The original title is "How AI answer engines decide which sources to cite"

Short summary

AI answer engines like ChatGPT, Perplexity, and Google AI Overviews use retrieval-augmented generation pipelines that pull 10-30 candidate pages but cite only 3-5, making citation optimization a two-stage battle: getting retrieved, then surviving reranking. ChatGPT relies heavily on Bing's index (87%+ of citations match Bing top results), Perplexity favors authority domains with hybrid retrieval and freshness signals, and Google AI Overviews cite extractable passages where structure and E-E-A-T can beat raw ranking position. The core levers across all three are indexability, extractable structure, schema markup, off-page trust, and content freshness.

• •ChatGPT citations map ~87% to Bing top organic results — Bing SEO is the primary lever
• •Perplexity uses hybrid retrieval with authority domain lists; freshness and relevance dominate reranking
• •Google AI Overviews cite passages not pages; E-E-A-T and structure can beat top-10 ranking position

Generated with AI, which can make mistakes.