YouTube mentions and AI visibility

Mentions of a brand in YouTube titles, descriptions and transcripts were the strongest correlate of AI visibility in Ahrefs' December 2025 study. The link is a correlation with plausible reasons behind it. You earn the mentions by being covered in other people's videos.

In short

  • YouTube mentions correlated 0.737 with brand visibility in ChatGPT, 0.740 in AI Mode and 0.712 in AI Overviews in Ahrefs' December 2025 study of 75,000 brands.
  • Ahrefs counts a YouTube mention when the brand name appears in a video title, transcript or description.
  • YouTube was 5.6% of all URLs cited in AI Overviews in Ahrefs' March 2026 analysis of 4 million cited URLs, the most of any domain.
  • Ahrefs states that the result is a correlation and that improving YouTube metrics will not automatically raise AI visibility.
  • Links in YouTube descriptions are nofollow by practitioner consensus, so the value is the mention and the citation, not ranking credit.

YouTube mentions track AI visibility closely for reasons that are partly mechanical and partly statistical. Models have been trained on YouTube transcripts, Google’s AI surfaces cite YouTube more than any other site, and brands that are widely discussed on video tend to be widely discussed everywhere. You earn the mentions mostly through other people’s videos, not your own channel. The evidence comes from one vendor and is correlational, so this lesson sets out what was measured, why it might hold and where it might not, and what to do about it without betting more on the finding than it can bear.

What the study measured

Ahrefs studied 75,000 brands and published the results on 12 December 2025. For each brand it compared a set of factors with how often the brand appeared in answers from ChatGPT, Google’s AI Mode and AI Overviews.

Factor ChatGPT AI Mode AI Overviews
YouTube mentions 0.737 0.740 0.712
YouTube mention impressions 0.717 0.726 0.714
Branded web mentions 0.664 0.709 0.656
Domain Rating 0.266 0.285 0.326

Two definitions matter. A YouTube mention is the brand name appearing in a video’s title, transcript or description. YouTube mention impressions weight those mentions by the views of each video. Both rows sit above branded web mentions on every engine, and a long way above Domain Rating, which came last of the four in this table.

The second row is the interesting one. Weighting by views did not improve the correlation. A brand named in many videos scored as well as a brand named in heavily watched ones. That fits the idea that assistants read the text around a video and do not care how popular it was. If that reading is right, it changes what a brand should aim for, since a mention in a small channel’s careful comparison leaves the same kind of text record as a mention in a video with a large audience. OtterlyAI’s YouTube citation study, published in March 2026, is summarised on its own site as concluding that being the best answer, and not popularity, seems to drive citations. We did not verify that study’s figures, so we give only its stated conclusion.

Ahrefs gives three explanations in the study. The first is training data. Ahrefs writes that both Google and OpenAI have trained their models on YouTube transcripts. A brand that is spoken about in thousands of videos is present in what the model learned, before any search happens. This is the route that works even when an assistant answers from memory without looking anything up.

The second explanation is retrieval. Google’s AI surfaces cite YouTube directly. In Ahrefs’ March 2026 analysis of 863,000 results pages and 4 million cited URLs, YouTube was the most cited domain in AI Overviews, at 5.6% of all cited URLs, and its citations had grown 34% in six months. YouTube made up 18.2% of the cited URLs that did not rank in Google’s top 100 for the query. So a video can be cited for a question it does not rank for, which is a different route to visibility from the one classic SEO offers. How AI search chooses sources explains how that happens.

The third is ownership. Google owns YouTube, AI Mode and AI Overviews. Ahrefs notes that this cannot be the whole story, because ChatGPT showed the same correlation while YouTube was only its sixth most cited domain. One possible reading of that detail is that it points back to the first explanation, with the content absorbed in training and not fetched at answer time, though the study does not test that.

Other vendors agree on the citation side. Peec AI’s March 2026 analysis of 30 million citations listed YouTube first among cited domains on Google AI Mode and AI Overviews and second on Gemini and Perplexity. Peec states that Reddit and YouTube are the only domains in every engine’s top five, but its own published ChatGPT list is Wikipedia, Reddit, Forbes, TechRadar and LinkedIn, with no YouTube. Profound’s analysis of 680 million citations, from August 2024 to June 2025, put YouTube at 1.9% of AI Overviews citations and 2.0% on Perplexity. The share is real and it is small.

Why it may be weaker than it looks

Ahrefs says plainly that correlation is not causation and that better YouTube numbers will not automatically raise AI visibility. There are three reasons to hold the figure loosely. Size explains a lot: large brands have more videos about them, more press, more searches and more AI visibility for one underlying reason. The study sample was limited to sites above Domain Rating 40, so it describes established brands and says little about a new one. YouTube may also be a proxy. People make videos about products they already use and talk about, so the mention count may measure real-world popularity more than it creates anything. And it is one study. No independent replication exists in our research, and Ahrefs sells an AI visibility product.

