ChatGPT Cites Primary Sources: What It Means for Your Visibility

PubMed climbed from 64th to 4th place among the sites most cited by ChatGPT in just six months. WebMD, which summarizes the same research, slipped from 18th to 79th. If your company publishes content that explains what others have already said, this shift affects you directly.

Does this shift really affect my business too, or only big publishers?

It affects anyone who publishes content online to get found: e-commerce stores, professional firms, manufacturers with a blog, agencies. If your articles explain topics others have already covered better, ChatGPT will ignore them.

The shift documented by Ahrefs and reported by Search Engine Journal isn't just about big media. It affects every company that uses content to get found online.

ChatGPT is ceasing to cite sites that summarize information already available elsewhere. Instead, it rewards those who produce original data, direct tests and their own technical documentation. It's a change of logic, not just of algorithm.

To find out whether you're at risk, ask yourself: does your content exist because you have something unique to say, or because you rewrote what others have already published?

What exactly happened: which sites gained and which lost?

Between March and September 2026, institutional sites, regulators and product testers rose in ChatGPT citations. Sites that explain and summarize fell, some out of the top 200.

Ahrefs analyzed 2.4 million ChatGPT answers in March and 3.8 million in September, looking at the 200 most-cited domains in US answers. The picture is clear.

PubMed (an archive of original medical research) went from position 64 to position 4. The FDA and other regulatory bodies rose. RTINGS, which runs direct tests on electronics, gained ground. Britannica, Investopedia and the New York Post dropped out of the top 200.

Reddit went from 19.1% to 2.3% of cited answers. Wikipedia from 14.1% to 2.9%. Both remain among the domains the model retrieves most, but they're cited far less in final answers.

Ahrefs verified that the movement is real: comparing about 25,000 identical prompts present in both March and September, every site moved in the same direction. It's not an effect of changed questions, but of how ChatGPT chooses its sources.

How does ChatGPT's citation mechanism really work?

ChatGPT doesn't cite everything it reads. It retrieves many pages in the background, but only surfaces a fraction of them. Understanding how this works helps you build content that makes the cut.

When it gets a question, ChatGPT runs background searches before answering. These searches are called fan-out queries: automated queries the model launches to gather up-to-date information.

Across 66.6 million free responses analyzed by Ahrefs between August and September, only 14.4% included a web search. When a search happens, the model runs an average of 1.6 queries per response.

But retrieving a page doesn't mean citing it. According to an AirOps study of over 548,000 pages, ChatGPT cites only about 15% of the pages it retrieves: the other 85% are read but never appear in the final answer.

One key finding for anyone writing content: 44.2% of citations come from the first third of the page, according to Kevin Indig's analysis of 1.2 million responses. This pattern is called the ski ramp: like a ski slope that starts steep and then flattens out, a page's value for ChatGPT is concentrated at the beginning.

Put the most important and original information in the first few lines. Don't build your article like a funnel leading to the conclusion: ChatGPT may never make it to the end.

Three concrete scenarios: what changes for an e-commerce, a professional firm, and a manufacturing company?

The impact differs depending on the type of content you produce. Here are three typical situations with priority actions.

Illustrative examples: the numbers and situations are meant to show the reasoning, they are not real customer data.

For e-commerce, the risk is real: if your buying guides summarize specs already available on the manufacturer's site, ChatGPT will cite the manufacturer, not you. The fix is to add data only you have: real delivery times, return rates by category, aggregated feedback from your customers.

For professional firms, the path is different. You can't compete with the Agenzia delle Entrate as the primary source of the rule. But you can be the primary source for its practical interpretation: what that rule means for an SME in your sector, with concrete numerical examples.

For manufacturing companies, the situation is often better than you'd think. Technical sheets, compliance certificates, and internal test data are already primary sources. The problem is often structure: this information is buried at the bottom of the page or in PDFs the model can't read.

What content do I need to produce to get cited by ChatGPT?

Ahrefs points in a clear direction: create pages only you can create. Original research, direct tests, proprietary data, exclusive technical documentation.

ChatGPT's logic rewards the source, not the one who describes it. This doesn't mean you have to become a research institute. It means every piece of content needs at least one element that doesn't exist anywhere else.

Why is ChatGPT's citation rate growing so fast?

In less than a year, the share of ChatGPT responses with at least one citation has grown more than fivefold. The trend is accelerating and changing the rules of online visibility.

