Your social team is already doing SEO. Nobody told them.

Key takeaways

• Google is right that likes and shares aren't ranking factors. That was never the point. Social platforms are now a big part of the search results and the AI citations themselves.
• YouTube and Reddit are the two most-cited sites in Google's AI Overviews. If your brand isn't on them, AI will describe your category using someone else's content.
• Merge the SEO and social roadmaps around shared intent, treat captions and spoken audio like metadata, earn Reddit honestly, and measure share of search instead of fighting over last click.

---

Picture two teams at the same big brand, two floors apart.

The SEO team is cleaning up schema, tuning category pages and arguing with engineering about page speed. The social team is in a different meeting, celebrating an engagement rate on a Reel. Neither team has seen the other's roadmap. When someone asks whether social helps search, the SEO lead gives the answer we've all given for 15 years: "Social signals aren't a ranking factor."

That answer is still technically true. Google's John Mueller put it plainly: "we don't use likes as a ranking factor." I believe him.

But that answer has quietly cost brands a fortune, because it trained a generation of marketers to treat social and search as separate worlds. They aren't anymore. If your brand isn't earning attention, conversation and watch time on social and community platforms, your SEO and AEO ceiling is capped, no matter how clean your technical setup is.

Social is bigger than Instagram and TikTok

Ask most executives what "organic social" means and you'll hear Instagram and TikTok. That definition is too small.

When people research a big purchase, a new hobby or a life decision, their journey isn't a straight line. It's a pinball machine of validation. They search, watch a video, read a thread, see a Reel, search again with a brand name attached, then finally visit a site. Here's the full stack I'd want every B2C brand to plan for:

• YouTube is the world's explainer. People go there to see proof: the unboxing, the setup, the side-by-side test. AI systems read its titles, descriptions and transcripts to build answers.
• Reddit is the internet's sanity check. People add "reddit" to searches to skip the marketing speak. Google's own CEO acknowledged this in 2023, saying people add it because they want "more comprehensive answers" than blue links. Google has since signed a data partnership with Reddit that includes showing more Reddit content in its products.
• Instagram and TikTok aren't walled gardens anymore. Since July 2025, public posts from Instagram professional accounts have been eligible to appear in Google and Bing, including photos, carousels and Reels. Captions and hashtags now work a lot like a landing page.

Four ways social shapes search and AI results

"It builds awareness" is true, but it's also the vague answer that gets social budgets cut. Here's what's actually happening.

1. Social platforms are the most-cited sources in AI answers

Your own website is still your most important asset. But look at who else AI is quoting. In Ahrefs' September 2026 analysis of more than 3 million US queries, YouTube was the most-cited domain in Google AI Overviews, with 22.9% of citations among the top 50 sites. Reddit was second at 18.5%. Facebook, Instagram, Quora and TikTok also made the top eight. Six of the eight most-cited domains are social or community platforms.

That fits a bigger pattern. Muck Rack found that earned media made up 84% of AI citations in May 2026. AI is going to describe your category either way. The only question is whether it uses your content and your fans, or someone else's.

2. Search results are crowded with social content

On many high-intent consumer searches, page one now mixes classic blue links with video results, "Discussions and forums" threads and short-video carousels. If you own one blue link, you own a sliver of the screen. If you own the page, the YouTube video and a positive Reddit thread, you surround the searcher.

3. AI learns your brand from what people say about it

Search engines and AI models map which words keep showing up next to your brand: which problems, which use cases, which adjectives. Those associations come from transcripts, creator reviews and forum threads as much as from your own site.

Ahrefs studied 75,000 brands and found that branded web mentions correlated with AI Overview visibility at 0.664, about three times the correlation for backlinks (0.218). Correlation isn't proof, but it's a strong hint about where to spend effort. Social is one of the biggest mention engines a brand has.

4. Social creates searches, and searchers create signals

A great video makes people curious, and curious people search your brand name. That matters twice. In the same Ahrefs study, branded search volume also correlated with AI visibility (0.392). And Google has confirmed, through testimony from its own search leadership, that it uses a system called Navboost that learns from clicks on past searches. When people already know and trust your name, they're more likely to pick you on the results page.

How social keeps a buyer in the loop

Here's a simple example. A family is planning its first real camping trip and needs a tent.

Dad searches "best family tent for beginners." An AI Overview cites a YouTube setup video and a Reddit thread. He watches the video, where a creator pitches the tent in four minutes using timestamped chapters for setup, rain test and packing. That night, he sees the brand's Reel about a tent that survived a mountain thunderstorm. A week later, he searches the brand name plus "vs," reads a Reddit thread where real owners answer questions, and finally clicks through to the brand's site and buys.

Last-click attribution gives all the credit to that final branded search. But the sale was won over three weeks, on four platforms, before the site ever got a visit. Social didn't just support the search journey. It was most of the journey.

The new enterprise playbook

You don't need double the headcount to fix this. You need the SEO and social teams working from one roadmap. Here are four places to start.

1. Map intent once, then publish everywhere. Take your top search queries and the questions people ask AI, and turn them into YouTube briefs and short-video hooks. Use clear titles, accurate transcripts and timestamped chapters. Google says it prioritizes key moments that creators set in YouTube descriptions, so that work pays off in search too. I wrote about why intent matters more than keyword lists in The end of the keyword.

2. Treat captions and spoken audio like metadata. Write captions, on-screen text and the first spoken line for human curiosity first and search second. Say the actual product name and the actual problem out loud. Just don't turn posts into keyword soup. People scroll past robots.

3. Earn Reddit. Never fake it. Sock-puppet accounts and disguised ads tend to get spotted fast, and the backlash becomes the thread that ranks. Participate openly as the brand, answer real questions, support the genuine fans you already have, and mine threads for content gaps. Reddit will tell you exactly what your website isn't explaining well.

4. Stop fighting over last click. Organic social works like brand building, so measure it that way. Track share of search, the share of branded searches in your category that go to your brand. Les Binet's research found it correlates with market share. Track how much of page one you own across your site, videos and threads. And watch AI citation share with prompt-tracking tools, but treat it as a directional barometer, not a hard number for the board deck.

None of this replaces the fundamentals. Your website is still the hub, and earned media still does a lot of the heavy lifting. But the brands that win the next few years will be the ones that stop treating social and search as two departments.

Likes were never the ranking factor. Being the brand people talk about, everywhere they look, is.