Brand Mentions Now Matter 3x More Than Backlinks for AI Visibility
A 75K-brand study reveals brand mentions correlate 3x more strongly with AI visibility than backlinks. Here's what this means for your strategy.
Brand Mentions Now Matter 3x More Than Backlinks for AI Visibility
For fifteen years, backlinks were the currency of search. More links from better domains meant higher rankings. SEO teams built entire strategies around link acquisition, and the industry grew a multi-billion dollar economy of outreach, guest posts, and digital PR, all measured by referring domain counts.
That playbook is breaking down. Not because backlinks stopped mattering for Google’s traditional results. They still do. But because AI search, the fastest-growing discovery channel in marketing, runs on a fundamentally different signal: brand mentions.
Ahrefs recently studied 75,000 brands across ChatGPT, Google AI Mode, and AI Overviews to identify which factors correlate most strongly with AI visibility. The top finding: branded web mentions showed a 0.664 correlation with AI Overview visibility. Backlinks? Just 0.218.
Brand mentions correlate three times more strongly with AI visibility than the metric SEO teams have obsessed over for a decade and a half.
Why Mentions Beat Links in AI Systems
The reason isn’t mysterious once you understand how AI platforms build answers.
Traditional search engines treat links as votes. Each link from Site A to Site B is an endorsement, weighted by the authority of Site A. The algorithm counts votes and ranks accordingly. It’s a graph problem.
AI systems don’t count votes. They synthesize information. When ChatGPT or Gemini answers a question, they read content from multiple sources, look for patterns, and construct a response that reflects what the collective internet seems to agree on. A brand mentioned positively across ten different articles carries more weight than a brand linked from ten directories, because the AI is reading the text, not following the links.
Think about it from the model’s perspective. If Reddit threads, industry publications, comparison sites, and forums all mention your brand when discussing a topic, the AI has multiple independent data points confirming your relevance. If you have backlinks from those same sources but no contextual mention, the AI gains nothing from the link itself.
As iProspect’s analysis puts it, LLMs analyze “surrounding context, sentiment, and frequency patterns to determine brand authority.” They care about what’s said about you, not who linked to you.
The Numbers Across Platforms
The Ahrefs study didn’t just find one correlation. It mapped visibility factors across three major AI surfaces, and the differences between platforms tell a strategic story.
| Factor | AI Overviews | AI Mode | ChatGPT |
|---|---|---|---|
| Brand web mentions | 0.664 | 0.66-0.71 | 0.66-0.71 |
| YouTube mentions | ~0.737 | High | High |
| Branded anchors | — | 0.628 | Lower |
| Brand search volume | — | 0.466 | 0.352 |
| Domain Rating | — | — | 0.266 |
| Backlinks | 0.218 | Lower | Lower |
Three patterns jump out.
YouTube mentions are the single strongest signal. Across all three platforms, being mentioned in YouTube content correlates more strongly with AI visibility than any other factor, at roughly 0.737. This aligns with recent citation data showing YouTube doubled its share of AI citations in the second half of 2025. YouTube transcripts give AI models clean, structured text to work with, and Google’s AI surfaces pull heavily from its own video platform.
Google’s AI favors branded authority signals. AI Mode showed the highest correlations with branded anchors (0.628) and branded search volume (0.466). If people search for your brand name on Google and if other sites use your brand name as anchor text, Google’s AI is more likely to mention you. This makes sense: Google has deep data on search behavior and uses it.
Traditional authority metrics matter least in ChatGPT. Domain Rating (0.266) and backlinks show the weakest correlations in ChatGPT’s results. ChatGPT relies more on the content itself and contextual mentions across the web, not on the link graph that traditional SEO tools measure.
What This Means (And What It Doesn’t)
The researchers were careful to note that correlation isn’t causation. Brands with more mentions tend to be more visible in AI, but that doesn’t prove mentions directly cause visibility. Both could be driven by a third factor: real-world brand strength.
That caveat matters. But the practical implication stands regardless.
If your team is spending 80% of its external SEO budget on link building and 20% on earning brand mentions, those ratios are probably backwards for AI visibility. The data suggests that the signals AI systems respond to are fundamentally different from the ones traditional search algorithms use.
And here’s the uncomfortable part for SEO teams: Ahrefs separately found that Google’s AI surfaces are significantly more biased toward big brands than ChatGPT or Perplexity. Brands in the top quartile of web mentions earn up to 10x more AI Overview mentions than the next quartile. The gap between large and small brands is wider in AI search than in traditional search.
If you’re a smaller brand, this sounds discouraging. But it also means the opportunity is clearest for brands willing to invest in building mention volume, because the returns compound. Every new mention pushes you up the curve.
The Perplexity Wrinkle
There’s one more data point worth noting. The ConvertMate AI visibility study, which analyzed over 80 million citations across 10,000 domains, found that Domain Rating actually shows a negative correlation (-0.18) with visibility in Perplexity.
Read that again. Higher domain authority is weakly associated with lower visibility in Perplexity.
