# OpenAIChatGPT Search Now Uses the site:operator at Scale On August 8, 2026, ChatGPT Search changed how it searches the web overnight. Fanout queries using the site:operator (searches scoped to a specific domain) jumped from 0.37% to 16.8% of all fanout queries, a roughly 46x increase in share within a single day. At the same time, the average number of searches per response nearly doubled: ChatGPT now runs more, and more targeted, background searches per answer. This report tracks the shift day by day. Klaas Foppen·August 10, 2026 ## OpenAIChatGPT Is Now Searching for site:domain.com Directly The percentage of all ChatGPT Search fanout queries that contain the site:operator, per day. The share hovered between 0.3% and 0.5% for weeks, dipped briefly to 0.15% on August 3 to 5 (consistent with a staged rollout or pre-launch experiment), then jumped to 16-17% on August 8. % of fanouts ### What this means for you For weeks, site:-scoped searches were a rounding error in ChatGPT's search behavior: about 1 in every 280 fanout queries. Since August 8, roughly 1 in 6 fanout queries is scoped to a specific domain. ChatGPT is no longer just searching the open web and seeing what comes back; it is deliberately going to specific sites to pull information from them. The shift happened within a single day, which points to a model or system-prompt rollout on OpenAI's side rather than a gradual behavior change. Platform-controlled search behavior can change overnight, so continuous monitoring matters more than one-off audits. #### How to act on it - Treat your own domain as a retrieval surface. When ChatGPT runs "site:yourdomain.com [topic]", it is searching your site directly. Thin category pages, broken internal search, and unindexed content now cost you answers on top of rankings. - Make sure your most important pages are crawlable and indexed. A site:-scoped query can only surface what search engines have indexed from your domain, so gaps in coverage translate directly into gaps in AI answers. - Watch for behavior shifts like this one before rewriting content strategy. This change happened platform-wide within a single day. It is a ChatGPT rollout, not something individual sites did. ## OpenAIAverage ChatGPT Query Fanouts Per Response The average number of fanout queries per ChatGPT Search response, per day. On August 8 the average jumped from ~1.08 to ~1.83, the same day site: usage surged. That timing suggests site:-scoped searches are additional searches layered on top of the generic ones, not a replacement for them: if ChatGPT had swapped generic queries for site: queries, the average would have stayed flat. Queries ### What this means for you ChatGPT Search held steady at about 1.08 fanout queries per response for weeks, then jumped to about 1.83 on August 8 and has stayed there. The fact that this happened in lockstep with the site: surge is the key signal: the new domain-scoped searches come on top of the generic web searches ChatGPT was already running, they do not replace them. For visibility, that is good news: your existing chances of appearing via generic searches are intact, and site:-scoped searches are a new, additional retrieval path into the answer. ChatGPT now searches both wider (more queries per answer) and deeper (queries scoped to specific domains) than it did before. #### How to act on it - Cover more angles per topic. At ~1.8 queries per response instead of ~1.1, each answer now draws on more distinct retrievals: comparisons, pricing, alternatives, and domain-scoped lookups are all separate chances to be found. - Re-run fanout analyses for your tracked prompts. The queries ChatGPT generates from a prompt changed materially on August 8; keyword mappings built before the shift are stale. ## How we collect this data We collect millions of prompt responses, citations, and click data from the actual user interfaces of major AI platforms: over 26 billion data points and growing. This gives us one of the largest datasets on how AI search engines cite sources and recommend brands. ### Real UI monitoring Data straight from the interfaces of ChatGPT, Gemini, Perplexity, Claude, AI Overviews, and more. ### 26B+ data points Over 26 billion analyzed citations, prompts, and responses, one of the largest AI search datasets available. ### Continuously updated Refreshed constantly so the trends you see reflect the latest behavior of AI search engines. ### Aggregated & public Published freely for the GEO community, based on aggregated, non-identifiable trends.