Reddit gets the attention in this data because it is first. The more actionable finding is second and third: YouTube at 145 citations and LinkedIn at 134, both ahead of every agency website in the category we measured.
Most B2B content budgets go to the domain that came fifth.
The numbers
Across 608 AI answers collected between May and August 2026:
| Domain | Answers citing it |
|---|---|
| reddit.com | 225 |
| aeoagency.org | 148 |
| youtube.com | 145 |
| linkedin.com | 134 |
| aeoengine.ai | 130 |
The first agency-owned domain appears at position two, and it is an exact-match domain running a large content operation. The next is at five. YouTube and LinkedIn sit between them.
Why video gets cited when nobody watches it
The model is not watching. It is reading the transcript, the title, the description and often the comments.
That turns a video into a text document with two useful properties. It is usually somebody talking through a problem in ordinary conversational language, which is the phrasing retrieval systems match against. And it carries an engagement signal that a page does not.
The practical consequence is that a video with a thin description is worth much less than the same video with a real one. Most channels treat the description as a place for links.
Why LinkedIn punches above its weight
Two things.
People write about problems there in the first person. A post describing a specific frustration, with replies arguing about it, has the same shape as the Reddit thread that gets cited: real language, real disagreement, and a visible signal about which answer people found useful.
It is unusually crawlable for a walled garden. Public posts and articles are indexed and retrievable in a way most social content is not.
What this changes
It does not mean stop publishing on your own site. Somebody has to be named in the answer, and the model needs a reason to believe the named business is real, which your site provides.
It means the sources deciding who gets named are mostly not yours, and a content plan that lives entirely on your own domain is addressing the half of the problem that is easiest to control.
What we would actually do
Put your best explanation on video, with a real description. Not a promotional video. The one where somebody who knows the subject talks through the question your buyers ask. Write the description as though it were the page.
Write on LinkedIn in the first person about specific problems. The posts that get cited are the ones that describe a real situation in real words, not the ones announcing a blog post.
Do not repost your blog. Both surfaces reward the language people actually use, and syndicated marketing copy has the wrong shape for both retrieval and for the audience.
Then keep your own pages good, so the check on your legitimacy passes.
The honest caveat
We measured one category, our own, which makes it the one we understand best and the one where we have an interest. The mechanism, that retrieval favors problems described in ordinary language, is not category-specific. Where that description happens will differ: in some industries it is YouTube, in others a trade forum or a professional association.
The generalisable instruction is not "make videos". It is: find where your buyers describe your problem in their own words, and be useful there.
The practical version for a business with no video team
The instinct on reading this is to conclude you need a content studio. You do not, and the reason is in why these sources get cited in the first place.
A model is not watching your video. It is reading the title, the description, the transcript and the comments. That means the production value is close to irrelevant and the words are close to everything.
What is worth doing. Record yourself answering the ten questions your buyers actually ask. One question per video, two to four minutes, screen or face, whatever is easier. Write a real description rather than a keyword line, and make sure captions are on so the transcript is accurate.
What LinkedIn needs. The same answers, written out, posted as articles rather than as short posts. Articles persist and get read. Short posts do not.
How much this is. Ten videos and ten articles is two weeks of intermittent work for one person, and it puts you in the two source types that out-cite every vendor website in the category.
What not to do. Repurposing your blog posts verbatim into both. The value is in the different phrasing a spoken answer produces, which is closer to how the question was asked.
What that presence contributes
The figures above are why the research on this page exists. We run this measurement continuously across every account rather than sampling it, which is the only way to reach a number like 82,000 measured answers and the only way to see a change in the week it happens rather than the month after.
A worked example
A B2B services firm with no video capability and a dormant LinkedIn page.
What they did. Ten videos, recorded on a laptop webcam, two to four minutes each, one buyer question per video. No editing beyond trimming. Titles written as the question. Descriptions written as a real paragraph answering it. Captions checked for accuracy rather than left auto-generated.
Then the same ten answers written up as LinkedIn articles, not short posts.
What that cost. About twelve hours in total, spread over three weeks, mostly the writing rather than the recording.
What happened. Two of the ten videos began appearing as cited sources within about six weeks, both on the narrowest questions. The LinkedIn articles took longer and two eventually appeared as well.
The part that surprised them. Production quality had no visible effect. The two that got cited were not the two that looked best. They were the two with the most precise titles and the most accurate captions, which is exactly what you would predict if you remember that nothing is watching the video.
What did not work. Three short LinkedIn posts published in the same period. None were cited. Articles persist and get read; short posts do not.
See which questions name you
A free AI visibility audit on your own buying questions, across all five answer engines.
Book your audit
Ashur Homa
Built and scaled a digital brand to $100M+ in sales with zero ad spend. Has helped businesses generate millions through AI go-to-market strategy. Leads growth at Omni Eclipse.
Connect on LinkedIn