This doesn’t mean authority hurts you. It more likely means Perplexity is pulling from a different content mix than Google, favoring niche and topically-specific sources over broadly authoritative domains. A focused industry blog with strong topical relevance might outperform a major publisher in Perplexity’s citations for a specialized query.
Each platform has its own weighting:
| Platform | Strongest Signal | Weight |
|---|---|---|
| Perplexity | Content freshness | ~40% |
| ChatGPT | Referring domains + mentions | ~30% |
| Gemini | E-E-A-T signals | ~35% |
| Claude | Entity verification | ~30% |
This per-platform variation is why a single “optimize for AI” strategy falls short. What works for Google AI Mode (branded anchors, search volume) is different from what works for Perplexity (fresh, topically relevant content from niche sources).
A Practical Strategy Shift
Knowing that mentions outweigh links is one thing. Shifting your actual strategy is another. Here’s what a rebalanced approach looks like.
Audit Where You’re Mentioned (Not Just Linked)
Most SEO tools track backlinks beautifully. Far fewer track unlinked brand mentions. But for AI visibility, an unlinked mention in a respected industry article may matter more than a link from a low-relevance domain.
Start by mapping your current mention landscape:
- Where does your brand appear in “best of” and comparison content?
- Which review platforms mention you?
- Do industry publications reference you in trend pieces?
- Are you mentioned in Reddit threads, YouTube videos, or forum discussions?
The gaps in this map are your highest-priority opportunities.
Prioritize Mention-Generating Activities
Some marketing activities generate mentions naturally. Others generate links but not mentions. Reweight toward the first category.
| High Mention Value | High Link Value (But Lower Mention Value) |
|---|---|
| Original research and data reports | Guest posts on unrelated sites |
| Expert commentary in industry press | Directory submissions |
| Participation in roundups and comparisons | Sponsored content |
| Community contributions (Reddit, forums) | Link exchanges |
| Being genuinely newsworthy | Broken link building |
The left column builds the kind of distributed, contextual brand presence that AI systems pick up on. The right column may still help traditional rankings, but does less for AI visibility.
Search Engine Land recommends leading with “referenceable assets” like original data, definition pages, and tools. These are the kinds of resources that earn mentions organically because other writers need to reference them.
Invest in YouTube Presence
The data is clear: YouTube mentions are the strongest single correlator with AI visibility. This doesn’t mean every brand needs a YouTube channel. But it means that for brands producing educational or comparison content, video versions of that content create an additional surface for AI citation.
As we covered in our YouTube citation analysis, AI models read YouTube transcripts as structured text. A well-transcribed explainer video can earn citations from Google AI Overviews, Gemini, and Perplexity simultaneously.
If you already create comparison guides or how-to content in written form, consider producing video versions. The written content becomes the transcript. The video becomes another citation surface.
Build Brand Search Volume
Brand search volume (people Googling your brand name) showed strong correlations with AI visibility, particularly in Google AI Mode (0.466). This is circular to some extent: well-known brands get searched more and get mentioned more in AI. But it suggests that brand awareness campaigns, even ones that don’t generate direct links or mentions, contribute to AI visibility indirectly.
PR campaigns, conference speaking, podcast appearances, social media visibility: these all build the kind of brand recognition that translates into search volume. And search volume, it turns out, is a signal AI systems use.
Don’t Abandon Backlinks Entirely
The data shows mentions correlate more strongly with AI visibility than links. It doesn’t show that links are worthless. For traditional Google rankings, backlinks still matter significantly. And traditional rankings feed into AI visibility, since AI systems like ChatGPT use search engines as their discovery mechanism.
The right move isn’t dropping link building. It’s rebalancing. If your external efforts are 100% link-focused, shift to maybe 50/50 between link acquisition and mention generation. If you’re already doing PR and community work, you may just need to measure those efforts differently.
Measuring What Matters Now
The biggest challenge in this shift is measurement. SEO teams have sophisticated tools for tracking backlinks, referring domains, and domain authority. Brand mention tracking is less mature.
Here’s a starter framework:
| Metric | What to Track | Tool Category |
|---|---|---|
| Unlinked brand mentions | Volume and sources of contextual mentions | Brand monitoring tools |
| Mention sentiment | Positive/negative/neutral context | Social listening + manual review |
| AI citation sources | Which sites AI platforms cite when mentioning you | AI visibility monitoring |
| YouTube mention frequency | How often your brand appears in video transcripts | YouTube search + monitoring |
| Brand search volume | Monthly branded search queries | Google Search Console, Ahrefs |
| Share of voice in AI | % of relevant queries where you appear | AI search monitoring |
The brands that will adapt fastest are the ones that start measuring mentions with the same rigor they currently measure links. Because the data is telling us: in the AI era, what people say about you matters more than who links to you.
That’s not a comfortable message for an industry built on link metrics. But the numbers from 75,000 brands don’t lie.
Want to know what AI platforms say about your brand? Try RivalHound free and find out.