Webinar: The Link Building Graveyard: What to Bury & What Still Wins

Reserve Your Seat
  1. SEJ
  2.  ⋅ 
  3. AI Search

ChatGPT’s Most-Cited Sites Tilt Toward Primary Sources, Data Shows

  • Government health, research, and regulator sites climbed ChatGPT's U.S. citation ranks while explainers fell.
  • The charted sites moved in the same direction on prompts that appeared in both months.
  • Pages that summarize others' research may lose ChatGPT citations to the original source.

Government health, research, and regulator sites climbed ChatGPT's U.S. citation ranks while explainers fell. Most free, logged-out answers skipped search.

ChatGPT’s Most-Cited Sites Tilt Toward Primary Sources, Data Shows

According to Ahrefs data, PubMed moved up from 64th to fourth place among the domains ChatGPT most frequently cited in U.S. answers from March to September, while WebMD dropped from 18th to 79th.

Most of the climbers in the report’s chart were government health and research sites, regulators, and hands-on product testers. Health and finance explainers lost ground, and Britannica, Investopedia, and the New York Post fell out of the top 200. Ahrefs writes that ChatGPT “seems to be prioritizing original sources over sites that summarize them.”

The report also looked at how frequently ChatGPT searches happen. Out of 66.6 million free, logged-out responses from August and September, 14.4% involved running a web search.

Which Sites Climbed And Which Fell

Ahrefs looked at the 200 domains most frequently cited by ChatGPT in U.S. answers, based on 2.4 million answers in March and 3.8 million in September.

The company describes WebMD and Verywell Health as health sites that summarize research, and it groups Healthline and Investopedia with them as health and finance explainers. Its longer list of sites cited more adds regulators such as the CFPB and FTC, plus support pages from Google and Microsoft.

The company’s tracking system adds prompts over time, meaning that a change in rank might be due to new questions rather than a change in what ChatGPT cites. To check, they compared around 25,000 prompts that appeared among their 150,000 highest-volume prompts in both March and September. They noted that every site on the chart moved in the same direction for those prompts.

Reddit And Wikipedia Appear In Fewer Answers

Reddit was cited in 19.1% of U.S. ChatGPT answers in March and 2.3% in September, in Ahrefs’ data. Wikipedia went from 14.1% to 2.9% over the same period. The company says both are still among ChatGPT’s most-cited domains.

I discussed a Reddit drop in August, noting that Promptwatch showed reddit.com’s share of ChatGPT Search citations decreased from an average of 3.83% in late July and early August to 0.52% in mid-August. Ahrefs measures the share of answers citing Reddit, which is a different way of looking at it.

In Ahrefs’ separate worldwide sample, reddit.com was still the ninth domain ChatGPT searched most often with site: searches, which limit a search to one website.

Most Free Answers Didn’t Search The Web

Out of the 66.6 million free, logged-out GPT-5.6 Luna responses in Ahrefs’ fan-out sample, 14.4% included a web search. This data covers August and September across all countries. On average, responses that involved searching had around 1.6 fan-out queries, which are background searches ChatGPT performs before giving an answer.

OpenAI’s help page mentions that ChatGPT might search the web on its own if a question would benefit from current information.

Of the 15 million fan-outs, 6.8% were site: searches, spread across 116,000 domains. Among these site: searches targeting the roughly 160 most-searched domains, 75.9% went to brand sites, which the report defines as company websites like apple.com or ikea.com.

The report highlights that other studies have observed much higher fan-outs per prompt. For example, Nectiv found an average of 7.61 fan-outs per prompt from about 4,000 prompts processed through GPT-5.6 Sol via the API. Compared to this, the figure of 1.6 only considers responses that involve searching. Since the two samples differ in models, prompts, and collection methods, their numbers aren’t directly comparable. The company explains that its broad prompts are drawn from search demand across every topic and country, which accounts for part of the gap. They also note that using the service for free and without logging in results in fewer searches compared to paid or reasoning models.

Why This Matters

Health and finance explainer sites like Healthline, WebMD, and Investopedia saw a decline in citation rank in this data. Meanwhile, PubMed, the FDA, and testing sites like RTINGS increased their rank. Ahrefs suggests focusing on pages that only you can create, such as original research, hands-on testing, and your own product information.

Ahrefs’ citation ranks don’t measure traffic. Checking how often ChatGPT cites your own pages for your core topics, and which sources it cites instead, will show whether this applies to you.

Looking Ahead

According to Ahrefs, their fan-out data comes from a snapshot of a single model taken between August and September. They also note that these patterns could change in future updates. If explainer pages are your main way to get visibility, it’s a good idea to check which sources ChatGPT cites for your important topics whenever OpenAI updates its model.

Featured Image: Tetiana Yurchenko/Shutterstock. OpenAI wordmark: OpenAI.

Category News AI Search
SEJ STAFF Matt G. Southern Senior News Writer at Search Engine Journal

Follow on YouTube. Matt G. Southern is the Senior News Writer at Search Engine Journal, where he’s covered Google, SEO, ...