In August 2026, Zoom’s Chief Marketing Officer said something most CMOs would rather not say out loud. Kimberly Storin had just signed a deal with journalist and podcast host Nayeema Raza, built around Zoom’s biggest brand problem: nobody doubts the product, but almost nobody thinks of Zoom as anything more than a meetings app.
The Raza partnership was pitched internally as an AI visibility play, a bet that a credible, indexable third-party voice could shape how ChatGPT and Gemini describe Zoom to the next buyer who asks. Then Storin was asked, plainly, whether the bet was working. “It’s not a direct correlation at this point,” she told Digiday, “but the influencers, we believe, is so important to that ‘be present and be believed’ aspect of the framework.”
That is a company spending real money on a strategy its own CMO cannot yet prove works. It is also, once you start looking, the industry’s open secret.
Table of contents
Jump to each section:
- The claim everyone in GEO is selling
- What the correlation data actually shows
- The one dataset that gets specific
- Betting ahead of the proof
- The honest synthesis
The claim everyone in GEO is selling
Every GEO vendor deck now makes some version of the same pitch: creator content drives AI citations, so buy the platform that tracks it, or buy the agency that produces it. What almost nobody selling that pitch does is separate the two very different kinds of evidence behind it. There is large-scale correlational research from independent analytics firms, and there is a smaller pile of vendor and practitioner case studies from companies with a direct commercial stake in the answer.
Sorted honestly, those two piles support very different levels of confidence. The strong version of the claim, the one on the vendor decks, is that creator content specifically causes AI visibility gains. Nobody has shown that.
The weaker, better-supported version is that third-party and earned content generally, of which creator content is one ingredient, dominates what AI systems choose to cite. That gap between the strong claim and the supported one is the actual story here.
What the correlation data actually shows
Ahrefs got closest to quantifying it. In a December 2025 report covering 75,000 brands across ChatGPT, AI Mode, and AI Overviews, Ahrefs researchers ran a Spearman correlation between dozens of brand signals and how often each brand turned up in AI answers:
- YouTube mentions came out on top, at roughly 0.737.
- Branded web mentions followed, at 0.66 to 0.71 depending on platform.
- Backlinks, the metric an entire generation of SEO strategy was built around, scored just 0.218.
- Site page count barely registered at all, around 0.194.
Ahrefs’ own researchers were careful to flag the obvious catch: correlation is not causation, and the data shows which brands already have visibility, not what put them there.
Sit with that YouTube number for a second, because it is the closest thing to a creator signal in the entire dataset. A “YouTube mention” is any time a brand name shows up in a video title, transcript, or description, and in practice that is dominated by creator-made content: reviews, unboxings, explainers, tutorials filmed by people who are not the brand.
Ahrefs never asked who made the video or why it got made. So this is not proof that creator content specifically drives the correlation. But of everything in the correlational tier, it is the strongest hint that creator-adjacent content sits closer to AI visibility than owned brand content ever does.
There is a related data point worth putting next to that one, and it moves the same direction. Ahrefs’ Brand Radar, which republishes its citation rankings every month, shows Instagram sitting in Google AI Overviews’ top five most-cited domains as of July 2026, at a 5.8% mention share, behind YouTube (21.1%), Reddit (18.5%), and Facebook (10.7%), but ahead of Wikipedia.
That does not prove Instagram content, let alone creator content on Instagram specifically, causes citations. But it is a real, measured sign that the pool of social platforms feeding AI answers keeps widening, not narrowing, and creator-produced content is what fills most of that space on every platform in the list.
Ahrefs followed up in May 2026 with a separate, larger Q1 2026 AI Search Benchmark Report, thirteen studies deep, pulling from 146 million search result pages and 730,000 AI responses across the same 75,000 brands. It confirmed YouTube mentions as the single strongest visibility signal, directionally, without restating the earlier report’s precise coefficients. Two Ahrefs publications, six months apart, pointing the same direction.