According to Similarweb data reported by Search Engine Journal, the share of US desktop ChatGPT responses with at least one citation rose from 1.3% in June 2025 to 6.8% in May 2026. That's a more than fivefold increase in eleven months.

This means more and more ChatGPT responses include links to external sources. If you're cited, you get direct visibility. If you're not, you don't exist for those users.

The rate varies widely by sector. Travel is the most cited, with 22.6% of responses including at least one link. Finance and Health are below the overall average of 6.8%, even though they're among the most searched sectors.

Another finding that shifts the perspective: 65.3% of the pages most cited by ChatGPT come from domains with a Domain Rating of 81 or higher, according to an Ahrefs analysis of 1.4 million prompts. Domain Rating measures a site's authority based on the links it receives from other sites. This doesn't mean small sites are never cited, but domain authority still counts for a lot.

To dig deeper into how visibility works on these tools, our SEO and AI visibility cluster collects all the updates that matter for Italian SMEs.

What are the risks and mistakes to avoid in this transition?

The main risk is chasing ChatGPT citations with artificial or inflated content, and losing credibility with real readers. Quality and uniqueness have to go hand in hand.

Here are some common traps we see when companies rush to adapt to these changes.

One point that is often underestimated: Ahrefs warns that its citation data does not measure traffic. Being cited by ChatGPT doesn't automatically mean getting visits. The value lies in visibility and perceived authority, not always in a direct click.

To understand how to measure your visibility in AI concretely, the article on how to earn citations and measure visibility in AI Answers offers a practical method for SMEs.

How do you get started: the steps to adapt your content strategy

Start with an audit of what you already have, identify what is unique, and restructure beginning with the openings of your most important pages.

Something worth keeping in mind: about 30 domains capture 67% of ChatGPT citations on a given topic, and the top 10 collect 46% of the total. Concentration is high. You don't have to be number one, but you need to be among the top thirty for your specific sector.

This is more achievable than it seems for an SME with a well-defined niche. An industrial valve manufacturer can be among the most cited sources on ChatGPT for technical valve questions, even without competing with the big generalist publishers.

Your ChatGPT visibility strategy overlaps with the one for Google AI Overviews: the article on AI Overviews and brand visibility explains how the two approaches overlap and where they differ.

How do I know if my strategy is working?

There is no single metric. You need to combine manual checks on ChatGPT, monitoring of referral traffic from AI, and analysis of your position in answers for your key topics.

Ahrefs points out that its ranking data does not measure traffic. Being cited is a signal of authority, but the commercial value depends on how many people see that citation and what they do next. Keep the two goals separate: AI visibility and actual traffic.

If you want to understand how traditional SEO is transforming in this context, the article on SEO without the Google index and site visibility for AI offers a complementary perspective on how indexing is changing.

So, is it worth changing your content strategy now?

It's worth it if you already have content with original data, direct tests or exclusive technical documentation. You just need to structure it better and put the most unique information up front.

A middle path: you have a mix of original and summary content. Start with the topics that matter most to your business and add one unique element to each priority page, without rewriting everything.

Not yet if your content is mostly summaries of information already available elsewhere. Before investing in new content, ask yourself what you truly know that others don't: that's where to start.

How we see it at SHM

For us, this shift confirms something we've believed for a long time: technology rewards those who have something real to say. ChatGPT simply makes visible a distinction that already existed: between those who produce knowledge and those who redistribute it.

Working with manufacturing and B2B companies on international positioning, we've seen time and again that the most specific technical content — detailed product sheets, test data, process documentation — outperforms any generalist guide, both on traditional search engines and, now, on AI. Not because it's better optimized, but because these are primary sources in the most literal sense: no one else can write them.

Our approach to SEO for AI starts right here: identifying what a company knows that its competitors can't copy, and building around that knowledge a content structure that AI models can read, understand and cite. If you want to see how to apply it to your case, the strategic AI integration service is the most useful starting point. And if you also want to measure results over time, the data processing and business intelligence service lets you track AI visibility systematically.

The change documented by Ahrefs is not an isolated development: it's part of a broader trend we've also followed while analyzing how digital PR is becoming a tool for earning citations on ChatGPT and Perplexity. Whoever builds real authority today will reap the rewards tomorrow.

Frequently Asked Questions

ChatGPT hasn't changed the rules of content: it has made it impossible to ignore the ones that already existed.