{"id":196421,"date":"2026-04-15T09:33:27","date_gmt":"2026-04-15T14:33:27","guid":{"rendered":"https:\/\/ahrefs.com\/blog\/?p=196421"},"modified":"2026-04-16T02:54:03","modified_gmt":"2026-04-16T07:54:03","slug":"why-chatgpt-cites-pages","status":"publish","type":"post","link":"https:\/\/ahrefs.com\/blog\/why-chatgpt-cites-pages\/","title":{"rendered":"Why ChatGPT Cites One Page Over Another (Study of 1.4M Prompts)"},"content":{"rendered":"
We\u2019ve all got used to the little numbered blue links in ChatGPT\u2019s responses. They\u2019re the citations that back up ChatGPT\u2019s responses with external information.<\/div>\n

But, although ChatGPT crawls dozens of pages to answer a single query, according to our research, it only ends up citing ~50% of them.<\/p>\n

\"Pie<\/p>\n

Why does one page get the credit while another, which the AI clearly retrieved, gets nothing?<\/p>\n

According to studies<\/a> by AI expert Dan Petrovic<\/a>, when ChatGPT retrieves results, each one comes back with the page title, a brief snippet or summary, the URL, and an ID number.<\/p>\n

\"Text<\/p>\n

ChatGPT uses this data to decide which pages are worth opening and eventually citing in its response.<\/p>\n

That means there\u2019s a gatekeeping layer before<\/em> ChatGPT opens and reads any of your actual page content. The title, snippet, and URL are doing the heavy lifting in that initial decision.<\/p>\n

So we wanted to know: what actually influences that decision?<\/strong> Does higher semantic similarity between a page\u2019s retrieval data and the user query increase citation likelihood? Which fields matter most? Do human-readable URLs outperform opaque ones?<\/p>\n

To find out, we analyzed 1.4 million ChatGPT 5.2 prompts from February 2025 (desktop) with the help of Ahrefs data scientist Xibeijia Guan<\/a>.<\/p>\n

But before we get into the findings, you need to understand how ChatGPT actually gathers its sources\u2014because not all URLs enter the system the same way.<\/p>\n

<\/a>