Muck Rack’s Generative Pulse team arrived at a strikingly similar answer from a completely different angle. Its May 2026 “What Is AI Reading?” report, the third edition of an ongoing study, analyzed more than 25 million links cited by ChatGPT, Claude, and Gemini across 17 industries:
- Earned media accounted for 84% of every citation.
- Paid and advertorial content accounted for 0.3%, close enough to zero to round down.
- Across all three editions of the study since July 2025, earned media’s share has held between 82% and 89%, and journalism specifically has stayed remarkably steady, between 25% and 27%.
What both studies actually prove, put together, is that third-party and earned content overwhelmingly beats brand-owned content in what gets cited. That much is about as settled as anything in this field gets right now.
What neither study proves is the creator-specific version of the claim. “Earned media,” in both datasets, is a wide bucket that swallows journalism, forum threads, review sites, and academic papers along with creator content. Nobody has pulled creator content out of that bucket and given it its own number.
The one dataset that gets specific
One study gets closer, at least on one platform. In January and February 2026, Semrush analyzed 89,000 LinkedIn URLs cited across ChatGPT Search, Google AI Mode, and Perplexity. LinkedIn turned out to be the second most-cited domain in the entire dataset, ahead of Wikipedia and YouTube.
But the more interesting number is buried further in: split by author type, Perplexity favors LinkedIn Company Pages 59% of the time, while ChatGPT Search and Google AI Mode do the opposite, citing individual members rather than brand accounts 59% of the time.
That is a real, measured line between brand-owned and individually-authored content, sitting inside a single citation dataset. Of everything currently on the table, it is the closest anyone has gotten to isolating the variable the whole industry keeps claiming to have already proven.
Dinda Anandita, Account Director at Content Collision, a content-led comms agency, sees the gap between the claim and the data play out in client meetings regularly, though she says the picture has genuinely shifted over the past year.
“Brands used to ask us to prove one creator post caused one citation, and we’d have to explain that no study has ever isolated it that cleanly,” she says. “What’s changed is that YouTube and LinkedIn are now established citation sources, and Instagram is climbing the same rankings. So it’s fair to say creators are becoming part of the information ecosystem these models actually pull from, not just adjacent to it.”
Betting ahead of the proof
Which brings the story back to Zoom, and to the strange position brands like it are choosing to occupy. Zoom’s creator strategy runs across Instagram and YouTube with Raza, a sponsored six-part podcast series, and a keynote slot at Zoom’s own conference in October.
None of it is built on a study that says this works. It is built on the correlational data above, plus a bet that authority and credibility, the two things Storin says the partnership is buying, are the variables AI systems already reward.
It is worth being fair to that bet, because the ground underneath it keeps shifting. Profound’s tracking shows LinkedIn’s citation rank on ChatGPT more than doubling in just three months, from roughly 11th to roughly 5th, between November 2025 and February 2026. Ahrefs and Semrush both describe monthly, not annual, movement in which domains and platforms models favor.
A partnership that looks unproven against today’s data could look prescient in six months, if the platforms keep drifting toward the kind of individually-authored content Zoom is betting on. That is a reason to keep watching this space, not a reason to call the bet validated already.
The honest synthesis
The correlation is strong enough to act on. Third-party and earned content, at the aggregate level, clearly and repeatedly beats brand-owned content in what AI systems choose to cite, across every independent study that has measured it since mid-2025. The causal, creator-specific version of the claim does not exist yet. Nobody has separated creator-produced content from the broader earned-media bucket and shown, on its own, that it causes AI visibility gains.
That gap is the reason to treat any single case study in this piece, Zoom included, as a data point and not a verdict. A brand acting on the correlational data today is making a reasonable bet, not following a proven playbook, and the difference matters for how much certainty gets attached to whatever results follow.
This piece will be updated as new studies land, and it should not take long. The field is producing fresh platform-specific citation data every few months. The LinkedIn breakdown above did not exist a year ago. Each new dataset chips slightly further at the gap between “earned media wins,” which is close to settled, and “creator content specifically wins,” which is still an open question waiting on someone to isolate the variable properly.
Leave a Reply