{ "studySlug": "citation-sources", "title": "Where AI Gets Its Answers", "description": "Source preference, category concentration, and domain hierarchy across AI citations. Answer pages, reference facts, and live trackers drawn from this study.", "lastUpdated": "2026-04-17", "sourceStudyTitle": "Where AI Gets Its Answers", "sourceStudyUrl": "/trakkr-research/citation-sources", "stats": [ { "label": "Answers", "value": "12" }, { "label": "Facts", "value": "8" }, { "label": "Trackers", "value": "2" } ], "featuredAnswer": { "kind": "answer", "slug": "which-sources-dominate-ai-citations", "title": "Which sources dominate AI citations?", "description": "The short answer is that a small set of domains dominates the head while the long tail still matters.", "path": "/trakkr-research/citation-sources/answers/which-sources-dominate-ai-citations" }, "sections": [ { "title": "Answer Pages", "description": "Narrow questions answered directly from the study.", "items": [ { "kind": "answer", "slug": "which-sources-dominate-ai-citations", "title": "Which sources dominate AI citations?", "description": "The short answer is that a small set of domains dominates the head while the long tail still matters.", "path": "/trakkr-research/citation-sources/answers/which-sources-dominate-ai-citations" }, { "kind": "answer", "slug": "how-concentrated-is-the-ai-citation-landscape", "title": "How concentrated is the AI citation landscape?", "description": "It is concentrated at the top and fragmented underneath. One domain can own more than 6% of all citations, yet the index still spans 208,567 unique domains and 744,579 cited URLs.", "path": "/trakkr-research/citation-sources/answers/how-concentrated-is-the-ai-citation-landscape" }, { "kind": "answer", "slug": "does-ai-rely-more-on-reference-sites-than-the-open-web", "title": "Does AI rely more on reference sites than the open web?", "description": "Mostly yes at the named-category level. Reference sources hold 6.75% of the current corpus, which makes them the clearest structured source family in the index, even though the largest share still sits in the long tail", "path": "/trakkr-research/citation-sources/answers/does-ai-rely-more-on-reference-sites-than-the-open-web" }, { "kind": "answer", "slug": "does-social-content-matter-for-ai-citations", "title": "Does social content matter for AI citations?", "description": "Yes. Social sources account for 3.82% of citations in the current index, led by YouTube, Reddit, and LinkedIn, which means community and creator content is not just discovery fluff - it is part of the citation layer.", "path": "/trakkr-research/citation-sources/answers/does-social-content-matter-for-ai-citations" }, { "kind": "answer", "slug": "do-review-sites-still-matter-to-ai-models", "title": "Do review sites still matter to AI models?", "description": "Yes, especially for commercial and comparative prompts. Reviews make up only 0.89% of the full citation index, but they are highly relevant on buying-intent queries where models need ranking, pros and cons, and category", "path": "/trakkr-research/citation-sources/answers/do-review-sites-still-matter-to-ai-models" }, { "kind": "answer", "slug": "how-often-do-docs-get-cited-by-ai", "title": "How often do docs get cited by AI?", "description": "Docs are a small but meaningful slice of the overall citation graph.", "path": "/trakkr-research/citation-sources/answers/how-often-do-docs-get-cited-by-ai" }, { "kind": "answer", "slug": "why-does-wikipedia-still-win-so-often", "title": "Why does Wikipedia still win so often?", "description": "Because it packages high-coverage facts in a format models already trust. In the current index, en.wikipedia.org alone captures 6.43% of all citations, which is a scale advantage few branded domains can match.", "path": "/trakkr-research/citation-sources/answers/why-does-wikipedia-still-win-so-often" }, { "kind": "answer", "slug": "how-big-is-the-competitive-set-for-ai-citations", "title": "How big is the competitive set for AI citations?", "description": "It is much bigger than most teams assume. Trakkr currently sees 208,567 unique domains competing for citations across 1,038 tracked brands, which means your real AI competition includes editors, creators, communities,", "path": "/trakkr-research/citation-sources/answers/how-big-is-the-competitive-set-for-ai-citations" }, { "kind": "answer", "slug": "what-source-categories-show-up-most-in-ai-answers", "title": "What source categories show up most in AI answers?", "description": "Reference and social are the clearest named categories, while the long tail still absorbs most