The fair reading is that YouTube is a place where being talked about leaves a text record that models and assistants can read. It is one more surface for brand mentions, with stronger numbers than most. That reading supports putting effort into being discussed on video. It does not support the stronger claim that a set number of new videos will produce a set rise in AI answers, and anyone selling that promise is going beyond what the study says.

How to earn mentions in other people’s videos

This is the part that matters most, because the measure counts all of YouTube and your own channel is a small fraction of it. These are working methods, not tested rules.

  1. Find who covers your category. Search YouTube for your category plus “review”, “best”, “vs” and “how to”. Note the channels that appear repeatedly and the competitors they name.
  2. Give creators something to cover. Offer product access, a review unit or a trial account with no script attached. A creator who chooses to cover you produces a credible mention.
  3. Supply facts they can read out. Current pricing, specifications and how you differ from the two nearest rivals. Comparison videos need this and often get it wrong.
  4. Be a guest. Interviews and podcast episodes uploaded to YouTube put your brand and your field into the title, description and transcript.
  5. Ask for the name to be in text. A mention that exists only as a logo on screen leaves nothing to read. The brand should be spoken, and ideally written in the title or description.
  6. Handle disclosure properly. If you pay, or give anything of value, the video is advertising. Influencer and creator collaborations covers the rules.

The fifth step follows directly from how Ahrefs defines a mention. Since the measure counts the brand name in the title, transcript or description, a product that appears on screen for ten minutes without being named has, for these purposes, not been mentioned at all. Product access for a review is also the approach behind sponsorships of creators and events, with the same condition: the creator stays in charge of what is said. That condition is what makes the mention credible to viewers, and it is also what keeps an unpaid review from turning into undisclosed advertising.

Your own videos count as mentions too, and they can be cited directly. Four habits make them more likely to be read correctly. Answer the questions assistants ask: assistants rewrite prompts into longer, specific queries, often adding the word “best”, and videos titled as plain answers to those questions are the ones retrieved. Write full descriptions that state the brand, the category and what the video covers in sentences. Check the transcript, because automatic captions often misspell brand names, and the transcript is what gets read, so a misspelt name is a mention that goes to nobody. And name the category aloud. “Acme, the invoicing tool for freelancers” gives a model the association. “Our product” gives it nothing.

As a link source for classic rankings the picture is plain. Links in descriptions and on channel pages are nofollow by practitioner consensus, so they pass little or no ranking credit. Videos rank in Google in their own right. When another site embeds your video, that produces a mention and sometimes a link. Do social media links help SEO covers the general case. The practical upshot is that a description link is there for viewers to click, and the case for YouTube rests on the mention and the citation.

The same logic suggests podcasts help. An episode produces show notes, a transcript and often a YouTube upload, and show notes usually link to the guest. We found no study that measures the effect. Treat it as plausible and low risk, and prefer shows that publish a transcript and upload to YouTube, since those are the ones that leave the text record the Ahrefs measure counts.

How to measure it, and where to go next

Three measures are enough to start. Count videos that mention your brand, and compare with your nearest competitors, since the gap between you and them tells you more than the raw number. Track whether YouTube URLs appear as sources for your priority prompts, using a tracker from the AI visibility tools comparison. Ahrefs is the vendor behind the mention measure used in this lesson. Then report mention rate in AI answers over many runs, before and after a creator campaign. Expect noise, because answers vary between runs and because other things change in the same period, so a single before-and-after comparison proves little and a trend across several campaigns is worth more.

For Reddit, the community surface that appears in all five of Peec AI’s published top-five lists, read Reddit for SEO and AI visibility. The glossary defines mention, citation and the other terms used here.

Common questions

Why do YouTube mentions correlate with AI visibility?

Ahrefs offers three reasons. Google and OpenAI have trained models on YouTube transcripts, Google's AI surfaces cite YouTube more than any other domain, and Google owns YouTube. A fourth reason is confounding, since well-known brands are discussed more everywhere.

Do YouTube links count as backlinks?

They are links, but links in video descriptions and on channel pages are nofollow by practitioner consensus and pass little or no ranking credit. Videos themselves can rank in Google and be cited in AI answers.

Do I need my own YouTube channel?

It helps, but the Ahrefs measure counts mentions across all of YouTube. Being named in other creators' videos is the larger part of it and the harder part to earn.

Do podcasts help AI visibility?

Possibly, by the same logic, since episodes produce transcripts, show notes pages and often YouTube uploads. We found no study that measures it.

Do views matter for being cited?

Ahrefs' measure that weights mentions by video views correlated about the same as the plain count, 0.714 to 0.726 against 0.712 to 0.740. That suggests how often a brand is named matters at least as much as how many people watched.

Vendors to look at

  • BazoomEditor's pick

    Sponsored content and link marketplace, managed service

  • MotherlinkEditor's pick

    Backlink services, guest posts, niche edits, full SEO