked questions', or 'common questions' adds 25; and three or more <details> or accordion-style elements add 15. Hitting all four boosters is the only way to score 100 on this dimension. This methodology mirrors what AI engines actually extract — structured Q&A pairs are among the most-cited content blocks because the question literally matches the user prompt.

The public /api/v1/aeo-score endpoint allows 20 requests per hour per IP, no authentication required. It runs the same 8-dimension analysis the paid scan uses and returns the full dimensions array, recommendations, and grade. If you need higher throughput, the authenticated Website Audit on a Foglift account is unlimited — audits do not consume token budget on any tier (Free 200 tokens, Launch 4,000, Growth 11,500, Enterprise 27,000); tokens are only spent on AI Visibility Checks against the 5 LLM engines.

Every time you ship a meaningful content change. Each scan runs live (8-15 seconds end-to-end) and is persisted to the geo_results table, so trends are queryable via /api/v1/scan/history. Frequency matters beyond just measurement — content updated within the last 30 days receives roughly 3.2x more AI citations than stale content (per Foglift's longitudinal AI-citation study tracked in our content-brand-brief). Stale technical content is one of the strongest negative signals AI engines use when ranking sources to cite.

AEO measures whether your content is structured for AI extraction — schema, headings, FAQs, tables, depth. GEO measures whether your brand is actually surfaced in AI answers across ChatGPT, Claude, Perplexity, Gemini, and Copilot. They are complementary: a page can have a perfect AEO score and still not appear in AI responses (low authority); conversely a brand can have strong GEO mentions on weak pages (high authority compensating for poor structure). The Foglift scan returns both because optimization without monitoring is blind, and monitoring without optimization is unactionable.

No. AEO is about content structure — schema markup, heading hierarchy, FAQ patterns, citation formatting, depth — not prose quality. An AI-generated 5,000-word article with no JSON-LD, no FAQ section, and no heading hierarchy will score worse on AEO than a 1,200-word post with FAQPage schema, an H2-organized outline, and a comparison table. AEO is closer to a developer/structural-editor concern than a copywriting concern. Foglift's recommendations engine pinpoints exactly which structural fixes will move each dimension's score (e.g. 'add FAQPage schema with 5 Q&A pairs to lift FAQ Quality from 25 to 100').

AEO is necessary but not sufficient. AEO is the structural readiness layer — without it, AI engines either cannot extract your content cleanly or have to paraphrase, which costs you attribution. But citations also depend on authority signals AEO doesn't measure: backlinks, brand mentions, Reddit/community presence (which carries roughly a 3.9x citation multiplier), and topical depth across multiple pages. The Foglift flywheel is built around this: improving AEO addresses the structural side; tracking GEO and AI Visibility tells you whether the structural improvements are translating to actual mentions, so you know which pages need authority work next.