citations.", "path": "/trakkr-research/citation-sources/answers/what-source-categories-show-up-most-in-ai-answers" }, { "kind": "answer", "slug": "do-brands-need-third-party-sources-to-win-ai-citations", "title": "Do brands need third-party sources to win AI citations?", "description": "Usually yes. The current citation graph is dominated by domains outside any single brand’s control, which means owned content alone rarely explains visibility.", "path": "/trakkr-research/citation-sources/answers/do-brands-need-third-party-sources-to-win-ai-citations" }, { "kind": "answer", "slug": "what-should-a-brand-prioritize-after-reading-the-source-data", "title": "What should a brand prioritize after reading the source data?", "description": "A brand should prioritize the source types that actually shape answers in its category.", "path": "/trakkr-research/citation-sources/answers/what-should-a-brand-prioritize-after-reading-the-source-data" }, { "kind": "answer", "slug": "does-raw-domain-count-matter-more-than-repeat-cited-authority", "title": "Does raw domain count matter more than repeat-cited authority?", "description": "No. The citation graph is broad, but models still repeat a relatively small set of trusted URLs and domains.", "path": "/trakkr-research/citation-sources/answers/does-raw-domain-count-matter-more-than-repeat-cited-authority" } ] }, { "title": "Reference Facts", "description": "Short, quotable claims with metrics and methodology context.", "items": [ { "kind": "fact", "slug": "wikipedia-leads-the-current-citation-index", "title": "Wikipedia leads the current citation index", "description": "The current live citation graph still has a strong head domain.", "path": "/trakkr-research/citation-sources/facts/wikipedia-leads-the-current-citation-index" }, { "kind": "fact", "slug": "the-citation-graph-spans-more-than-two-hundred-thousand-domains", "title": "The citation graph spans more than 200,000 domains", "description": "The long tail is enormous even when the head is concentrated.", "path": "/trakkr-research/citation-sources/facts/the-citation-graph-spans-more-than-two-hundred-thousand-domains" }, { "kind": "fact", "slug": "reference-sites-are-the-clearest-named-source-family", "title": "Reference sites are the clearest named source family", "description": "Reference is the strongest named source class even though the long tail is larger in aggregate.", "path": "/trakkr-research/citation-sources/facts/reference-sites-are-the-clearest-named-source-family" }, { "kind": "fact", "slug": "social-sources-are-material-not-noise", "title": "Social sources are material, not noise", "description": "YouTube, Reddit, and LinkedIn keep appearing in the citation layer.", "path": "/trakkr-research/citation-sources/facts/social-sources-are-material-not-noise" }, { "kind": "fact", "slug": "review-sites-are-small-in-share-but-high-in-intent", "title": "Review sites are small in share but high in intent", "description": "Review domains are a smaller slice of raw volume than many teams expect.", "path": "/trakkr-research/citation-sources/facts/review-sites-are-small-in-share-but-high-in-intent" }, { "kind": "fact", "slug": "docs-are-a-niche-but-important-citation-layer", "title": "Docs are a niche but important citation layer", "description": "Docs are not broad recommendation winners, but they show up when the prompt becomes technical.", "path": "/trakkr-research/citation-sources/facts/docs-are-a-niche-but-important-citation-layer" }, { "kind": "fact", "slug": "the-long-tail-still-holds-most-citations", "title": "The long tail still holds most citations", "description": "Most citation volume sits outside the obvious named categories.", "path": "/trakkr-research/citation-sources/facts/the-long-tail-still-holds-most-citations" }, { "kind": "fact", "slug": "the-current-index-tracks-more-than-seven-hundred-thousand-cited-urls", "title": "The current index tracks more than 700,000 cited URLs", "description": "AI systems do not just cite domains. They return to a large but finite set of pages.", "path": "/trakkr-research/citation-sources/facts/the-current-index-tracks-more-than-seven-hundred-thousand-cited-urls" } ] }, { "title": "Trackers", "description": "Live benchmark views built from the study’s most reusable dimensions.", "items": [ { "kind": "tracker", "slug": "source-category-share-tracker", "title": "AI citation share by source category", "description": "Current category mix in the live Trakkr citation index.", "path": "/trakkr-research/citation-sources/trackers/source-category-share-tracker" }, { "kind": "tracker", "slug": "top-cited-domain-tracker", "title": "Top cited domains in the live index", "description": "Current head domains in Trakkr’s live public citation graph.", "path": "/trakkr-research/citation-sources/trackers/top-cited-domain-tracker" } ] } ] }